Hype Cycle for Application Architecture and

Integration, 2021
Published 15 July 2021 - ID G00747497 - 82 min read
By Analyst(s): Eric Thoo, Massimo Pezzini
Initiatives: Software Engineering Technologies; Software Engineering Strategies

Composable, digital business is driving application architecture

and integration to modernize and transform. Applications and
software engineering leaders can leverage the technologies and
practices in this Hype Cycle to meet the increasing demand for
innovation, agility and scalability.

Gartner, Inc. | G00747497 Page 1 of 78

Analysis
What You Need to Know
The escalating demand for innovation, agility and scalability from application and
integration capabilities compels application and software engineering leaders to pursue
technological advances and practices better suited to their organizations.

However, increasingly complex, ever-changing application portfolios make it more and

more challenging for software engineering leaders to modernize and deliver new
application capabilities. The need for greater business agility, remaining competitive and
innovation hinges on integration infrastructure composed as part of a broad, cohesive
strategy to deal with the disruptive changes associated with digital business
transformation.

At the same time, as businesses transition through the COVID-19 pandemic and
associated economic changes, leaders responsible for modernizing application and
integration infrastructure are under pressure to reduce the cost and improve the
performance of their application portfolios. It is becoming critical to focus on improving
business process efficiency via integration and automation.

This Hype Cycle, along with Hype Cycle for Software Engineering, 2021, reflects the
position, rate of adoption and speed of maturation of innovative technologies and
practices that will affect the evolution of application and integration infrastructure. Many
of these innovations can have a short- to medium-term impact on application and
software engineering leaders’ strategies and tactics, but all collectively pave the way for
the composable business revolution.

The Hype Cycle

This year’s Hype Cycle for Application Architecture and Integration highlights a number of
important trends:

Gartner, Inc. | G00747497 Page 2 of 78

■ Frictionless sharing of applications and data — Capabilities for integration and
application infrastructure are evolving with a growing emphasis on democratized
low-code and no-code approaches, API-enablement, data fabric, and support for
event-driven design. Implementing these capabilities will lay the foundations for user
organizations’ application portfolios, to readily access, assemble and provision a
broad range of business functions and data for diverse use. Such implementations
have led to further developments in packaged integration processes, integration
platform as a service (iPaaS), data integration tools and full life cycle API
management to support the changing nature of business.

■ Modernizing application and integration infrastructure — As organizations navigate

pathways to the future of applications, they will require more advanced yet, at times,
less mature techniques. These will include implementing application functionality as
a service mesh tied to emerging elements of application architecture such as event-
driven architecture, event broker platform as a service (PaaS), along with hybrid
integration platform (HIP) capabilities.

■ Optimizing application services and composition — Motivated to prepare for the

composable business, software engineering leaders are increasingly interested in
microservices, event stream processing, Internet of Things (IoT) integration and
cloud-native application architecture. Additionally, the requirement to provide large-
scale, high-throughput and low-latency API platforms is driving the emergence of
digital integration hubs, which also provides application leaders an opportunity to
reduce costs by offloading and decoupling expensive legacy systems.

■ Harnessing artificial intelligence (AI)-augmentation — Software engineering leaders

responsible for integration seeking to simplify integration development and improve
time to value are investigating innovations through a growing use of AI in
integration. Digital integrator technologies and self-integrating applications
capitalize on opportunities for using augmented integration to rebalance the work of
humans and AI.

■ Extensible platform architecture and deployment — Maturing cloud-based and hybrid

offerings are making integration and application infrastructure more broadly
applicable and easier to build and manage. Leaders responsible for low-code
approaches continue to adopt application PaaS, low-code application platforms and
various PaaS technologies, for faster time to value and increased developer and
integrator productivity. This trend is spawning diverse opportunities to exploit PaaS
options across a wide range of cloud-based technologies featured in Hype Cycle for
Platform as a Service, 2021.

Gartner, Inc. | G00747497 Page 3 of 78

Figure 1: Hype Cycle for Application Architecture and Integration, 2021

Source: Gartner (July 2021)

Downloadable graphic: Hype Cycle for Application Architecture and Integration, 2021

The Priority Matrix

Application and software engineering leaders should closely monitor the following
innovations, which will provide the greatest benefits with the shortest times or at current
mainstream adoption:

■ iPaaS

■ Data integration tools

■ Full life cycle API management

■ IoT integration

■ Low-code application platform (LCAP)

Some innovations will take longer to achieve mainstream adoption relative to the ones
above, but have proven to deliver high or even transformational value, including:

■ Data hub iPaaS

Gartner, Inc. | G00747497 Page 4 of 78

■ Digital integration hub

■ Digital integrator technologies

■ Event stream processing

■ Event broker PaaS

■ Integration strategy empowerment team

■ Packaged integration processes

Adoption of these innovations requires investment in skills and presents some risks
because of their intrinsic complexity or still-limited industry support. However, their market
penetration is growing due to many successful deployments and associated lessons
learned.

Other innovations in this Hype Cycle are either relatively mature but have a moderate
impact, or their low level of industry adoption dilutes their potentially high benefits.

Finally, a small number of innovations are still in the initial stages of their life cycle, so
application and software engineering leaders should assess their risks and rewards
associated with adoption.

Gartner, Inc. | G00747497 Page 5 of 78

Table 1: Priority Matrix for Application Architecture and Integration, 2021
(Enlarged table in Appendix)

Gartner, Inc. | G00747497 Page 6 of 78

On the Rise
Self-Integrating Applications
Analysis By: Keith Guttridge, Eric Thoo

Benefit Rating: Transformational

Market Penetration: Less than 1% of target audience

Maturity: Embryonic

Definition

Self-integrating applications will use a combination of automated service discovery,

automated metadata extraction and mapping, automated process definition and
automated dependency mapping to enable applications and services to integrate
themselves into an existing application portfolio with minimal human interaction.

Why This Is Important

Integrating new applications and services into an application portfolio is complex and
expensive. Gartner research shows that up to 65% of the cost of implementing a new ERP
or CRM system is attributable to integration. The technology to enable applications to
self-integrate exists in pockets, but no vendor has yet combined all the elements
successfully. As applications develop the ability to discover and connect to each other, the
amount of basic integration work will dramatically reduce.

Business Impact

■ Improved agility, as the time to onboard applications and services is massively

shortened.

■ Costs savings of up to 65% when onboarding new applications and services.

■ Reduced vendor lock-in, as platform migration becomes simpler.

■ Greater ability to focus on differentiation and transformational initiatives, as the

“keep the lights on” burden is dramatically reduced.

Drivers

■ Cloud hyperscalers providing features, such as service discovery, metadata

extraction, intelligent document processing, natural language processing.

Gartner, Inc. | G00747497 Page 7 of 78

■ Automation/integration vendors providing features, such as intelligent data
mapping, metadata extraction, data fabric, next best action recommendations,
process discovery, automated decisioning.

■ SaaS vendors providing features, such as process automation, packaged integration

processes, portfolio discovery, platform composability.

■ A new era in which intelligent application portfolio management is placed on top of

augmented integration platforms, in order to be where the challenge is finally
addressed.

Obstacles

■ Embedded integration features within SaaS being good enough to enable

organizations to get started quickly, thus stalling investment in improving self-
integration capabilities.

■ A general lack of awareness of the availability of augmented integration

technologies to enable self-integrating applications. Many organizations still view
integration as a complex issue requiring specialist tools.

■ The lack of a clear market leader that is looking to push this technology forward as
the major application vendors look to protect their customer bases.

User Recommendations

Application leaders should:

■ Ask their major application vendors about the interoperability of applications within
their portfolios. This is the area where self-integrating applications are most likely to
emerge first.

■ Investigate integration vendors that have augmented artificial intelligence features to

automate the process of onboarding applications and services into a portfolio.

■ Manage their expectations. Self-integrating applications will provide just enough

integration with the rest of the application portfolio to enable a new application to
work efficiently.

Sample Vendors

Boomi; Informatica; Microsoft; Oracle; Salesforce; SAP; SnapLogic; Workato

Gartner, Inc. | G00747497 Page 8 of 78

Gartner Recommended Reading

Innovation Insight for Self-Integrating Applications

Data Fabrics Add Augmented Intelligence to Modernize Your Data Integration

Integration Strategy Empowerment Team

Analysis By: Massimo Pezzini

Benefit Rating: High

Market Penetration: 5% to 20% of target audience

Maturity: Emerging

Definition

An integration strategy empowerment team (ISET) is responsible for designing,

implementing and delivering an organization’s integration strategy. Hence, it’s a “service
provider” responsible for deploying the integration technology platform; disseminating
best practices; delivering training, support and consulting services; and running an
integration community of practice. It serves different personas — integration specialists,
as well as ad hoc and citizen integrators — across the organization.

Why This Is Important

Most organizations have set up an integration competency center (ICC) focused on

centrally delivering integration projects for the organizational units. Although highly
efficient, this model shows limited ability to scale and meet organizations’ fast and agile
integration needs.

An ISET’s goal is to overcome these limits by empowering units and individual business
users to fulfill integration work in a self-service way by providing them a shared
technology platform and a proper set of services.

Gartner, Inc. | G00747497 Page 9 of 78

Business Impact

■ An ISET enables decentralized units to reduce time to value and increase business
agility by performing integration work by themselves, while keeping their integration
costs under control and ensuring overarching integration governance.

■ Longer term, the ISET will empower composable business by supporting

decentralized fusion teams wishing to collaboratively build new applications by
composing packaged business capabilities in an agile way via orchestration,
integration and automation tools.

Drivers

An ISET empowers a democratized, self-service and multipersona approach to integration

by providing application teams, business units, departments, subsidiaries and business
work teams with:

■ A set of shared integration technologies, typically based on the hybrid integration

platform (HIP) framework.

■ The services (training, consulting, support, mentoring and service desk) required to
take advantage of these shared technologies.

A growing number of midsize and large organizations will pursue a democratized

integration approach, and therefore establish an ISET driven by:

■ The ever-increasing amount of work needed to support the differentiated integration

use cases stemming from business-efficiency-focused initiatives, from digitalization
and from the need to address the transformation required in the postpandemic
world.

■ Application teams’ desire to incorporate self-service integration work in their agile,

DevOps-enabled application delivery processes.

■ Organizations’ growing adoption of composable business, which aims at enabling a

wide range of personas to build new, highly focused and customized applications by
assembling and integrating predefined “building blocks” (or “packaged business
capabilities”).

Gartner, Inc. | G00747497 Page 10 of 78

■ The widespread vendors’ commitment to provide integration platforms designed to
appeal to a range of personas, including developers who occasionally need to
perform integration work (“ad hoc integrators”) and business users (“citizen
integrators”).

■ The increasingly facilitated access to integration capabilities within SaaS

applications and the growing offering of packaged integration processes, which
appeals to ad hoc and citizen integrators.

■ The expanding availability of ISET implementation methodologies and services from

integration platform providers, consulting companies and system integrators.

Obstacles

■ Designing, developing, deploying, managing, maintaining and evolving a suitable

HIP, which requires investments in technology and technical, methodological and
organizational skills.

■ Effectively supporting a potentially large population of “integrators” (up to hundreds

or even thousands) with highly differentiated IT skills (from advanced for integration
specialists to minimal for citizen integrators).

■ Preventing excessive duplication of integration efforts by encouraging knowledge,

best practices and artifact (processes, transformation maps, adapters) sharing
across different units via a “community of practice.”

■ Managing governance and compliance in a highly decentralized environment.

■ The still-limited industry experience, which makes it relatively difficult to find support
and skills to help the ISET set up and quickly climb the relevant organizational,
methodological and technical learning curves.

User Recommendations

■ Establish your ISET by taking into account its size can vary from a few full-time
employees to dozens (or more), depending on your organization’s nature (midsize or
large) and the scale and complexity of your integration challenges.

■ Explicitly and unequivocally position the ISET as a service provider focused on

empowering self-service integration and not as a “regulator” setting rules, processes
and policies.

Gartner, Inc. | G00747497 Page 11 of 78

■ Staff the ISET with personnel with the skills and mindset needed to act as an
enabling entity rather than an “integration factory.”

■ Define what integration personas are in scope (and what are not) to ensure due
diligence for democratized use cases.

■ Establish KPIs to measure the ISET’s ability to help its constituents become more
innovative, creative and empowered via self-service integration.

■ Implement the ISET model in a stepwise approach, which makes it easier to justify
investments in terms of business or technical benefits.

Sample Vendors

Boomi; Informatica; Mindtree; MuleSoft; PACE; Quinnox; SAP; TCS; Wipro

Gartner Recommended Reading

Integration Teams for the Digital Era Must Support New Delivery Models

Ensure Your Integration Strategy Supports Modern Integration Trends

Gartner, Inc. | G00747497 Page 12 of 78

At the Peak
Event Broker PaaS
Analysis By: Yefim Natis, Keith Guttridge

Benefit Rating: High

Market Penetration: 5% to 20% of target audience

Maturity: Adolescent

Definition

Event broker platform as a service (ebPaaS) plays the role of the intermediary in event-
driven architecture (EDA), configuring the event topics, registering event publishers and
subscribers, facilitating capture and distribution of event notifications. Event brokers are
built on message-oriented middleware (MOM, also known as message brokers) that
delivers the essential publish-subscribe capability, then extended with additional EDA-
oriented mediation and governance capabilities.

Why This Is Important

Most of the outcomes of digital business transformation depend, in part, on an

organization’s continuous awareness of relevant business events and its ability to
respond in business real time. Event broker services facilitate detection and distribution of
event notifications to application services for automated response, dashboards for human
action or data stores for further analysis. The alternative to the use of a broker is custom
design, which is less effective, more expensive and a higher risk.

Business Impact

Organizations that are aware of their relevant business ecosystem events are better
prepared to manage unexpected interruptions and capitalize on opportunities in business
moments. They are equipped for broadcasting notable events for simultaneous,
multitargeted response. Event broker services enable organizations’ versatility in
monitoring multiple sources of events and communicating to many responders in parallel,
with strong scalability, integrity and resilience.

Drivers

■ Increased demand for real-time insights drives organizations to manage event

streaming and stream analytics, leading in turn to event brokers for governance and
coordination of event traffic.

Gartner, Inc. | G00747497 Page 13 of 78

■ Increased adoption of Apache Kafka, by both businesses and leading technology
vendors, promotes organizational awareness of the benefits and opportunities that
event-driven application design brings.

■ The migration of business applications to the cloud demands new platforms and
communication infrastructure, driving many organizations to evaluate and adopt
event broker services, paired with integration and API management offerings.

■ The availability of multiple vendors’ ebPaaS, based on open-source standards such

as Pulsar, Kafka, NATS, MQTT and AMQP, provides competitive and differentiated
options in event broker services for better-tuned fit to customers’ use cases.

■ Increased maturity of ebPaaS offerings supports more advanced capabilities in

performance, data and process management, and optimization of event-driven
applications.

■ Most leading SaaS offerings support some event processing, increasing awareness
of benefits and opportunities of event-driven application design in a large number of
mainstream business and government organizations.

■ Open-source event brokers are easier to operate and scale, reducing the cost of early
experimentation with event-driven architecture and attracting more start-ups and
other advanced software engineering teams.

■ The increasing popularity of digital integration hubs and other data consolidation
approaches gets the near-real-time data accuracy when consuming event streams,
instead of database lookups.

Obstacles

■ Desire to keep control of all aspects of infrastructure deployment leads some

organizations to manual implementation of event-driven communications.

■ ebPaaS offerings become too expensive as more proprietary features are added to
help differentiate from the competition.

■ Event broker functionality, embedded in some platform and application services,

fragments control of event streaming across the organization, while delaying a
systematic investment in event brokering.

■ Some software engineering teams use webhooks and websockets tools to set up
event notifications, delaying the full many-to-many experience of EDA that’s
implemented via an event broker technology.

Gartner, Inc. | G00747497 Page 14 of 78

■ Lack of universally supported standards for protocols or APIs for EDA
implementation increases costs and complexity of managing a large event-driven
application infrastructure.

User Recommendations

■ Apply the complementary strengths of service-oriented architecture (SOA) and EDA,

and encourage every new project to consider the combined use of both, as
appropriate in advanced mesh app and service architecture (MASA).

■ Pilot experimental projects using event brokers to gain insight and skills for the
upcoming more advanced projects. Even a basic pub/sub middleware service is
sufficient as a precursor for a full-featured event broker.

■ Give preference to ebPaaS vendors demonstrating the understanding of the full life
cycle of event brokers’ functionality and responsibility.

■ Plan for coordinated use of an event broker and a stream analytics platform. The
technologies are different and are used in combination in most advanced event
broker use cases.

Sample Vendors

Amazon Web Services (AWS); Confluent; Google; IBM; Microsoft; Solace; TIBCO Software;
Vantiq

Gartner Recommended Reading

Innovation Insight for Event ThinkingInnovation Insight for Event BrokersThe 5 Steps
Toward Pervasive Event-Driven ArchitectureThe Impact of Event-Driven IT on API
ManagementApplying Event-Driven Architecture to Modern Application Delivery Use Cases

Choosing Event Brokers: The Foundation of Your Event-Driven Architecture

Event-Driven Architecture
Analysis By: Yefim Natis, Paul Vincent

Benefit Rating: High

Market Penetration: 5% to 20% of target audience

Maturity: Adolescent

Gartner, Inc. | G00747497 Page 15 of 78

Definition

Event-driven architecture (EDA) is a style of application design in which application

components communicate indirectly by passing event notification messages via an
intermediary (an event broker). EDA is a long-standing architecture model. The demands
of digital business, including IoT, continuous intelligence and contextual decisions, are
reintroducing EDA as newly relevant to the current generation of mainstream application
designers and planners, and placing it back onto the Hype Cycle.

Why This Is Important

EDA provides advanced opportunities for scale, extensibility and resilience in applications
through its asynchronous, intermediated, pub/sub design model. Monitoring business and
technical events in real time enables continuous analysis of context for advanced
intelligence in decision management. Organizations that are interested in digital business
innovation will inevitably discover event stream analytics and EDA as powerful
components of their application design.

Business Impact

EDA is a key enabling architecture pattern for a number of leading trends in digital
business. An event-aware organization is more responsive in its ecosystem, more
empathetic in its customer experience and more intelligent in its decision making than a
purely transaction-centric business. Competence in EDA accelerates transition to digital
business. Lacking event awareness, organizations may struggle to support business at
competitive speeds, agility, continuous innovation and cost-efficiency.

Gartner, Inc. | G00747497 Page 16 of 78

Drivers

■ Digital business demands real-time context awareness through stream analytics to

support intelligent business decisions. Applications that adopt EDA become sources
of such context and empower their business decision makers.

■ Cloud-native application architectures, often using microservices principles,

frequently use EDA to implement more flexible and scalable interservice
communication.

■ The popularity of Apache Kafka is creating greater awareness of EDA among

mainstream organizations and their software engineering leaders.

■ Many major application vendors, including Salesforce and SAP, upgraded support of
EDA to their applications and application platforms in recent years, enabling more
intelligent monitoring of business processes.

■ IoT applications use EDA to monitor states of devices. As the use of IoT software
continues to increase, so does the adoption of EDA.

■ All cloud hyperscalers have added or upgraded their support for EDA by adding and
extending their messaging and event brokering services.

■ Application integration continues to gain adoption in mainstream organizations, and

EDA is a popular model for integration design.

Obstacles

■ The lack of productivity and governance tools dedicated to EDA limits the design of
EDA-based applications to more advanced engineering teams, and thus delays
broader adoption

■ The diversity of protocol and API formats and standards for event processing limits
adoption and increases implementation costs.

■ The design principles of EDA are less well-understood by most development teams,
in part because of the familiarity bias in favor of the common and ubiquitous
request/reply model, often implemented using REST APIs.

■ Event-driven communications can deliver only eventual consistency. Applications

that require synchronization of distributed database updates must choose a
different architecture.

Gartner, Inc. | G00747497 Page 17 of 78

User Recommendations

■ Develop an inventory of EDA-related technologies and practices currently deployed;

consolidate and extend EDA capabilities in technology, skills and policies.

■ Aim for a pragmatic mixed use of request-driven APIs following the SOA model and
EDA, including application design, software life cycle and production management.

■ Adopt EDA gradually, as the industry develops required standards, best practices,
and improved productivity design and management tools.

■ Aim to establish EDA, along with SOA, as the common and complementary
architecture patterns, both considered for all application initiatives.

■ Work with business stakeholders to coordinate the discovery and analysis of

business events; aim for synergy in business and technical modelling of event-driven
solutions.

■ Manage and mediate event channels aggressively, and understand that their value
represents an in-motion view of key business processes and happenings.

Gartner Recommended Reading

Innovation Insight for Event Thinking

Innovation Insight for Event Brokers

Maturity Model For Event Driven Architecture

The Impact of Event-Driven IT on API Management

Choosing Event Brokers: The Foundation of Your Event-Driven Architecture

Digital Integration Hub

Analysis By: Massimo Pezzini, Eric Thoo

Benefit Rating: High

Market Penetration: 1% to 5% of target audience

Maturity: Adolescent

Gartner, Inc. | G00747497 Page 18 of 78

Definition

A digital integration hub (DIH) provides API-/event-based data access by aggregating and
replicating multiple system-of-record sources into a low-latency, high-throughput data
management layer that synchronizes with the systems of record via event-driven patterns.
A DIH enables scalable, 24/7 data access; reduces workloads on the systems of record;
and improves business agility. Organizations can reap additional value by leveraging a
DIH in analytics, data integration and composition scenarios.

Why This Is Important

Digital initiatives massively leverage APIs and, increasingly, events, to unlock core
business applications data and business logic. However, their success can be undermined
when traditional integration architectures face severe performance, scalability and
availability issues that often stem from an excessive workload generated in the systems
of record. A DIH is an increasingly popular alternative to these approaches as it is able to
fix these issues while delivering additional benefits.

Business Impact

■ Provides digital application audiences with a rich and responsive experience

■ Reduces the cost of running systems of record or limits the fees paid to SaaS
providers for API consumption

■ Helps enable 24/7 operations

■ Improves business agility and favors composability by decoupling the API layer from
the systems of records

■ Maintains an up-to-date picture of fast-changing data used for analytics-based

services, notification services and data integration

Drivers

DIH architectures are typically used to deliver an API platform featuring a data
management layer between the systems of record and the API service layer itself. In this
way, the inquiry workload generated by the API calls doesn’t hit the systems of record,
which are impacted only when their data must be updated.Therefore, interest in and
adoption of DIH-enabled API platforms is fast growing in:

Gartner, Inc. | G00747497 Page 19 of 78

■ The organizations that want to:

■ Offload the systems of record to reduce their operational costs, optimize

expansive upgrades or reduce the API-limit fees paid to SaaS providers needed
to support the high workload generated by digital applications frontending the
systems of record themselves.

■ Improve customers’ satisfaction by delivering a more responsive and data-rich

user experience.

■ Accelerate the transition to composable business, digital and API economy by

implementing a comprehensive set of APIs and events.

■ The banking, insurance, retail, energy and utilities, higher education, transportation,
hospitality and telecom industry sectors. However, other industries (for example,
government and healthcare) are also showing interest in this architecture as they are
subject to the above-mentioned drivers.

■ Large and midsize organizations with limited skills attracted by vendors addressing
the opportunity by repackaging their technology portfolios in DIH-oriented value
propositions or by coming to the market with packaged DIH-enabled API platforms,
at times focused on specific use cases.

Obstacles

Although the emerging packaged DIHs will make implementation easier and faster, the
technical complexity in current DIH implementation limits its adoption to leading-edge
organizations with sufficiently advanced skills and financial resources.

Such complexity stems from the following issues:

■ Dealing with an architecture still not well known in the industry, which implies a
scarcity of know-how, experience and skills in turn leading to high costs

■ Assembling and managing the varied set of DIH building blocks (API gateways,
application platforms, integration platforms, event brokers, data management and
metadata management tools)

■ Keeping the data management layer in sync with the systems of record by
leveraging event-based integration tools (for example, change data capture)

Gartner, Inc. | G00747497 Page 20 of 78

■ Addressing the data governance issues deriving from the creation of yet another
copy or data structure out of the systems-of-record data

User Recommendations

■ Adopt a DIH-enabled API platform when addressing a combination of the following

requirements:

■ Providing a responsive and rich omnichannel experience for large audiences

(hundreds of thousands of users or greater)

■ Reducing the cost associated with sustaining the API-generated workload

hitting the systems-of-record

■ Enabling API “pull” and event “push” services to access data scattered across
multiple back-end systems

■ Drastically decoupling the API services from the systems-of-records to enable

composable business applications

■ Maintaining an up-to-date “single source of truth” for fast-changing data, which

can be used to provide additional services (for example, custom analytics or
search) or can be analyzed in real time to detect “business moments”

■ Embed DIH initiatives into the overall data hub strategy for governance and
integration to avoid ending up with yet another data silo.

Sample Vendors

Cinchy; Fincons Group; IBM; Informatica; Mia-Platform; Microsoft; Oracle; SAP; Sesam;
Software AG

Gartner Recommended Reading

Innovation Insight: Turbocharge Your API Platform With a Digital Integration Hub

Data Fabric
Analysis By: Ehtisham Zaidi, Robert Thanaraj, Mark Beyer

Benefit Rating: Transformational

Market Penetration: 1% to 5% of target audience

Gartner, Inc. | G00747497 Page 21 of 78

Maturity: Emerging

Definition

A data fabric is an emerging data management design for attaining flexible and reusable
data integration pipelines, services and semantics. A data fabric supports various
operational and analytics use cases delivered across multiple deployment and
orchestration platforms. Data fabrics support a combination of different data integration
styles and leverage active metadata, knowledge graphs, semantics and ML to automate
and enhance data integration design and delivery.

Why This Is Important

A data fabric leverages both traditional and emerging technologies in enterprise

architectural design and evolution. It is composable and supports flexibility, scalability
and extensibility in an infrastructure used by humans or machines across multiple data
and analytics use cases. It abstracts data management infrastructure to disintermediate
any incumbent platforms, and enables data integration and delivery regardless of the
number of on-premises or CSP-based data assets in use.

Business Impact

Organizations benefit as data fabric:

■ Provides insights to data engineers and ultimately automates repeatable tasks in

data integration, quality, data delivery, access enablement and more.

■ Adds semantic knowledge for context and meaning, and provides enriched data
models.

■ Evolves into a self-learning model that recognizes similar data content regardless of
form and structure, enabling broader connectivity to new assets.

■ Monitors data assets on allocated resources for optimization and cost control.

Drivers

■ A data fabric enables tracking, auditing, monitoring, reporting and evaluating data
use and utilization, and data analysis for content, values, veracity of data assets in a
business unit, department or organization. This results in a trusted asset capability.

Gartner, Inc. | G00747497 Page 22 of 78

■ Demand for rapid comprehension and adaptation of new data assets has risen
sharply and continues to accelerate — regardless of the deployed structure and
format. The data fabric provides an operational model that permits use cases, users
and developers to identify when data experience varies from the data expectations
depicted in system designs.

■ A shortage of data management professionals is increasing the demand for

accurate and actively utilized metadata to make system design, data availability and
data trust decisions.

■ Catalogs alone are insufficient in assisting with data self-service. Data fabrics
capitalize on machine learning to resolve what has been a primarily human labor
effort using metadata to provide recommendations for integration design and
delivery.

■ Business delivery and management professionals find it difficult to identify adjacent,

parallel and complementary data assets to expand their analytical models. Data
fabrics have the capability to assist with graph data modeling capabilities (which is
useful to preserve the context of the data along with its complex relationships), and
allow the business to enrich the models with agreed upon semantics.

■ Significant growth in demand and utilization of knowledge graphs of linked data as

well as ML algorithms to provide actionable recommendations and insights to
developers and consumers of data can be supported in a data fabric.

■ Organizations have found that one or two approaches to data acquisition and
integration are insufficient. Data fabrics provide capabilities to deliver integrated
data through a broad range of combined data delivery styles including bulk/batch
(ETL), data virtualization, message queues, use of APIs, microservices and more.

Obstacles

Data fabrics are just past the Peak of Inflated Expectations. The main challenges
surrounding broad adoption are:

■ Diversity of skills and platforms to build a data fabric present both technical and
cultural barriers. It requires a shift from data management based upon analysis,
requirements and design to one of discovery, response and recommendation.

■ Intentional market hype by providers and services organizations purporting a data

fabric delivery is adding to market cynicism.

Gartner, Inc. | G00747497 Page 23 of 78

■ Misunderstanding and lack of knowledge in how to reconcile and manage a data
fabric and a legacy data and analytics governance program that assumes all data is
equal will lead to failure.

■ Proprietary metadata restrictions will hamper the data fabric, which is wholly
dependent upon acquiring metadata from a wide variety of data management
platforms. Without metadata, the fabric requires analytic and machine learning
capabilities to infer missing metadata, and while possible, will be error prone.

User Recommendations

Data and analytics leaders looking to modernize their data management with a data
fabric should:

■ Invest in an augmented data catalog that assists with creating a flexible data model.
Enrich the model through semantics and ontologies for the business to understand
and contribute to the catalog.

■ Invest in data fabrics that can utilize knowledge graph constructs.

■ Ensure subject matter expert support by selecting enabling technologies that allow
them to enrich knowledge graphs with business semantics.

■ Combine different data integration styles into your strategy (bulk/batch, message,
virtualization, event, stream, replication and synchronization).

■ Evaluate existing tools to determine the availability of three classes of metadata:

design/run, administration/deployment and optimization/algorithmic metadata.
Rate existing and candidate platforms and favor those that share the most
metadata.

■ Focus on a similar transparency and availability of metadata between PaaS and

SaaS solutions.

Sample Vendors

Cambridge Semantics; Cinchy; CluedIn; Denodo; IBM; Informatica; Semantic Web

Company; Stardog; Talend

Gartner Recommended Reading

Top Trends in Data and Analytics for 2021: Data Fabric Is the Foundation

Gartner, Inc. | G00747497 Page 24 of 78

What Is Data Fabric Design?

Top Trends in Data and Analytics for 2021: Data Fabric Is the Foundation

Emerging Technologies: Data Fabric Is the Future of Data Management

HIP
Analysis By: Massimo Pezzini

Benefit Rating: High

Market Penetration: 5% to 20% of target audience

Maturity: Adolescent

Definition

The hybrid integration platform (HIP) is an architectural framework that defines

integration and governance capabilities and enables differently skilled personas to tackle
multiple integration use cases across hybrid, multicloud setups. A HIP implementation
typically consists of an assembly of diverse integration tools, from one or more providers,
which are managed as a cohesive, federated and integrated whole, typically by an
integration strategy empowerment team.

Why This Is Important

As organizations pursue digital and composable business initiatives, they find that the
integration challenges they must address are growing in complexity and quantity. Cloud
services, cloud data warehouses, ecosystems, mobile apps and Internet of Things (IoT)
devices are new endpoints that they must integrate with traditional applications and data
sources. The HIP helps software engineering leaders implement the integration and
governance capabilities needed to integrate all their IT assets.

Business Impact

Each organization’s HIP implementation will differ to reflect specific requirements. But in
all cases, it will alleviate integration challenges by:

■ Supporting centralized control and governance, while leveraging decentralized and

collaborative integration delivery

Gartner, Inc. | G00747497 Page 25 of 78

■ Improving business groups’ self-sufficiency and agility by reducing their reliance on
specialist integrators of limited availability

■ Accelerating the time to value for integration-intensive business initiatives

Drivers

A HIP implementation typically consists of an assembly of on-premises and cloud-

delivered Integration platforms, API management platforms, event brokers, metadata
management tools and other use case-specific components, often from different
providers. Despite the complexity of such a setup, a growing number of midsize and large
organizations are implementing HIP-inspired platforms to:

■ Enable a range of diverse integration personas to perform integration work in a self-

service fashion. These personas include: integration specialists (professional
integration developers), “ad hoc” integrators (application developers, SaaS
administrators and business technologists who occasionally have to perform
integration work), and citizen integrators (business users who want to automate
personal or workgroup processes).

■ Integrate a wide variety of endpoints residing in cloud environments, on-premises

data centers, ecosystem partners, and mobile and IoT devices by leveraging APIs,
events and batch mechanisms.

■ Support a differentiated set of use cases, including, but not limited to, application,
data, B2B, process, IoT, API and event integration; robotic process automation; and
digital integration hub

■ Deploy integration platform capabilities in a hybrid, multicloud scenario — that is,

one featuring a combination of public and private clouds and on-premises data
centers — and embed them in applications and edge systems.

Although not all organizations need to address all these requirements, almost all
organizations will have to tackle some of them. Therefore, most midsize, large and global
organizations will have to deploy at least a subset of the capabilities defined in the HIP
framework.

Obstacles

Organizations will face key challenges when implementing an HIP-inspired platform:

Gartner, Inc. | G00747497 Page 26 of 78

■ A growing number of providers have released integrated technology suites mirroring,
at least in part, the HIP framework. In many instances, though, a HIP implementation
requires the aggregation of multiple products from different providers — a daring
technological deployment and skills-building effort for the less technically skilled
organizations.

■ Such a technology aggregation poses operational challenges. Use of a wide range of

product-specific tools leads to suboptimal outcomes and skills duplication. However,
implementing a single, cross-product “control plane” may require notable
investments in technologies and skills.

■ A HIP is often deployed to enable self-service integration by a variety of

organizational units. To avoid chaotic duplication of efforts and high costs, software
engineering leaders must define and enforce well-balanced governance policies.

User Recommendations

Software engineering leaders responsible for integration should:

■ Modernize their strategy by implementing a HIP-inspired infrastructure to enable

collaborative and decentralized integration delivery, carried out by a variety of
personas and addressing diverse use cases.

■ Implement a HIP, If they work for a large organization, by federating different

vendors’ products, instead of buying an out-of-the-box HIP. This will make it easier to
maintain backward-compatibility with in-place integration platforms and mitigate
the risk of single-vendor lock-in.

■ Implement a HIP, If they work for a midsize organization, by adopting an iPaaS,

whenever possible, to reduce the complexity of effort. Many iPaaSs provide a subset
of the HIP framework capabilities that is generally sufficient for such organizations.

■ Adopt a stepwise, initiative-by-initiative HIP implementation strategy, which is much

easier to justify than a “big bang” approach and reduces complexity and risk.

Sample Vendors

Boomi; IBM; Informatica; Jitterbit; MuleSoft; Oracle; SAP; SnapLogic; Software AG; TIBCO
Software

Gartner Recommended Reading

How to Deliver a Truly Hybrid Integration Platform in Steps

Gartner, Inc. | G00747497 Page 27 of 78

How to Justify Strategic Investments in Integration Technology

Digital Integrator Technologies

Analysis By: Eric Thoo, Keith Guttridge

Benefit Rating: Transformational

Market Penetration: 5% to 20% of target audience

Maturity: Emerging

Definition

Digital integrator technologies apply artificial intelligence (AI) techniques, such as

machine learning (ML) and natural language processing (NLP), to assist integration
design and delivery. Areas these technologies focus on include engagement via chatbots
or voice, assistance of flow automation via next best action and intelligent data mapping,
and insight for processing optimization and intelligent platform operations.

Why This Is Important

Digital integrator technologies aim to simplify integration development efforts. By

anticipating user needs and making next-best-step recommendations for designing an
integration flow, inference algorithms identify suitable prepackaged integration content,
help rectify errors in flows and improve performance. Advanced digital integrator
technologies dynamically optimize integration processing and platform operations, with
capabilities to auto-adjust runtime, auditing and self-healing.

Business Impact

■ AI-enabled integration platforms provide automated guidance for integrating

applications and data — thereby enabling less-technical integrator roles, to perform
integration tasks as well as simplifying tasks for integration specialists.

■ Initiatives to modernize integration platforms using AI can adopt a low-code or no-

code paradigm to empower all integrator personas.

Gartner, Inc. | G00747497 Page 28 of 78

Drivers

■ Delivery of integration is becoming pervasive rather than a specialist task. Digital

integrator technologies empower a broad range of specialist, ad hoc and citizen
integrators — thus advancing the notion for democratizing integration and enabling
the composable enterprise (see Future of Applications: Delivering the Composable
Enterprise and The Applications of the Future Will Be Founded on Democratized,
Self-Service Integration).

■ Digital integrator technologies in the form of a conversational user experience

expedites efficiencies in creating integration processes or in monitoring the
operational state of the integration platform.

■ Increasing use by line-of-business teams connects software and makes

independently designed applications and data structures work as integrated
solutions.

Obstacles

■ Rather than a breakthrough moment, organizations deploy incrementally when being

conservative about embracing this evolution.

■ Governance challenges reign when there is limited or no availability of

comprehensive lineage/metadata management capabilities that track the activities
and outcome. It may be difficult to ensure the traceability of integration flows or
potentially substantial, consequential damages created by flawed next best steps
guided by flawed data.

■ Experiences that AI learns come from a variety of integrators, not only specialists but
also citizen integrators who may not offer proven techniques. Poor design practices
that become popular through overuse will misdirect the recommendation engine.

Gartner, Inc. | G00747497 Page 29 of 78

User Recommendations

■ Provide self-guided integration designs in ways that will make implementation of

integration flows easier, faster and less expensive.

■ Support self-service integration tasks by business roles.

■ Target simpler scenarios where past experience can be used to train ML systems in
integration.

■ Apply algorithms to learn and analyze integration processes to autogenerate end-to-

end integration flows, understand the performance characteristics of the services
involved and provide suggestions to optimize the integration process going forward.

Sample Vendors

Boomi; IBM; Informatica; Microsoft; Oracle; SAP; SnapLogic; TIBCO Software; Tray.io;
Workato

Gartner Recommended Reading

Innovation Insight for AI in Integration Technologies

Gartner, Inc. | G00747497 Page 30 of 78

Sliding into the Trough
Data Hub iPaaS
Analysis By: Keith Guttridge, Eric Thoo

Benefit Rating: High

Market Penetration: 5% to 20% of target audience

Maturity: Adolescent

Definition

Data hub integration platform as a service (iPaaS) supports integration between

applications and system endpoints via a centralized intermediary store that persists the
(often-normalized) data before delivery to the destination. This is different to the pass-
through architecture that most established iPaaS offerings utilize today and provides
additional information management and analytics capabilities as well as API access to
the data store.

Why This Is Important

The data landscape for most organizations is fragementing across on-premises

applications and datastores and an ever increasing number of SaaS applications and
cloud data stores. This is causing increasingly complex integration and governance
challenges. iPaaS vendors are converging application and data integration technologies,
including data stores, to increase the appeal of their offerings as a one-stop shop for all
integration needs.

Business Impact

Data Hub iPaaS provides the following benefits:

■ Simplification of integrating applications and data sources.

■ Improved data management and data governance.

■ Improved resilience of the production environment with record/replay capabilities for

integration errors and system availability.

■ Simplified access to the centralized data model via APIs instead of connecting direct
to application APIs.

Gartner, Inc. | G00747497 Page 31 of 78

■ Analytics of the data in real time to gain business insight and potentially enable
business moments.

Drivers

■ Organizations looking to improve data management across on-premises and cloud

based applications and data sources.

■ Organizations looking to build a customer engagement hub.

■ Organizations looking to build a digital integration hub.

■ Organizations looking to create a hybrid transactional analytics platform.

■ Organizations looking to reduce the number of vendors providing integration and

data management technologies.

■ iPaaS vendors converging various integration technologies.

Obstacles

■ Regional compliance policies for data stores.

■ Industry compliance policies for data stores .

■ Organizations compliance policies for data stores.

■ Preference for best-of-breed integration and data management technology.

■ Organizational structure impeding unified approach to integration and data

management.

■ Vendor landscape is mostly small startups with only a handful of large vendors
providing this service.

User Recommendations

■ Recognize that this is currently still a relatively new market. The few vendors that do
provide this capability often do so for relatively niche use cases. It may take several
years before data hub iPaaS becomes general-purpose enough for most clients.
Given that the data is stored within the data hub iPaaS, this brings with it extra
challenges such as security, resilience and compliance that regular iPaaS vendors do
not have to worry about.

Gartner, Inc. | G00747497 Page 32 of 78

■ Combine offerings from several technology categories and vendors (such as iPaaS +
data store + analytics) if the current offerings in the data hub iPaaS market are not
suitable for your needs. Once established though, the combination of iPaaS, data
management, real-time analytics and machine learning has the potential to
significantly disrupt how organizations integrate their application and data
portfolios as well as their B2B partners.

Sample Vendors

Cinchy; Domo; ForePaaS; IBM, Informatica; SAP; Sesam

Gartner Recommended Reading

 Magic Quadrant for Enterprise Integration Platform as a Service

Magic Quadrant for Data Integration Tools

Packaged Integration Processes

Analysis By: Abhishek Singh

Benefit Rating: High

Market Penetration: 5% to 20% of target audience

Maturity: Adolescent

Definition

Packaged integration processes (PIPs) are predefined integration solutions designed to

automate and standardize common business processes that require integration of
multiple endpoints. Examples include PIPs for order to cash, hire to retire and procure to
pay. Multiple integration platform and SaaS providers, as well as service providers, make
available a portfolio of PIPs to support ERP, HCM, CX and other integration requirements
for both cloud and on-premises applications.

Why This Is Important

Packaged integration processes are important for delivering integrations faster in

common business scenarios as they provide near to complete out-of-the-box solutions.
However, they are not plug-and-play solutions and might require some customization
work. PIPs help reduce time to value for “commodity” integration scenarios, increase
productivity in integration delivery and facilitate democratized, self-service integration.

Gartner, Inc. | G00747497 Page 33 of 78

Business Impact

The notion of PIPs has been in the market for many years, often known as “accelerators”
or “recipes” for integration between applications. However, PIPs are now becoming
popular because of the prevalence of SaaS offerings. PIPs enable ad hoc and citizen
integrators, as well as integration specialists, to deliver integration, thus enabling the self-
service model of integration delivery.

Drivers

■ Recently, the total number of vendors providing PIPs as a part of their offerings has
steadily increased, primarily driven by customers' adoption of SaaS offerings.
Examples include S4/HANA integration with Salesforce, NetSuite integration with
Shopify and many more.

■ In addition to the increased investment in PIPs by vendors, a key factor contributing

to hype is the growing customer demand for very fast integration, which is shifting
the integration development toward practices that facilitate citizen integrators and
therefore promote the use of PIPs.

■ PIP offerings are especially attractive to midsize organizations and LOBs of large
organizations that have limited IT skills and cannot handle overly complex
integration requirements. Many of these organizations see PIPs as a way to rapidly
deliver integrations without investing massively in new skills.

Obstacles

The obstacles to the adoption of PIPs are:

■ Some of the PIPs provided by the vendors pose a risk of vendor lock, which can
negatively affect your integration strategy.

■ Lack of flexibility of the PIP, leading to rigid integrations. Perfectly acceptable for
nondifferentiating use cases, if you want to integrate the same as everyone else, but
you have to change your working practices to match the PIPs.

User Recommendations

Leverage PIPs when trying to automate the integration of common and undifferentiating
business processes to reduce implementation costs and accelerate time to value.
Software engineering leaders responsible for integration should:

Gartner, Inc. | G00747497 Page 34 of 78

■ Identify business application integration processes in the backlog that are
nondifferentiating, and can therefore be implemented as quickly and inexpensively
as possible via PIPs.

■ Empower new integration personas, such as ad hoc and citizen integrators, to take
on responsibility for integration in part by providing them with approved PIPs that
they can customize and deploy themselves.

■ Test your ability to implement a PIP efficiently and effectively by performing a proof
of concept (POC) for each PIP identified for implementation.

Sample Vendors

Boomi; Celigo; Informatica; Jitterbit; Mulesoft; Oracle; SAP; Snaplogic; Workato; Zapier

Gartner Recommended Reading

Accelerate you Integration Delivery by Using Packaged Integration Processes

Innovation Insight for AI in Integration Technologies

Choose the Best Integration Tool for Your Needs Based on the Three Basic Patterns of
IntegrationChoosing Application Integration Platform TechnologyToolkit: RFP Templates
for Application Integration Platforms

Service Mesh
Analysis By: Anne Thomas

Benefit Rating: Moderate

Market Penetration: 1% to 5% of target audience

Maturity: Adolescent

Definition

A service mesh is a distributed computing middleware that optimizes communications

between application services within managed container systems. It provides lightweight
mediation for service-to-service communications, and supports functions such as
authentication, authorization, encryption, service discovery, request routing, load
balancing, self-healing recovery and service instrumentation.

Gartner, Inc. | G00747497 Page 35 of 78

Why This Is Important

A service mesh is lightweight middleware for managing and monitoring service-to-service

(east-west) communications, especially among microservices running in ephemeral
managed container systems, such as Kubernetes. It provides visibility into service
interactions, enabling proactive operations and faster diagnostics. It automates complex
communication concerns, thereby improving developer productivity and ensuring that
certain standards and policies are enforced consistently across applications.

Business Impact

■ A service mesh helps ensure resilient and secure request-response communication

between services deployed in Kubernetes and other managed container systems.

■ Service mesh middleware is one of many management technologies that provide

software infrastructure for distributed applications deployed in managed container
systems.

■ This type of middleware, along with other management and security middleware,
helps provide a stable environment that supports “Day 2” operations of
containerized workloads.

Drivers

■ Service mesh adoption is closely aligned with microservices architectures and

managed container systems like Kubernetes. Service mesh supports needed
functionality in ephemeral environments, such as service discovery and mutual
Transport Layer Security between services.

■ As microservice deployments scale and grow more complex, DevOps teams need
better ways to track operations, anticipate problems and trace errors. Service mesh
automatically instruments the services and feeds logs to visualization dashboards.

■ A service mesh implements the various communication stability patterns (including

retries, circuit breakers and bulkheads) that enable applications to be more self-
healing.

■ Many managed container systems now include a service mesh, inspiring DevOps
teams to use it. The hyperscale cloud vendors provide a service mesh that is also
integrated with their other cloud-native services.

■ Independent vendors, such as Buoyant, HashiCorp and Kong provide service meshes
that support multiple environments.

Gartner, Inc. | G00747497 Page 36 of 78

Obstacles

■ Service mesh technology is immature and complex, and most development teams
don’t need it. It can be useful when deploying microservices in Kubernetes, but it’s
never required.

■ Users are confused by the overlap in functionality among service meshes, ingress
controllers, API gateways and other API proxies. Management and interoperability
among these technologies hasn’t yet been addressed by the vendor community.

■ Many people associate service mesh exclusively with Istio, even though it isn’t the
most mature product in the market and has a reputation for complexity.

■ Independent service mesh solutions face challenges from the availability of

platform-integrated service meshes from the major cloud and platform providers.

User Recommendations

■ Delay adoption of service mesh until your teams start building applications that will
get value from a mesh, such as applications deployed in managed container
systems with a large number of service-to-service (east-west) interactions.

■ Favor the service meshes that come integrated with your managed container system
unless you have a requirement to support a federated model.

■ Reduce cultural issues and turf wars by assigning service mesh ownership to a
cross-functional PlatformOps team that solicits input and collaborates with
networking, security and development teams.

■ Accelerate knowledge transfer and consistent application of security policies by

collaborating with I&O and security teams that manage existing API gateways and
application delivery controllers.

Sample Vendors

Amazon Web Services; Buoyant; Decipher Technology Studios; Envoy; F5; Google;
HashiCorp; Istio; Kong; Microsoft; Red Hat; Solo.io; Tetrate; VMware

Gartner Recommended Reading

 How a Service Mesh Fits Into Your API Mediation Strategy

 Assessing Service Mesh for Use in Microservices Architectures

Gartner, Inc. | G00747497 Page 37 of 78

 Emerging Technology Analysis: Service Mesh

Microservices
Analysis By: Anne Thomas

Benefit Rating: High

Market Penetration: 5% to 20% of target audience

Maturity: Adolescent

Definition

A microservice is a tightly scoped, strongly encapsulated, loosely coupled, independently

deployable and independently operated application service. Microservices architecture
(MSA) applies the principles of service-oriented architecture (SOA), DevOps and domain-
driven design (DDD) to the delivery of distributed applications. MSA has three core
objectives: Continuous delivery, precise scalability and improved stability.

Why This Is Important

Microservices architecture promises powerful application agility, scalability and resilience.

It is a way to build cloud-native applications, and it facilitates continuous delivery
practices. But the architecture is complex, with disruptive cultural and technical impacts.
Misconceptions about microservices often push software engineering teams to use them
indiscriminately, leading to overly complex architectures that fail to deliver anticipated
benefits and often make things worse.

Business Impact

■ Microservices increase business agility by enabling teams to incrementally deliver

new features and capabilities in their software products in response to changing
business requirements.

■ Microservices improve the scalability of the software engineering organization by

enabling small teams to work independently to deliver different services within an
application.

■ Microservices allow teams to change one part of an application, without the delay
and cost of changing the entire application.

Gartner, Inc. | G00747497 Page 38 of 78

Drivers

■ Software engineering teams adopt microservices architecture to facilitate a

continuous delivery practice. The architecture must be combined with strong DevOps
practices to enable teams to safely deploy small, independent features to production
systems at the frequency at which they are delivered.

■ When applied well, the architecture increases the independence of different parts of
a large application, enabling multiple development teams to work autonomously and
on their own schedules.

■ Microservices architecture facilitates the building of cloud-native applications that

support robust scalability and resiliency requirements.

■ Microservices are frequently deployed in managed container systems, which can

dynamically scale service instances in response to load requirements and
automatically recover services that have failed.

■ When combined with chaos engineering and resiliency practices, microservices

architecture enables self-healing systems that can continue to operate through
partial outages.

Obstacles

■ Microservices architecture and its benefits are often misunderstood, and many
software engineering teams struggle to deliver outcomes that meet senior
management expectations. For example, microservices should not be shared, and
they will not save you money.

■ If you aren’t trying to implement or improve your continuous delivery practice, you
will almost certainly be disappointed with the microservices cost-benefit equation.

■ Microservices architecture is complex. Developers must acquire new skills and adopt
new design patterns and practices to achieve its benefits.

■ Microservices disrupt traditional data management models.

■ Microservices require new infrastructure.

■ Microservices are related to but not the same as APIs or containers.

■ Many software engineering leaders underestimate the cultural prerequisites. Success

depends on applying mature agile and DevOps practices and changing team
structures to align with service domains.

Gartner, Inc. | G00747497 Page 39 of 78

User Recommendations

■ Set clear expectations by defining business goals and objectives for microservices
architecture adoption based on realistic cost-benefit analysis of the architecture.

■ Use microservices architecture as a tool to help you attain those goals. Don’t view
microservices as a destination.

■ Avoid “microservice washing” conventional SOA, three-tier architecture and

integration. Recognize the difference.

■ Improve outcomes by creating guidelines for where and when software engineering
teams should and should not use microservices architecture.

■ Keep application architecture as simple as possible to achieve your goals.

■ Address cultural concerns by aligning teams along business domain boundaries,

investing in distributed computing architecture skills and improving DevOps
practices.

Gartner Recommended Reading

Leading Teams to Success with Microservices Architecture

Designing Services and Microservices to Maximize Agility

10 Ways Your Microservices Adoption Will Fail — and How to Avoid Them

Serverless fPaaS
Analysis By: Anne Thomas

Benefit Rating: Low

Market Penetration: 5% to 20% of target audience

Maturity: Early mainstream

Gartner, Inc. | G00747497 Page 40 of 78

Definition

Function platform as a service (fPaaS), also known as function as a service (FaaS), is a

serverless execution platform for event-triggered application components known as
functions. Like all serverless platforms, fPaaS enables you to run code without
provisioning or managing the underlying system or application infrastructure. fPaaS
pricing models allow users to pay in microincrements only for actual usage, rather than
preprovisioned resources required to support projected peak loads.

Why This Is Important

fPaaS can deliver significant savings for certain types of workloads via its consumption-
based micropricing model. The programming model also enables software engineers to
rapidly deploy and configure new functions with little or no assistance from operations
teams.

Business Impact

■ An fPaaS can offer significant cost savings and virtually unlimited scalability for
applications with highly variable capacity requirements.

■ Serverless technologies like fPaaS abstract and commoditize infrastructure

technologies, increasing developer and operator productivity, enabling organizations
to respond rapidly to digital business moments and deliver new applications and
features faster.

Drivers

fPaaS is gaining momentum because of:

■ Potential cost savings — The micropricing model charges for small increments of
compute time per invocation, which can be advantageous for small, spiky
workloads. The model is less favorable for large, consistent workloads.

■ Rapid solution delivery — The serverless model reduces the amount of work
developers and operations teams need to do to build, deploy and configure
solutions.

■ Integration with hyperscale xPaaS — The hyperscale vendors make it easy to use
their fPaaS with their other cloud-native xPaaS offerings.

■ Broad use-case support — fPaas can support a broad spectrum of application use
cases, from basic websites to complex analytical processes.

Gartner, Inc. | G00747497 Page 41 of 78

■ Edge computing efficiencies — Deploying functions at the edge enables processing
close to the source.

■ Embedding within other xPaaS — Some xPaaS vendors embed an fPaaS in their
platforms to host code components, such as rules and workflow routines. Examples
include the InRule decision management platform and the Zoho low-code
application platform. These systems hide the complexity of the fPaaS programming
and operating model.

Obstacles

fPaaS is facing challenges because:

■ Cost savings don’t always materialize — fPaaS pricing isn’t favorable for
applications with consistently high invocation rates. Also, fPaaS-based applications
often require other xPaaS, such as API management, data management and
notifications.

■ Latency — fPaaS-based applications can suffer from cold-start issues.

■ Lock-in — fPaaS-based applications aren’t portable across vendor solutions.

■ Resource constraints — fPaaS is inappropriate for memory- and compute-intensive

workloads.

■ Lacking infrastructure — DevOps teams require development frameworks, testing

and debugging tools, security services, and management technology. A fledgling
ecosystem is emerging to address these requirements, although most ecosystem
players focus only on Amazon Web Services (AWS) Lambda, and tooling for other
fPaaS offerings is limited.

■ Alternative solutions — Many developers prefer to use more general-purpose

platforms, such as Kubernetes or low-code application platforms.

Gartner, Inc. | G00747497 Page 42 of 78

User Recommendations

■ Minimize vendor lock-in by ensuring that your software engineering teams don’t limit
their skills and practices to the proprietary features of a single fPaaS.

■ Estimate fPaaS costs based on expected invocation rates, memory requirements,

execution times and other xPaaS dependencies. Don’t presume that fPaaS is always
a less expensive option.

■ Evaluate whether fPaaS is a good fit for your applications and your teams’
development skills. Consider aPaaS or managed container solutions as alternatives.

■ Identify use cases where fPaaS offers a strategic benefit from a cost or agility
perspective. Consider fPaaS for microservices deployments and for applications
with highly variable or unpredictable capacity requirements.

■ Avoid using fPaaS for high-memory or compute-intensive workloads. Don’t port

existing monolithic applications to fPaaS.

■ Use fPaaS if it’s an integral part of another solution, such as edge computing,
decision management or low code.

Sample Vendors

Amazon Web Services; Cloudflare; Google; InRule; Microsoft; Netlify; Red Hat; Vercel; Zoho

Gartner Recommended Reading

A CIO’s Guide to Serverless Computing

Security Considerations and Best Practices for Securing Serverless PaaS

Decision Point for Selecting Virtualized Compute: VMs, Containers or Serverless

Citizen Integrator Tools

Analysis By: Massimo Pezzini, Tim Faith

Benefit Rating: Moderate

Market Penetration: 5% to 20% of target audience

Maturity: Early mainstream

Gartner, Inc. | G00747497 Page 43 of 78

Definition

Citizen integrator tools are typically cloud-hosted services providing very intuitive, no-code
integration process development tools. This way expert business users with minimal IT
skills can handle relatively simple application, data and process integration tasks (or
“automations”) by themselves. Citizen integrator tools also provide a rich set of packaged
integration processes (PIPs) that business users can rapidly configure and run with no
assistance from integration specialists.

Why This Is Important

Organizations must address a growing amount of integration challenges in shorter and

shorter timeframes, which implies having at their disposal several “integrators” equipped
with high productivity tools.

Citizen integrator tools enable business users with minimal IT skills to perform self-service
integration work, thus increasing the organization’s overall delivery capacity. However their
ungoverned proliferation can lead to security and compliance risks and duplicated costs.

Business Impact

Citizen integrator tools enable business users to automate tasks currently integrated via
slow and error-prone manual methods. Integration specialists or ad hoc integrators
(developers, SaaS administrators), also use these tools to quickly sort out simple tasks
instead of using more powerful, but expensive and complex tools. Therefore, citizen
integrator tools contribute to improving organizations’ efficiency, productivity, agility and
innovation by reducing the relevant integration costs.

Gartner, Inc. | G00747497 Page 44 of 78

Drivers

■ Citizen integrator tools may help deliver business value faster, reduce integration
costs and support tactical or strategic digital initiatives. These outcomes are
achieved by enabling rapid, pervasive integration by a wide range of employees
within (and potentially also outside) the organization. However they are available in
many forms, which address different markets and needs: PIPs — At times called
“recipes,” these are prepackaged and configurable sets of integration flows, available
stand-alone (at times for free), as embedded capabilities in SaaS or as add-ons to
integration platforms. As such buyers are typically application owners or SaaS
administrators. Integration software as a service (iSaaS) — Cloud services that
enable users to implement brand new PIPs and to deploy, run and customize existing
ones. They are typically sold to individual business users or work teams. Integration
platform as a service (iPaaS) — These are targeted to professional integrators, but
several iPaaS provide an iSaaS-like development environment and/or make
available collections of configurable PIPs atop their platform.

■ iSaaS tools have achieved notable traction in the consumer and SMB markets,
thanks to their very low cost of entry, intuitive user experience, low skills demand and
their rich set of PIPs. However, they have failed to penetrate other segments due to
their lack of enterprise capabilities and services (for example, consulting).

■ PIPs and iPaaS providing citizen-integrator-oriented capabilities are becoming more

and more popular in midsize, large and global organizations. The growing use of AI,
ML, NLP and chatbots in iPaaS offerings to facilitate integration development is
augmenting their appeal for citizen integrators, thus further favoring adoption.

Gartner, Inc. | G00747497 Page 45 of 78

Obstacles

■ Business users are increasingly technology savvy and often driven by time-to-market
pressures, especially in the post-pandemic era that requires fast reaction to sudden
changes in the business environment. This will increasingly urge them to adopt
cloud citizen integrator tools, rather than wait for their IT colleagues to methodically
perform integration work for them. However, this will create a few challenges: If not
framed in a proper governance model, citizen integrator tools adoption by business
users will inevitably lead to security, compliance, management and governance
issues.

■ Although some central IT departments will adopt a positive attitude and proactively
address these challenges, others will try to stop business users from leveraging
these tools to prevent these risks. In addition, excessive expectations for ultra-easy,
super-fast integration and the simplistic nature of some citizen integrator tools may
still lead to disappointment, thus hindering their more widespread adoption.

User Recommendations

Software engineering leaders responsible for integration should:

■ Engage with business teams to understand their automation needs and identify to
what extent citizen integrator tools can improve their responsiveness and
productivity.

■ Approve, certify and support a set of citizen integrator tools that meet these needs
and make them available to internal users in a self-service way. This will help to
prevent the uncontrolled proliferation of similar tools and maintain a degree of
centralized governance and monitoring.

■ Beware when selecting citizen integrator tools that: some tools are rather simplistic
and lowest-common-denominator in nature; and PIPs provided by SaaS vendors
may have been designed for a professional IT developer audience.

■ Give preference to providers that can support both “professional” and citizen
integrator requirements when selecting an iPaaS.

■ Frame citizen integrator tools, including those embedded in SaaS applications, in

your hybrid integration platform (HIP) strategies.

Gartner, Inc. | G00747497 Page 46 of 78

Sample Vendors

Adeptia; Celonis (Integromat); elastic.io; IFTTT; Microsoft; Quickbase; Tray.io; Workato;

Zapier

Gartner Recommended Reading

Accelerate Your Integration Delivery by Using Packaged Integration Processes

The Applications of the Future Will Be Founded on Democratized, Self-Service Integration

Quick Answer: When to Use (or Not Use) Embedded Integration Features Provided by Your
SaaS Vendor

Gartner, Inc. | G00747497 Page 47 of 78

Climbing the Slope
Event Stream Processing
Analysis By: W. Roy Schulte, Pieter den Hamer

Benefit Rating: Transformational

Market Penetration: 20% to 50% of target audience

Maturity: Early mainstream

Definition

Event stream processing (ESP) is computing that is performed on streaming data

(sequences of event objects) for the purpose of stream analytics or stream data
integration. ESP is typically applied to data as it arrives (data “in motion”). It enables
situation awareness and near-real-time responses to threats and opportunities as they
emerge, or it stores data streams for use in subsequent applications.

Why This Is Important

ESP is a key enabler of continuous intelligence and related real-time aspects of digital
business. ESP’s data-in-motion architecture is a radical departure from conventional data-
at-rest approaches that historically dominated computing. ESP products have progressed
from niche innovation to proven technology and now reach into the early majority of
users. ESP will reach the Plateau of Productivity within several years and eventually be
adopted by multiple departments within every large company.

Business Impact

ESP transformed financial markets and became essential to telecommunication networks,

smart electrical grids and some IoT, supply chain, fleet management, and other
transportation operations. Most of the growth in ESP during the next 10 years will come
from areas where it is already established, especially IoT and customer experience
management. Stream analytics from ESP platforms provides situation awareness through
dashboards and alerts, and detects anomalies and other significant patterns.

Drivers

Five factors are driving ESP growth:

Gartner, Inc. | G00747497 Page 48 of 78

■ Companies have ever-increasing amounts of streaming data from sensors, meters,
digital control systems, corporate websites, transactional applications, social
computing platforms, news and weather feeds, data brokers, government agencies
and business partners.

■ Business is demanding more real-time, continuous intelligence for better situation

awareness and faster, more-precise and nuanced decisions.

■ ESP products have become widely available, in part because open-source ESP
technology has made it less expensive for more vendors to offer ESP. More than 40
ESP platforms or cloud ESP services are available. All software megavendors offer
at least one ESP product and numerous small-to-midsize specialists also compete in
this market.

■ ESP products have matured into stable, well-rounded products with many thousands
of applications (overall) in reliable production.

■ Vendors are adding expressive, easy-to-use development interfaces that enable

faster application development. Power users can build some kinds of ESP
applications through the use of low-code techniques and off-the-shelf templates.

Obstacles

■ ESP platforms are overkill for most applications that process low or moderate
volumes of streaming data (e.g., under 1000 events per second), or do not require
fast response times (e.g., less than a minute).

■ Many ESP products required low-level programming in Java, Scala or proprietary

event processing languages until fairly recently. The spread of SQL as a popular ESP
development language has ameliorated this concern for some applications,
although SQL has limitations. A new generation of low-code development
paradigms has emerged to further enhance developer productivity but is still limited
to a minority of ESP products.

■ Many architects and software engineers are still unfamiliar with the design
techniques and products that enable ESP on data in motion. They are more familiar
with processing data at rest in databases and other data stores, so they use those
techniques by default unless business requirements force them to use ESP.

User Recommendations

Gartner, Inc. | G00747497 Page 49 of 78

■ Use ESP platforms when conventional data-at-rest architectures cannot process
high-volume event streams fast enough to meet business requirements.

■ Acquire ESP functionality by using a SaaS offering, IoT platform or an off-the-shelf

application that has embedded CEP logic if a product that targets their specific
business requirements is available.

■ Use vendor-supported closed-source platforms or open-core products that mix open-

source with value-added closed-source extensions for mainstream applications that
require enterprise-level support and a full set of features. Use free, community-
supported, open-source ESP platforms if their developers are familiar with open-
source software and license fees are more important than staff costs.

■ Use ESP products that are optimized for stream data integration to ingest, filter,
enrich, transform and store event streams in a file or database for later use.

Sample Vendors

Amazon; Confluent; Google; IBM; Informatica; Microsoft; Oracle; SAS; Software AG; TIBCO
Software

Gartner Recommended Reading

Market Guide for Event Stream Processing

Adopt Stream Data Integration to Meet Your Real-Time Data Integration and Analytics
Requirements

Market Share Analysis: Event Stream Processing (ESP) Platforms, Worldwide, 2020

Cloud Native Architecture

Analysis By: Anne Thomas

Benefit Rating: Moderate

Market Penetration: 5% to 20% of target audience

Maturity: Adolescent

Gartner, Inc. | G00747497 Page 50 of 78

Definition

Cloud native architecture is the set of application architecture principles and design
patterns that enables applications to fully utilize the agility, scalability, resiliency, elasticity,
on-demand and economies of scale benefits provided by cloud computing. Cloud native
applications are architected to be latency-aware, instrumented, failure-aware, event-driven,
secure, parallelizable, automated and resource-consumption-aware (LIFESPAR).

Why This Is Important

Many organizations are moving to cloud native architecture as they shift their application
workloads to cloud native application platforms. Cloud native principles and patterns
enable applications to operate efficiently in a dynamic environment and make the most of
cloud benefits. Organizations that simply “lift and shift” legacy applications to cloud
native platforms often find that the applications perform poorly, consume excessive
resources and aren’t able to fail and recover gracefully.

Business Impact
■ Cloud native architecture ensures that applications can take full advantage of a
cloud platform’s capabilities to deliver agility, scalability and resilience.

■ It enables DevOps teams to more effectively use cloud self-service and automation
capabilities to support continuous delivery of new features and capabilities.

■ It can also improve system performance and business continuity, and it can lower
costs by optimizing resource utilization.

Drivers

■ Organizations want to make the most of cloud computing to support their digital
business initiatives, but they can’t fully exploit cloud platform benefits without cloud
native architecture.

■ Software engineering teams are adopting cloud native architecture to support cloud
native DevOps practices, including self-service and automated provisioning,
blue/green deployments, and canary deployments. A basic set of rules known as the
“ twelve-factor app” ensures that applications can support these practices.

■ Cloud native architecture includes practices such as application decomposition

(following the mesh app and service architecture [MASA] structure), containerization,
configuration as code, and stateless services.

Gartner, Inc. | G00747497 Page 51 of 78

Obstacles
■ Cloud native architecture adds a level of complexity to applications, and
development teams require new skills, new frameworks, and new technology to be
successful.

■ Without proper education, architects and developers can apply the principles poorly
and deliver applications that fail to deliver the expected benefits. This leads to
developer frustration in adopting the new patterns and practices.

■ Not every cloud-hosted application needs to be fully cloud-native and developers

may be confused about when they need to use particular patterns to address their
specific application requirements.

User Recommendations

■ Use the twelve-factor app rules and the LIFESPAR architecture principles to build
cloud native applications.

■ Incorporate cloud native design principles in all new applications irrespective of

whether you currently plan to deploy them in the cloud. All new applications should
be able to safely run on a cloud platform, even if it doesn’t fully utilize cloud
characteristics.

■ Apply cloud native design principles as you modernize legacy applications that you
plan to port to a cloud platform to ensure that they can tolerate ephemeral or
unreliable infrastructure. Otherwise, they are likely to experience stability and
reliability issues.

■ Select an application platform that matches your cloud native architecture maturity
and priorities. Recognize that low-code platforms enable rapid development of
cloud-ready applications, but they won’t provide you with the full flexibility to apply
LIFESPAR and twelve-factor principles.

Sample Vendors

Amazon Web Services; Google; Microsoft; Red Hat; VMware

Gartner Recommended Reading

How to Help Software Engineering Teams Modernize Their Application Architecture Skills

How to Modernize Your Application to Adopt Cloud-Native Architecture

Gartner, Inc. | G00747497 Page 52 of 78

A Guidance Framework for Modernizing Java EE Applications

Guidance Framework for Modernizing Microsoft .NET Applications

MASA
Analysis By: Anne Thomas

Benefit Rating: High

Market Penetration: 5% to 20% of target audience

Maturity: Early mainstream

Definition

Mesh app and service architecture (MASA) is a composition-based application

architecture that enables application delivery teams to respond rapidly to changing
business demands and support multiple experiences. A MASA application is implemented
as a mesh of distributed, loosely coupled, autonomous and shareable components,
including multiple fit-for-purpose apps supporting unique experiences and composable
multigrained back-end services. Apps and services communicate via mediated APIs.

Why This Is Important

MASA describes the foundation for modern business application architecture. It enables
multiexperience applications. It supports agility and rapid delivery of new capabilities. It is
a technical architecture that enables composability. It facilitates incremental
modernization of legacy applications while providing mechanisms that ensure security
and robust operations.

Business Impact

MASA enables organizations to respond rapidly to opportunities and disruptions through

extension and recomposition. It enables multiexperience. It enables cloud-native
architectures. MASA is an architecture for individual applications, as well as a strategy for
modernizing the application portfolio. It provides an evolutionary approach that enables
development teams to iteratively modernize their applications in direct response to
business priorities.

Gartner, Inc. | G00747497 Page 53 of 78

Drivers

The initial impetus to shift to MASA was to enable existing applications to add support for
mobile experiences. But MASA enables many other critical application capabilities, such
as:

■ Multiple experiences for different types of devices and modalities, such as voice,
touch, wearables and immersive technologies.

■ Distinct, optimized experiences for the different personas that use an application.

■ Rapid response to disruptive events and changing business priorities via

composition of existing services and creation of new experiences.

■ Greater flexibility through loose coupling of components.

■ Improved application performance, scalability, security and resilience through

intelligent mediation.

Obstacles
■ The biggest obstacle to MASA is the extensive technical debt embedded in existing
application portfolios.

■ MASA requires application functionality to be encapsulated and exposed via APIs.

Those legacy applications must be modernized and refactored to convert the
embedded business logic into composable services.

■ The architecture enables iterative modernization, but it will take years (perhaps
decades) to modernize the entire application portfolio.

■ MASA also requires an investment in API mediation and multiexperience

technologies.

User Recommendations

Software engineering leaders responsible for architecture and infrastructure:

■ Ensure that development teams have competence in user experience design, service-
oriented architecture, API design and domain-driven design.

■ Task your architects with updating existing technical architectures, governance

mechanisms and success metrics to align them with using a MASA approach to
modernize application delivery.

Gartner, Inc. | G00747497 Page 54 of 78

■ Analyze your business’s digital transformation roadmap and identify and prioritize
applications to modernize to support those needs.

■ Encapsulate data and functionality in existing applications and expose them via
APIs to enable composition.

■ Mediate API traffic to apply governance, performance and security policies.

■ Take a pragmatic approach to creating services: encapsulate, extend or refactor

existing applications or build new services.

■ Determine appropriate service granularity based on your objectives. Don’t presume

that all services within MASA must be microservices.

Gartner Recommended Reading

Adopt a Mesh App and Service Architecture to Power Your Digital Business

3 Key Practices to Enable Your Multiexperience Development Strategy

Leading Teams to Success with Microservices Architecture

Mediated APIs: An Essential Application Architecture for Digital Business

How to Apply Design and Architecture to Multiexperience Application Development

Accelerate Digital Transformation With an API-Centric (Headless) Architecture for

Enterprise Applications

Full Life Cycle API Management

Analysis By: Shameen Pillai

Benefit Rating: High

Market Penetration: More than 50% of target audience

Maturity: Early mainstream

Gartner, Inc. | G00747497 Page 55 of 78

Definition

Full life cycle API management involves the planning, design, implementation, testing,
publication, operation, consumption, versioning and retirement of APIs. API management
tools enable API ecosystems and publishing APIs that securely operate and collect
analytics for monitoring and business value reporting. These capabilities are typically
packaged as a combination of developer portal, API gateway, API design, development
and testing tools as well as policy management and analytics.

Why This Is Important

APIs are widely used and accepted as the primary choice to connect systems, applications
and things to build modern composable software architectures. The use of APIs as digital
products monetized directly or indirectly is also on the rise. Advancing digital
transformation initiatives across the world have emphasized the need for creation,
management, operations and security of APIs and made full life cycle API management
an essential foundational capability every organization must have.

Business Impact

Full life cycle API management provides the framework and tools necessary to manage
and govern APIs that are foundational elements of multiexperience applications,
composable architectures and key enablers of digital transformations. It enables the
creation of API products, which may be directly or indirectly monetized, while its security
features serve to protect organizations from the business impact of API breaches.

Gartner, Inc. | G00747497 Page 56 of 78

Drivers
■ Organizations are facing an explosion of APIs, stemming from the need to connect
systems, devices and other businesses. Use of APIs in internal, external, B2B, private
and public sharing of data is driving up the need to manage and govern APIs using
full life cycle API management.

■ APIs that package data, services and insights are increasingly being treated as
products that are monetized (directly or indirectly) and enable platform business
models. Full life cycle API management provides the tooling to treat APIs as
products.

■ Digital transformation drives increased use of APIs, which in turn increases the
demand for API management.

■ APIs provide the foundational elements required for growth acceleration and
business resilience.

■ Developer mind share for APIs is growing. Newer approaches to event-based APIs,
design innovations and modeling approaches such as GraphQL, are driving interest,
experimentation and growth in full life cycle API management.

■ Cloud adoption and cloud-native architectural approaches to computing (including

serverless computing) are increasing the use of APIs in software engineering
architectures, especially in the context of microservices, service mesh and serverless.

■ Regulated, industry-specific initiatives such as open banking and connected

healthcare, along with nonregulated, opportunistic approaches in other industries are
increasing the demand for full life cycle API management.

Gartner, Inc. | G00747497 Page 57 of 78

Obstacles
■ Lack of commitment to adequate organizational governance processes hinders
adoption of full life cycle API management. This can be due to lack of skills or know-
how, or due to too much focus on bureaucratic approaches rather than federated
and automated governance approaches.

■ Lack of strategic focus on business value (quantifiable business growth or

operational efficiencies) and too much focus on technical use cases can disengage
business users and sponsors. This is particularly apparent in cases where API
programs fail to deliver promised return on investment.

■ Traditional, single-gateway approaches to API management do not fit well to a

modern, distributed application environment.

■ Partial or full set of API management capabilities provided by vendors in other

markets such as application development, integration platforms, security solutions,
B2B offerings, etc., can create confusion and potentially shrink the market
opportunities.

User Recommendations
■ Use full life cycle API management to power your API strategy that addresses both
technical and business requirements for APIs. Select offerings that have the ability
to address needs well beyond the first year.

■ Treat APIs as products managed by API product managers in a federated API

platform team.

■ Choose a functionally broad API management solution that supports modern API
trends, including microservices, multigateway and multicloud architectures. Ensure
that the chosen solution covers the entire API life cycle, not just the runtime or
operational aspects.

■ Use full life cycle API management to enable governance of all APIs (not just APIs
you produce), including third-party (private or public) APIs you consume.

■ Question full life cycle API management vendors on their support for automation of
API validation and other capabilities, as well as their support for a modern, low-
footprint API gateway.

Gartner Recommended Reading

Magic Quadrant for Full Life Cycle API Management

Gartner, Inc. | G00747497 Page 58 of 78

The Evolving Role of the API Product Manager in Digital Product Management

How to Use KPIs to Measure the Business Value of APIs

API Security: What You Need to Do to Protect Your APIs

Top 10 Things Software Engineering Leaders Need to Know About APIs

Data Integration Tools

Analysis By: Ehtisham Zaidi, Mark Beyer, Robert Thanaraj

Benefit Rating: High

Market Penetration: More than 50% of target audience

Maturity: Early mainstream

Definition

Data integration tools enable the design and implementation of data access and delivery
capabilities that allow independently designed data structures to be leveraged together.
Data integration tools have matured from supporting traditional bulk/batch integration
scenarios to now supporting a combination of modern delivery styles (such as data
virtualization and stream data integration). Data integration tools are now expected to
support hybrid and multicloud integration scenarios.

Why This Is Important

Data integration tools are needed by organizations to support distributed data

management and deliver data at all latencies across a range of use cases. These include
MDM, analytics, data science, data warehousing, and multicloud and hybrid cloud
integration.

Data integration tool suites are expected to deliver simpler interfaces to support less-
skilled roles like citizen integrators. Growing requirements for automated data integration
also require support for data fabric architectures.

Business Impact

Organizations adopting mature data integration tools increasingly exploit comprehensive

data access and delivery capabilities. They get immediate benefits in the form of:

Gartner, Inc. | G00747497 Page 59 of 78

■ Reduced time to integrated data delivery

■ Cost savings (by reduced integration technical debt)

■ Quality enhancements (for analytics/data science products)

■ Flexibility (to access new data sources)

Integration tools that support data fabric designs will increase the productivity of data
engineering and data science teams.

Drivers
■ Ability to execute data integration in a hyperconnected infrastructure (irrespective of
structure and origins) and the ability to automate transformations through
embedded ML capabilities are the most important drivers for organizations investing
in modern data integration tools.

■ Traditional data integration architectures and tools (which focused solely on

replicating data) are slow in delivering semantically enriched and integrated
datasets that are ready for analytics. This exacerbates data integration tools to
provision a mix of variable latency, granularity, physical and virtualized data delivery.
This is another major reason to invest in these tools.

■ Activities for self-service data access and data preparation by skilled data engineers,
citizen integrators and other non-IT roles spur requirements for new data integration
tools.

■ While traditional data integration tools have now become mature on technical
metadata ingestion and analysis to support data integration activities, there is still
room for maturity for data integration vendors to introduce capabilities to harness
and leverage “active” metadata. Organizations must therefore investigate and adopt
data integration tools that can not only work with all forms of metadata, but also
share it bidirectionally with other data management tools (e.g., data quality tools) to
support data fabric architectures for automation.

■ Dynamic data fabric designs bring together physical infrastructure design, semantic
tiers, prebuilt services, APIs, microservices and integration processes to connect to
reusable integrated data. Vendors will continue to add data integration functionality
or acquire technology in these areas.

Gartner, Inc. | G00747497 Page 60 of 78

Obstacles

■ Tightly integrated data integration tool suites in which all components share
metadata (both active and passive), design environment, administration and data
quality support remain an area for improvement in the data integration tools market.

■ The popularity of data preparation (and other self-service ingestion tools), with the
sole focus on analytics use cases demonstrated, will create some confusion in the
market, slowing the advance of data integration tool suites.

■ The demand for a seamless integration platform that spans and combines multiple
data delivery styles (batch with data virtualization, for example), multiple
deployment options (hybrid and multicloud) and multiple personas currently exceeds
the capabilities of most offerings.

■ Most existing data integration tools are limited in their ability to collect and analyze
all forms of metadata to provide actionable insights to data engineering teams to
support automation.

User Recommendations
■ Assess your data integration capability needs to identify gaps in critical skill sets,
tools, techniques and architecture needed to position data integration as a strategic
discipline at the core of your data management strategy.

■ Review current data integration tools to determine if you are leveraging the
capabilities they offer. These may include the ability to deploy core elements
(including connectivity, transformation and movement) in a range of different data
delivery styles driven by common metadata, modeling, design and administration
environments.

■ Identify and implement a portfolio-based approach to your integration strategy that

extends beyond consolidating data via ETL to include stream data integration, event
recognition and data virtualization.

■ Make automation of data integration, ingestion and orchestration activities your

primary goal for the year, and focus on those data integration tools that can support
data fabric designs.

Sample Vendors

Denodo; Fivetran; HVR; IBM; Informatica; Matillion; Precisely (Syncsort); Qlik; Talend;
TIBCO

Gartner, Inc. | G00747497 Page 61 of 78

Gartner Recommended Reading

Magic Quadrant for Data Integration Tools

Critical Capabilities for Data Integration Tools

Market Share Analysis: Data Integration Tools, Worldwide, 2019

Position Your Product to Benefit From the Rise of Intercloud Data Integration

Data Fabrics Add Augmented Intelligence to Modernize Your Data Integration

LCAP
Analysis By: Paul Vincent, Jason Wong, Yefim Natis

Benefit Rating: High

Market Penetration: More than 50% of target audience

Maturity: Mature mainstream

Definition

A low-code application platform (LCAP) is an application platform that supports low-code

declarative and often visual, programming abstractions, such as model-driven and
metadata-based programming. An LCAP supports end-user interface creation, includes a
database, and is used for rapid application development with simplified software
development life cycle tooling.

Why This Is Important

LCAPs are one of the most popular types of development tools supporting the low-code
paradigm. They support general web and mobile application development with high
productivity while reducing the need for deep developer skills, and are mostly cloud-based.
They are widely adopted for developer personas ranging from enterprise software
developers to citizen developers. Over 200 vendors support a wide variety of business use
cases and industry specializations for digital business automation.

Gartner, Inc. | G00747497 Page 62 of 78

Business Impact

Speeding up application development while reducing developer effort can be

transformative for business IT. Businesses adopt LCAP to deliver more automation and
reduce their application backlogs as well as enable more self-service application
development. Mostly cloud-based, LCAP vendors are also accelerating development of
new capabilities to increase their use-case coverage and justify their subscription costs.

Drivers

■ LCAPs have evolved from rapid application development, business process

management technologies and SaaS extension platforms through their evolution of
common capabilities around user interface, database, business logic definition,
process orchestration and integration of now-ubiquitous API services. The demand
to deliver new business automations through applications continues to outstrip
conventional application development capacity. This is despite the rise of SaaS
usage for standard business services — indeed the latter has resulted in more
demands for custom-made extensions that has resulted in a large part of the LCAP
marketing being SaaS vendors’ LCAP: the market is dominated by Salesforce.

■ Through the requirement for LCAP to enable competitive SaaS and complete
applications, they have evolved toward multifunction capabilities. LCAPs overlap
with the business process automation/iBPMS market for workflow use cases, and
the MXDP market for user-interface-driven use cases.

■ Some vendors have recently focused on cloud-native scalability to support larger

B2C deployments and deeper governance tooling to support remote and distributed
developers to enable postpandemic business and IT development fusion team
structures. Through support for composing applications from multiple API and
service types, LCAPs can cover an increasingly large set of enterprise application
requirements, with some enterprises starting to choose them as a strategic
application platform.

Gartner, Inc. | G00747497 Page 63 of 78

Obstacles

■ Current LCAP market share is heavily biased toward some very large hyperscalers
and a few successful independent vendors. However, Gartner commonly speaks with
clients that have multiple LCAP offerings deployed across the enterprise.

■ LCAPs have been implemented by the main SaaS platform vendors whose market
dominance and deep pockets could diminish the opportunities for a large number of
small LCAP vendors. However, this really means that for most enterprises the
question is not whether to adopt LCAP, but which LCAP(s) will they focus on and
invest in.

■ LCAP like most low-code trades productivity for vendor lock-in (of both applications
and developer skills). Vendor cancellations (like Google App Maker) do occur.

■ Licensing models vary across vendors and can be regularly updated by vendors, and
may not scale for new use cases. This can lead to vendor disillusionment!

User Recommendations

Software engineering leaders, CIOs and CTOs should:

■ Evaluate application lock-ins due to the lack of portability or standards for low-code
models. This technical debt will accumulate fast, and means that vendor
relationships (and contracts) need to be considered strategic. Architecture needs
should be considered — for example whether to use the built-in database for all use
cases.

■ Weigh annual subscriptions against the productivity benefits during application

development (and maintenance). Subscription costs for LCAP are typically per end
user, encouraging maximum LCAP adoption per paid-up user.

■ Ensure developers have access to the tools that make them productive, whether
LCAP or others, and are governed accordingly. Different developers with different
skill sets will vary in their successful adoption of LCAP.

■ Assess LCAP vendors. The large number of vendors implies possible future market
instability, although to date there have been few cases of LCAP retirements.

Sample Vendors

Appian; Betty Blocks; Kintone; Mendix; Microsoft; Oracle; OutSystems; Quickbase;

Salesforce; ServiceNow

Gartner, Inc. | G00747497 Page 64 of 78

Gartner Recommended Reading

Magic Quadrant for Enterprise Low-Code Application Platforms

Critical Capabilities for Enterprise Low-Code Application PlatformsIdentify and Evaluate

Your Next Low-Code Development Technologies

IoT Integration
Analysis By: Benoit Lheureux

Benefit Rating: High

Market Penetration: 20% to 50% of target audience

Maturity: Early mainstream

Definition

IoT integration refers to the integration strategies and technologies needed to assemble
end-to-end IoT-enabled business solutions. IoT-specific integration challenges include
integrating IoT devices, operational technology (OT), digital twins and multiple IoT
platforms. More traditional IoT project integration challenges include integrating IoT
applications and digital twins with enterprise applications, data, business processes,
SaaS applications, B2B ecosystem partners and mobile apps.

Why This Is Important

Every IoT project requires significant integration work — some unique to IoT projects — to
enable IoT devices, IoT applications and various existing business applications to work
well together. In a recent survey, a majority (71%) of companies reported that they made
moderate to major investments in their integration strategy to support IoT projects (see
Survey Analysis: Companies Recognize Integration as a Key Competency for Internet of
Things Projects).

Gartner, Inc. | G00747497 Page 65 of 78

Business Impact
■ IoT integration is an essential functional requirement for all IoT projects.

■ All software engineering leaders (SWELs) and application leaders responsible for IoT
projects must address IoT integration, and to successfully deliver IoT products, they
will either have to train or hire software engineers with unique-to-IoT integration
skills.

■ Special integration skills and tools are often needed for IoT projects (e.g., for OT
integration).

Gartner, Inc. | G00747497 Page 66 of 78

Drivers

■ Extraordinary IoT project technology heterogeneity — e.g., multiple types and OEMs
offering IoT devices, brand-new and decades-old products and equipment, diverse
IoT device data heterogeneity, and diverse applications systems to be integrated.

■ A proliferating desire to ingest and analyze IoT data to support data-driven business
decisions.

■ Proliferation of IoT projects (for which IoT integration is always required).

■ IoT integration is a key challenge for IoT projects. A Gartner survey found that
companies can’t rely on a “one-size-fits-all” approach to IoT device integration, and
had to integrate their IoT projects with many different types of IT endpoints (see
Survey Analysis: Companies Recognize Integration as a Key Competency for Internet
of Things Projects).

■ To fully realize the benefits of IoT, companies will eventually need to integrate new
IoT technologies with legacy (i.e., pre-IoT) business applications and software using
new, enhanced workflow (see How Can Organizations Integrate IoT Digital Twins
and Enterprise Applications?).

■ Complex, distributed IoT projects often involve a mix of IoT devices, IoT platforms,
business applications, mobile apps, cloud services and (often) external business
partners. Such complex IT projects are needed to enable new IoT-enabled outcomes
— e.g., self-diagnosing and self-repairing assets and equipment, “lights-out-
manufacturing,” or product-as-a-service.

■ The need for owner-operators in heavy-asset industries (e.g., manufacturing, utilities,

oil and gas production) to integrate IoT-connected devices and digital twins hosted
on multiple IoT platforms.

■ A growing need to align time-series data generated by various IoT-connected assets

and equipment with traditional EAM master data (e.g., BOM) for the same assets
and equipment.

■ Performance and scalability — that is, potentially large numbers of IoT devices,
products and equipment with high API throughputs and large volumes of time series
data must be integrated.

Gartner, Inc. | G00747497 Page 67 of 78

Obstacles
■ SWELs tend to focus on building software engineering teams for IoT projects with
skills in IoT data, applications and analytics — rather than skills in IoT integration.

■ Few engineers have IoT software development skills, and even fewer have IoT
integration skills.

■ TSPs investing in IoT products (e.g., IoT platforms) tend to focus more on IoT data,
applications and analytics, rather than on integration, which creates integration
functionality gaps.

■ A function gap among many general-purpose integration tools (e.g., ESB, iPaaS) for
many of the IoT-specific integration needs of IoT projects (e.g., IoT devices, OT
equipment or LOB applications such as MES). While many integration tools support
modern IoT device protocols (e.g., APIs, MQTT and OPC-UA), most cannot connect to
older, “brownfield” OT equipment.

■ IoT integration products focused on OT integration (e.g., OSIsoft, Skkynet) may be

needed and must be licensed separately.

■ Perceived high cost of IoT-specific integration tools or services.

User Recommendations

SWELs for IoT projects should:

■ Clearly identify what IoT integration functionality is needed for IoT projects (see
Survey Analysis: Companies Recognize Integration as a Key Competency for Internet
of Things Projects).

■ Avoid simplistic approaches to IoT integration (e.g., “APIs = integration”) that cannot,
alone, address all your needs (e.g., does not also address functionality such as IoT
data translation, OT integration).

■ Hire and/or train software engineers with IoT integration skills.

■ Confirm the availability of required IoT integration capabilities for any IoT product or
service (see Critical Capabilities for Industrial IoT Platforms).

■ Modernize your B2B integration strategy (either via EDI or APIs — see Use APIs to
Modernize EDI for B2B Ecosystem Integration) to enable IoT project integration with
business partners.

Gartner, Inc. | G00747497 Page 68 of 78

■ Align your IoT integration skills with your company's overall integration strategy (see
How to Deliver a Truly Hybrid Integration Platform in Steps).

Sample Vendors

Alleantia; Dell Boomi; Informatica; Microsoft; Reekoh; Salesforce (MuleSoft); Sky Republic;
SnapLogic; Software AG; Solace

Gartner Recommended Reading

 Survey Analysis: Companies Recognize Integration as a Key Competency for Internet of

Things Projects,

 What Should I Do To Ensure Digital Twin Success?

 Critical Capabilities for Industrial IoT Platforms

 Use APIs to Modernize EDI for B2B Ecosystem Integration

How to Deliver a Truly Hybrid Integration Platform in Steps

 Use the IoT Platform Solution Reference Model to Help Design Your End-to-End IoT
Business Solutions

Gartner, Inc. | G00747497 Page 69 of 78

Entering the Plateau
iPaaS
Analysis By: Massimo Pezzini

Benefit Rating: High

Market Penetration: 20% to 50% of target audience

Maturity: Early mainstream

Definition

An integration platform as a service (iPaaS) is a vendor-managed suite of cloud services

that delivers a mix of application, data, B2B and other integration functionality, as well as
API management and event brokering. An iPaaS targets multiple personas: integration
specialists, application developers and business users. Organizations adopt iPaaS as
their first move into integration technology, and to complement or replace classic
integration software so as to enable democratized integration.

Why This Is Important

Organizations’ accelerating shift to the cloud is boosting the iPaaS market (up 38% in
2020) and has made it the biggest segment ($3.7 billion) of the integration platform
technology market. Its functional breadth makes it the natural alternative to classic
integration software (ESB, ETL and B2B gateway software) for large organizations. But,
unlike the classic software, iPaaS also attracts midsize organizations and lines of
business, due to its ease of access, versatility and low initial cost.

Business Impact

By rapidly and cost-effectively addressing integration needs, iPaaS enables organizations

to improve efficiency, provide real-time business insights, increase business agility and
introduce innovation faster. iPaaS adoption helps software engineering leaders achieve
these goals cost-effectively, efficiently and with less costly skills than are needed for
classic integration software. Also, iPaaS makes these benefits accessible to midsize
organizations that cannot afford classic platform costs.

Drivers

Gartner, Inc. | G00747497 Page 70 of 78

■ The vast iPaaS installed base and the COVID-19 pandemic notwithstanding, the
iPaaS market grew quickly in 2020, driven by several business factors. These
included organizations’ pressing need to automate processes, accelerate digital
transformation, react to the dramatic business changes forced by the pandemic, and
speed up plans to move to the cloud in order to contain costs and increase agility.

■ These factors were strongly at play among midsize organizations — traditionally,

heavy users of iPaaS — at least in the less-COVID-19-impacted verticals. Domain-
specific iPaaS targeting particular industries, SaaS ecosystems, business processes
or geographies has been reasonably successful in this sector, because of its appeal
to time-, skill- and resource-constrained organizations.

■ The main goal of iPaaS providers now is to maximize opportunities to upsell and
cross-sell to their vast installed base. Therefore, they are evolving their offerings into
enterprise-class suites that address a wide range of hybrid, multicloud scenarios.
Hence, large and global organizations now position iPaaS as a strategic option to
complement, but increasingly also to replace, classic integration platform software,
which drives more widespread adoption.

■ A growing number of SaaS providers “embed” in their applications their own iPaaS,
or one from a third party, which they typically extend with a rich portfolio of
packaged integration processes (PIPs). This makes embedded iPaaS offerings
attractive to organizations that need to quickly address SaaS application integration.

■ Providers will keep investing to improve developers’ productivity, reduce time to value
and shorten the learning curve. The goal is to further expand their potential audience,
to include business users. Hence, providers’ R&D efforts will focus on using AI,
machine learning and natural language processing to assist development and
operation, enrich PIP portfolios, and enable CI/CD and DevOps to entice professional
developers.

Obstacles

Obstacles to even faster iPaaS adoption include:

■ The market’s extreme fragmentation (over 150 providers and counting). This makes
it hard for user organizations to select the best-fit iPaaS for their needs, could
generate a proliferation of diverse, stand-alone and embedded iPaaS offerings, and
risks fragmenting service providers’ investments in skills building.

Gartner, Inc. | G00747497 Page 71 of 78

■ The top five PaaS providers’ command of about 60% of the market, and the fact that
only seven providers have over 2% share. Hence, the vast majority of providers are of
dubious viability, which may discourage the most risk-averse organizations from
making strategic investments in iPaaS.

■ The API rhetoric of seamless “plug and play” integration, the confusion among less
technically savvy users about the differences between iPaaS, RPA and API
management platforms, and the growing trend for code-based integration
encouraged by OSS integration frameworks. These factors could reduce iPaaS’s
appeal, at least to large organizations.

User Recommendations

Despite the risks relating to market fragmentation, software engineering leaders

responsible for integration should adopt iPaaS when looking for:

■ An integration platform for midsize organizations moving to the cloud and for
“greenfield” integration initiatives.

■ A strategic complement to traditional integration platforms — increasingly in the

context of hybrid integration platform (HIP) strategies — in order to empower a
collaborative, democratized approach to integration.

■ An enabler of self-service integration for ad hoc integrators (such as application

developers and SaaS administrators) or citizen integrators.

■ A platform to support well-defined, tactical integration projects with low budgets,

severe time constraints, and informally defined and incrementally formulated
requirements.

■ A potential replacement for classic integration platforms that are obsolete or cannot
support their changing requirements.

Sample Vendors

Boomi; IBM; Informatica; Jitterbit; Microsoft; MuleSoft; Oracle; SAP; TIBCO Software;
Workato

Gartner Recommended Reading

Magic Quadrant for Enterprise Integration Platform as a Service

Critical Capabilities for Enterprise Integration Platform as a Service

Gartner, Inc. | G00747497 Page 72 of 78

Choose the Best Integration Tool for Your Needs Based on the Three Basic Patterns of
Integration

Gartner, Inc. | G00747497 Page 73 of 78

Appendixes
Figure 2. Hype Cycle for Application and Integration Infrastructure, 2020

Source: Gartner (July 2020)

Gartner, Inc. | G00747497 Page 74 of 78

Hype Cycle Phases, Benefit Ratings and Maturity Levels
Table 2: Hype Cycle Phases
(Enlarged table in Appendix)

Gartner, Inc. | G00747497 Page 75 of 78

Table 3: Benefit Ratings

Benefit Rating Definition

Transformational Enables new ways of doing business across

industries that will result in major shifts in
industry dynamics

High Enables new ways of performing horizontal

or vertical processes that will result in
significantly increased revenue or cost
savings for an enterprise

Moderate Provides incremental improvements to

established processes that will result in
increased revenue or cost savings for an
enterprise

Low Slightly improves processes (for example,

improved user experience) that will be
difficult to translate into increased revenue
or cost savings

Source: Gartner (July 2021) (required)

Gartner, Inc. | G00747497 Page 76 of 78

Table 4: Maturity Levels
(Enlarged table in Appendix)

Document Revision History

Hype Cycle for Application and Integration Infrastructure, 2020 - 30 July 2020

Hype Cycle for Application and Integration Infrastructure, 2019 - 1 August 2019

Hype Cycle for Application and Integration Infrastructure, 2018 - 26 July 2018

Hype Cycle for Application Infrastructure and Integration, 2017 - 7 August 2017

Hype Cycle for Application Infrastructure, 2016 - 7 July 2016

Hype Cycle for Application Infrastructure, 2015 - 30 July 2015

Recommended by the Authors

Some documents may not be available as part of your current Gartner subscription.

Understanding Gartner’s Hype Cycles

Create Your Own Hype Cycle With Gartner’s Hype Cycle Builder

Gartner, Inc. | G00747497 Page 77 of 78

© 2021 Gartner, Inc. and/or its affiliates. All rights reserved. Gartner is a registered trademark of
Gartner, Inc. and its affiliates. This publication may not be reproduced or distributed in any form
without Gartner's prior written permission. It consists of the opinions of Gartner's research
organization, which should not be construed as statements of fact. While the information contained in
this publication has been obtained from sources believed to be reliable, Gartner disclaims all warranties
as to the accuracy, completeness or adequacy of such information. Although Gartner research may
address legal and financial issues, Gartner does not provide legal or investment advice and its research
should not be construed or used as such. Your access and use of this publication are governed by
Gartner’s Usage Policy. Gartner prides itself on its reputation for independence and objectivity. Its
research is produced independently by its research organization without input or influence from any
third party. For further information, see "Guiding Principles on Independence and Objectivity."

Gartner, Inc. | G00747497 Page 78 of 78

Table 1: Priority Matrix for Application Architecture and Integration, 2021

Benefit Years to Mainstream Adoption

Less Than 2 Years 2 - 5 Years 5 - 10 Years More Than 10 Years

Transformational Digital Integrator Data Fabric

Technologies Self-Integrating Applications
Event Stream Processing

High Data Integration Tools Data Hub iPaaS Digital Integration Hub
Full Life Cycle API Event Broker PaaS Event-Driven Architecture
Management Integration Strategy HIP
IoT Integration Empowerment Team
LCAP iPaaS
MASA
Microservices
Packaged Integration
Processes

Moderate Citizen Integrator Tools

Cloud Native Architecture
Service Mesh

Low Serverless fPaaS

Source: Gartner (July 2021)

Gartner, Inc. | G00747497 Page 1A of 4A

Table 2: Hype Cycle Phases

Phase Definition

Innovation Trigger A breakthrough, public demonstration, product launch or other event

generates significant media and industry interest.

Peak of Inflated Expectations During this phase of overenthusiasm and unrealistic projections, a flurry of
well-publicized activity by technology leaders results in some successes, but
more failures, as the innovation is pushed to its limits. The only enterprises
making money are conference organizers and content publishers.

Trough of Disillusionment Because the innovation does not live up to its overinflated expectations, it
rapidly becomes unfashionable. Media interest wanes, except for a few
cautionary tales.

Slope of Enlightenment Focused experimentation and solid hard work by an increasingly diverse
range of organizations lead to a true understanding of the innovation’s
applicability, risks and benefits. Commercial off-the-shelf methodologies and
tools ease the development process.

Plateau of Productivity The real-world benefits of the innovation are demonstrated and accepted.
Tools and methodologies are increasingly stable as they enter their second
and third generations. Growing numbers of organizations feel comfortable
with the reduced level of risk; the rapid growth phase of adoption begins.
Approximately 20% of the technology’s target audience has adopted or is
adopting the technology as it enters this phase.

Years to Mainstream Adoption The time required for the innovation to reach the Plateau of Productivity.

Gartner, Inc. | G00747497 Page 2A of 4A

Phase Definition

Source: Gartner (July 2021) (required)

Table 3: Benefit Ratings

Benefit Rating Definition

Transformational Enables new ways of doing business across industries that will result in
major shifts in industry dynamics

High Enables new ways of performing horizontal or vertical processes that will
result in significantly increased revenue or cost savings for an enterprise

Moderate Provides incremental improvements to established processes that will result

in increased revenue or cost savings for an enterprise

Low Slightly improves processes (for example, improved user experience) that will
be difficult to translate into increased revenue or cost savings

Source: Gartner (July 2021) (required)

Gartner, Inc. | G00747497 Page 3A of 4A

Table 4: Maturity Levels

Maturity Levels Status Products/Vendors

Embryonic In labs None

Emerging Commercialization by vendors First generation

Pilots and deployments by industry leaders High price
Much customization

Adolescent Maturing technology capabilities and process Second generation

understanding Less customization
Uptake beyond early adopters

Early mainstream Proven technology Third generation

Vendors, technology and adoption rapidly evolving More out-of-box methodologies

Mature mainstream Robust technology Several dominant vendors

Not much evolution in vendors or technology

Legacy Not appropriate for new developments Maintenance revenue focus

Cost of migration constrains replacement

Obsolete Rarely used Used/resale market only

Source: Gartner (July 2021) (required)

Gartner, Inc. | G00747497 Page 4A of 4A