Your Organization's

Guide to Data Maturity

Skill up at scale.
Data training designed for your business.
Join 1,600+ companies and 80% of the Fortune 1000
who use DataCamp to upskill their teams.

Visit DataCamp
Table of
Contents The state of data-driven digital transformation 1

The five data maturity levers 2

Data maturity is a spectrum 3

What data reactive looks like 4

The importance of scaling people 5

The path to data scaling 6

What data scaling looks like 7

Strengthening foundations will be key 8

The path to data progressive 8

What data progressive looks like 10

The path to data fluency requires a multipronged approach 11

A blueprint for data fluency 14

Data fluency is driven by people 15

How DataCamp supports organizations on 16

their path to data fluency
 In 2012, Harvard Business Review dubbed data The path to becoming a data-driven organization is a
science the “sexiest job of the 21st century”. In it, D.J. long and arduous process (Harvard Business Review),
Patil and Thomas H. Davenport outlined the nascent as it requires investments in scaling various levers such
set of skills and attributes that make up a data as infrastructure, people, tools, and more.

1 scientist, and how leveraging data and analytics will

be the main competitive advantage that sets
Organizations also sit on a spectrum of data maturity
and depending on where they sit on this spectrum,
organizations apart from their competitors. need to allocate their focus to some levers over others.

The state of data- Now, almost a decade later, achieving digital The north star for any aspiring data-driven
driven digital transformation, becoming data-driven, and scaling
data science capabilities is on every organization’s
organization is to become data fluent, where everyone
has the access, and skills they need to work with data
transformation mind. According to a 2021 NewVantage Partners and make data-informed decisions. This ultimately
survey, 99% of firms are reporting active investment in means the democratization of data and skills across a
data science and machine learning. However, fewer range of roles and the maturation of a data culture.
than 30% of organizations are experiencing
transformational business outcomes as a result of In this guide, we outline a framework for evaluating,
these investments, and only 24% of them claim they and scaling data maturity throughout the
have created a data-driven organization. Similarly, organization, define the various data maturity stages
research from BCG shows that more than 80% of an organization goes through, and draw a path of
companies are accelerating their digital initiatives organizations can take to achieve data
transformation plans, but only 30% of them are fluency.
achieving their stated objectives.

Your Organization's Guide to Data Maturity 1

 When scaling data science throughout an organization, there are five key levers one should
consider to go from one data maturity stage to another. These levers are the key elements of
our IPTOP framework for scaling data science, and they can be defined as such:

Infrastructure — entails creating a data infrastructure that ensures data is collected, discoverable,
reliable, understood, compliant, and actionable throughout the organization.
2
People — entails both forging a data culture where everyone understands the value of data and the
The five data importance of continuous learning, as well ensuring that everyone has the skills to work effectively with data.

maturity levers Tools — which refers to the tools data practitioners across all skill levels use, as well as the development of
frameworks that enable easier work with data throughout the organization.

Organization — which refers to how data talent is organized, and adopting organizational models that
promote scalable data science throughout the organization.

Processes — the workflows that data experts and teams adopt to make their work more predictable and
collaborative, and to ensure alignment with business objectives.

Infrastructure and people are the base foundation supporting the other levers.
Without data culture, skills, and access to data, organizations won’t be able to leverage
data tools, scale data processes, and organize data talent.

Your Organization's Guide to Data Maturity 2

 3
Data maturity is a spectrum

The path to data fluency is comprised of four data maturity It’s important to note that this does not mean that everyone needs
stages. It’s important to think of data maturity as a to code, as there are various data personas found in every data-
spectrum, and that some organizations may find themselves driven organization who possess a completely different relationship
relatively more mature in one element of the IPTOP with data to do their best work. For example, data leaders and
framework and less mature in another. With that in mind, we consumers are often non-technical roles that need to be data
can break down the stages of data maturity through the literate in order to be conversational with data experts and identify
number of people in an organization that rely on data to do where data can be used to answer a business question. Data
their best work, and who have the skills and access to make engineers, on the other hand, are highly technical individuals who
data-driven decisions. are responsible for scaling and maintaining data infrastructure
throughout the organization.

Data Fluent
For a deep-dive into all eight Everyone knows how to
data personas found in every access the data they
data-driven organization, Data Progressive need to do their job and
download our white paper on the make data-driven
Every team has at least
L&D Guide to Data Fluency. decisions.
one data practitioner who
Data Scaling can analyze, report, and
Very few people in an present their data
organization have the regardless of their role.
Data Reactive skills and access they
No one accesses or uses need to analyze, report,
data in their daily work. and confidently present
An organization rarely data insights.
reports or presents data.
 4
In data reactive organizations, no one accesses or uses data in their daily work,

What data reactive looks like and a company rarely reports on or presents data. There is no data culture or
strategy to support maturing data competencies.

Infrastructure Tools Processes

Data is collected on an ad hoc basis with disparate tools, There are a variety of ad hoc legacy tools for working Any form of data work is done on an ad hoc basis and
and there is no trusted, centralized data storage with data that are rarely leveraged. there are no established processes for working with data.
throughout the organization.

Data
People Organization Fluent
Data
Progressive
No one has the skills to work with data on a daily basis There is no data team and no data experts throughout Data
Scaling
— and the organization does not have nor does it the organization. A lack of data strategy hinders the Data
Reactive
prioritize developing a data culture. allocation of resources for data talent.

At the data reactive stage, organizations suffer from low On the people side, data reactive organizations are
maturity in infrastructure and people, ultimately limiting characterized by cultures that value experience and gut
their maturity ceiling in terms of tools, organization, and instinct over data-driven decision making, there is no
processes. On the infrastructure side, a data reactive dedicated data talent working on scaling data maturity,
organization collects data on an ad hoc basis and stores it the majority of the organization does not value the
in siloes. The data often sits in departmentally owned flat opportunities of employing data analytics in their daily
files, spreadsheets, or legacy tools. There are no data work, and learning and development for data skills is not a
quality standards to support data linkage between priority on the organizational roadmap.
departments, and there is minimal data strategy that
incorporates scaling data infrastructure.

Your Organization's Guide to Data Maturity 4

 “Building a data culture is not just an
option, it is business-critical. Literally 100%
of data and analytics investments depend
on having a Data Culture.”
Sudaman Thoppan Mohanchandralal, Chief Data Officer
at Allianz Benelux

The importance of scaling people

Arguably the most important lever to scale when scaling data maturity in the early
stage of data maturity is people, specifically data culture and skills. According to the
same NewVantage Partners survey covered earlier, 92% of organizations believe that
the biggest impediment towards becoming data-driven is people and culture. This is
because regardless of technological investments in data infrastructure and tools,
people drive the success of data and analytics initiatives (Deloitte).

Moreover, scaling data skills and culture means that there is organizational alignment
and excitement around the importance of data initiatives (McKinsey). This excitement
will be the lifeblood that sustains future initiatives when scaling data maturity and will
be the catalyst of growth for organizations on their path to data fluency.

Your Organization's Guide to Data Maturity 5

4. WHAT DATA REACTIVE LOOKS LIKE

The path to data scaling • Build strong executive support — According to

McKinsey, one of the key differentiators between
• Hire the appropriate data talent — As excitement
begins to build around data initiatives, organizations
The path to scaling starts by focusing exclusively on the organizations who succeed at scaling data initiatives should start looking to hire data experts to start
two base levers of the IPTOP framework: infrastructure versus those who lag behind is strong executive support working on key pilot data projects that could have
and people. The initiatives laid out below are centered and ensuring senior-management participation in transformational business outcomes. organizations
around forging data culture and talent, the development analytics activities. More importantly, organizations must hire data talent that focuses on the foundations
of a data strategy that takes into account scalable data with high-performing data initiatives report higher that enable scalable data science in the long run. A key
infrastructure alongside learning and development, and direct sponsorship of data initiatives by the CEO. anchor for organizations to leverage is the Data
ultimately setting the stage for greater growth in the long Developing a top-down approach for driving Science Hierarchy of Needs developed by Monica
term. excitement around data initiatives is essential to Rogati, which articulates that the path towards using
sustaining data culture change. Practically speaking, machine learning and advanced data science at scale
People this could begin with public championing of successful starts with fostering a solid foundation in terms of data
proof of concepts, developing and articulating an collection, storage, and transformation. While no
• Prove the value with a proof of concept — In order to
organization-wide data strategy, ensuring senior organization is the same, hiring data engineering talent
create excitement around data initiatives and to drive
management are putting data on their departmental should be top of mind for organizations looking to
executive sponsorship of organization-wide data
agenda, and hiring data talent to champion and enact launch their path into data fluency (Quantum Black).
culture change, organizations should look to develop
change (Harvard Business Review).
a proof of concept that showcases the potential return
on investment of data projects. The goal here is to
Infrastructure
• Develop a data strategy that puts learning at the
showcase a quick win that is simple and robust in center — Once executive support is acquired and an • Develop a data strategy with scalable infrastructure
order to show the potential of wider investments in organization is aligned on the importance of scaling in mind — In order to enable data access at scale for
data. These analytics low-hanging fruit can range data science capabilities, leaders should start the long run, organizations need to also center their
from developing a simple customer churn model that developing a data strategy that puts learning and data strategy around scalable data infrastructure.
showcases the potential of optimized marketing development at the center. A holistic approach that Depending on the size and complexity of the
spend, developing a rudimentary dashboard that integrates business strategy, data infrastructure, organization, such a strategy could mean defining how
provides descriptive analytics for finance governance, and implementation outcomes is essential and which data will be collected, outlining initiatives for
departments, or even leveraging A/B testing on email to drive success (Forbes). However, a key component modernizing IT infrastructure (McKinsey), scoping data
subject lines when managing email campaigns to of scalable data strategies is learning and development architecture and its requirements (Deloitte), thinking
optimize open rates. as it will empower executives to navigate complex data about data quality, identifying vendors for cloud
landscapes with improved data literacy, lay the migration, and more. In the next data maturity stage,
groundwork for sustainable organization-wide data we will outline concrete steps organizations can take to
maturity in the long run with data skill assessments, scale their data infrastructure.
evolve the skill set of subject matter experts to meet
future data challenges, and be a key component of any
Your Organization's Guide to Data Maturity 6
data culture program (Deloitte).
 5 In data scaling organizations, very few people in an organization have the
skills and access they need to analyze, report, and confidently present data
What data scaling looks like insights. There is minimal data culture with very few champions and very few
skilled practitioners, and there is no organization-wide access or trust in data.

Infrastructure Tools Processes

Only a few key experts understand how data is accessed There are mostly legacy tools for working with data, with Very few, ad-hoc and limited data processes exist in
in the organization. There is no organization-wide access ad hoc use of modern tooling for data work. siloed teams.
to or trust in data.
Data
Fluent
People Organization Data
Progressive
Data
There is minimal data culture with very few champions While a data strategy is in place, the organizational Scaling
Data
believing in the importance of leveraging data, and few structure of data talent is still undefined and change Reactive

data experts that are working on data initiatives. management is still taking place.

At the data scaling stage, organizations have made the in the way of joining datasets across various departments. As for the remaining components of data maturity (tools,
first steps by creating a data strategy that aligns data More importantly, some departments have gotten a head organization, and processes), organizations at this stage
science with business objectives, is centered around data start over others in terms of accessing and utilizing data of maturity do not have a modern data tooling stack.
culture and skills, and incorporates scalable data to push key strategic pilot projects. Instead, most data work is done in legacy tools such as
infrastructure. Moreover, they have some data talent in spreadsheets, and only a few key data experts leverage
place to drive some strategic projects. Having made these On the people side, data scaling organizations can be advanced data tooling. More importantly, organizations
first steps, data scaling organizations will need to start characterized as having minimal data culture with very have still not developed an organizational model for data
strengthening maturity of tools, organization, and few people who believe in the importance of data, talent, and senior leadership is not yet spearheading or
processes. evidenced by not having the skills or mindset to make sponsoring change management. Finally, only a few ad
data-driven decisions. This can be manifested as a data hoc data processes exist on siloed teams, with no
Infrastructure for data scaling organizations is not yet culture that is still undefined, and whose exact organization-wide norms set in place to work with data or
mature, as only a few key experts understand how data is components are not yet articulated, and where self- collaborate with data experts.
accessed, and there is still no widespread trust in data. motivated learning is king. Moreover, as data initiatives
This means that there is no centralized data storage that start gaining traction alongside data experts, change
allows easy access for data experts and enthusiasts alike, agents and data champions will start to crop up in
datasets are still siloed, and data experts face data quality forward-thinking departments (DataIQ). Your Organization's Guide to Data Maturity 7
issues that lengthen the data cleaning process and stand
4. WHAT DATA SCALING LOOKS LIKE

Strengthening foundations will be key The path to data progressive

In this stage of data maturity, organizations will still need Infrastructure People
to focus on the foundations. As discussed earlier, people
• Start creating a single source of truth by centralizing • Clearly define what a data culture is and what it isn’t
are arguably one of the most important levers when
data storage — A key component in scaling access to — As leadership starts to communicate the
scaling data maturity, in terms of their skills and culture.
data that is discoverable, reliable, understood, importance of data culture, they must also define
Scaling people will not require just organizational
compliant, and actionable throughout the what a data culture entails. In a data-driven culture,
alignment around the importance of data science or data
organization, is centralizing data storage and creating organizations set clear expectations on how to use
initiatives, but a behavior and mindset change that trickles
a single source of truth for the organization data responsibly, executives exemplify data-driven
down from the top to the bottom of an organization
(McKinsey). While just centralizing data storage won’t behavior by arguing and persuading with data, and a
(Harvard Business Review).
solve an organization’s data quality woes, it’s an data-driven mindset becomes a habit (Tableau). As
excellent first step in avoiding siloed datasets that do such, organizations should start defining and tracking
“Data Science is a way of doing things, and not not adhere to organization-wide data quality important metrics around projects, encourage agility

a thing to do [...] Data Competency is a way of standards. Centralizing data storage is most and the use of data, tie data-driven decision-making
commonly done by leveraging centralized data to business outcomes, and drive continuous
thinking, and not a thing to think about”
storage solutions like data lakes and warehouses from education.
Kirk Borne, Chief Data Scientist at Booz Allen Hamilton cloud providers such as Amazon Web Services,
Microsoft Azure, Google Cloud, Snowflake, and more. • Create an upskilling strategy that’s made for the
long-term — A key component of creating and
As such, organizations need to articulate and define a • Start scaling data quality — One of the main reinforcing a data culture is making sure everyone has
data culture by defining important metrics for business obstacles towards organization-wide trust in data is the skills they need to work with data effectively to do
initiatives, defining and incentivizing new behaviors (e.g., data quality (Monte Carlo). Moreover, data quality their role. At this stage, organizations should set the
reporting on project updates with data points), developing issues hamper productivity as data experts spend stage for an organization-wide data upskilling
a wider learning and development strategy to shore up excessive time on data quality issues rather than strategy that is inclusive and takes into account all
data skills, and rewarding data champions and change advancing an organization’s data maturity types of roles. This begins with identifying relevant
agents. Strengthening infrastructure is the second (McKinsey). As such, organizations looking to mature data personas in the organization, assessing current
foundation for data to permeate throughout the their infrastructure should make data quality and skill sets and identifying skill gaps, and most
organization, which will allow organizations to scale data governance a key aspect of scaling infrastructure on importantly, aligning the upskilling objectives with
access. This includes centralizing data using a centralized the path to data progressive. This includes developing business strategy (DataCamp).
data storage solution, defining data access structure for data quality standards for various departments,
high-impact teams, and establishing data governance and creating documentation around accessing data,
quality mechanisms and policies. investing in data quality monitoring mechanisms, and Learn how AXA XL is using DataCamp
iteratively increasing the scope of data governance to upskill its people
initiatives. Learn more
5. WHAT DATA SCALING LOOKS LIKE
Centralized Model
• Reward the change agents — A data scaling between the centralized data team model and the H OW
MUC ..?
H…? WE.
ULD
organization is typically characterized by the decentralized model vary. The centralized model FINANCE SHO PRODUCT

presence of data champions and change agents who allows the data team to function as a center of
are playing an active role in evolving the status quo. excellence, with the ability to push the data agenda IS...
?
D ATA S C I E N C E
H OW
CAN
W H AT …?

Organizations should reward these change agents by throughout the organization. Moreover, it allows
MARKETING ENGINEERING
increasing the visibility of their work, promoting and collaboration and knowledge sharing between data
empowering them to be more cross-functional, and experts who are working on completely different
Benefits
evangelizing the successes they’ve led across the problems. However, a drawback is that it could limit
Data science functions as a center of excellence
organization. This further catalyzes culture change collaboration and alignment with subject matter
Promotes collaboration and knowledge sharing
and increases the adoption of data-driven buy-in experts, and risks isolating the data team as a support
Data Science Manager has domain knowledge
throughout the organization (Tableau). function. The decentralized model, on the other hand, Incentivizes consistent technology stack and better tooling
allows data experts to be directly embedded in
Drawbacks
Tools business units. This allows better cross-functional
Limits coordination between data science and other stakeholders
alignment between business stakeholders and data
• Modernize your data stack — As data becomes Risk of misalignment between data science and Business Units
scientists. However, this also means that it’s more
centralized and providing scalable access becomes Risks isolating data science as a Support Function
difficult to move data experts between teams, and
top of mind, giving access to data tooling that is
data experts are managed by business leaders.
inclusive to relevant data personas for your
organization will be paramount. This could include
Processes Decentralized Model
open-source programming languages like Python and
R, business intelligence tools such as Tableau or • Create processes for collaboration between data
Microsoft Power BI, easy access to a SQL database talent and business units — Depending on the model FINANCE PRODUCT

interface, and more. chosen for organizing data talent, organizations

should start thinking about creating scalable
Organization processes to enable better collaboration between MARKETING ENGINEERING

data scientists and subject matter experts and to

• Define how data talent is organized — With
increase the productivity of data experts. This could Benefits
increased hiring of data talent (i.e., data scientists,
start with a ticketing system, creating a Quality Each team has a dedicated data scientist
data engineers, data analysts, etc.), a key component
Assurance process before publishing data products, Cross-functional alignment
of scaling data maturity is defining an organizational
adopting agile project management methodologies, Data science has a more natural “seat at the table”
model for data experts. Depending on the size and the Fewer dependencies across teams
or creating template project structures.
objectives of an organization, data talent can be
Drawbacks
organized under a central data team or can be
Harder to move data science resources between teams
distributed across business units. The pros and cons
Manager of the team may not have domain knowledge
Harder for data scientists to collaborate and drive long-term projects
6 In data progressive organizations, every team has at least one data fluent
employee who can analyze, report, and reason with data regardless of role.
What data progressive looks like Data is accessible but not easily discoverable. Data is seen as a strategic
asset, but there is no organization-wide data literacy.

Infrastructure Tools Processes

Data is accessible, and data infrastructure is maturing. Modern tooling is accessible, but there are no Mature processes for high data competency teams, but
However, data is not readily discoverable, compliant, frameworks that democratize working with data. are yet to be democratized for the entire organization.
understood, or actionable throughout the organization.

Data
People Organization Fluent
Data
Progressive
Data is seen as a strategic asset and there is alignment The data team is set in place, but their impact is siloed Data
Scaling
around its importance. However, organization-wide data and data talent is seen as a support function. Data
Reactive
literacy is lacking and data training is limited to pockets
of the organization.

At the data progressive stage, organizations have made strides to changes have been done to it, nor to find a clear explanation of As for tools, organization, and processes, data progressive
elevate data maturity. This translates into an environment where dataset metadata and other key components of data discovery. organizations have a modern tooling stack but have yet to create
there is organization-wide buy-in over the importance and Moreover, scalable organization-wide data quality standards are scalable frameworks that further democratize working with data.
strategic nature of leveraging data, and where data is largely still lacking, which hurts overall trust in data. Finally, This means that working with relatively advanced tools has a high
accessible. There is an established organizational model for data operationalizing machine learning and data science is still barrier to entry for non-data experts. The data team has an
talent, and modern tooling is accessible across the board for nascent, and many tools aimed at productionizing, monitoring, organizational model (centralized or decentralized), but data
various data personas. However, the final set of challenges here and refining models in production have yet to be adopted. talent is limited to being a support function for other business
are all about democratization—empowering everyone to make units. This limits the data team from advancing the data agenda
data-driven decisions, freeing up data talent from becoming a On the people side, strides in data culture have led to by pursuing high-impact projects that scale throughout the entire
support function, and focusing instead on highly strategic projects organization-wide buy-in on the strategic importance of data. organization. Data progressive organizations also have high
that capture value and raise the organization's data maturity However, data skills training is still departmental. Insufficient maturity processes for streamlining data talent workflows, but
(O’Reilly). organization-wide data literacy slows down innovation and limits ways of working around operationalizing machine learning and
productivity (Accenture). Most importantly, it hinders the data science are still nascent, and processes for scalable data
On the infrastructure side, data is accessible and data adoption of data science as a methodology for solving problems, work are not inclusive to the remainder of the organization.
infrastructure is maturing. However, there is no easy, scalable rather than a skill to acquire. This ultimately results in an
way to easily discover who owns a particular dataset and what organization where there is no common data language, an over-
reliance on data talent to provide data insights even if it’s basic
reports, and where people do not feel empowered to make data Your Organization's Guide to Data Maturity 10
informed decisions.
6. WHAT DATA PROGRESSIVE LOOKS LIKE

The path to data fluency requires a powerful search functionality, a deep overview of • Strengthen data quality and governance — In order
dataset ownership, lineage, metadata, and to create trust and alignment around data,
multipronged approach
programmatic functionality to scale data quality organizations looking to democratize data should look
The arrival at data fluency is arguably the most difficult checks and descriptions (Edmunds). at data governance as a key pillar for facilitating that.
step to make for organizations. At this point, organizations As such, organizations should invest in scaling their
are likely to have scaled their infrastructure to process data quality strategy and processes. This means
terabytes of data daily, hired, organized, and derived investing in data governance platforms, treating data
value from data talent, and enabled access to modern as a product to which it abides by software
tooling. engineering best practices like data service level
agreements (Monte Carlo) and creating data quality
The difficulty is scaling data science to become an certifications for datasets (Airbnb).
organization-wide methodology for solving business
challenges, and where data and data skills are • Invest in operationalization capabilities and tooling

democratized. To reach this point, organizations should — As machine learning models become more

adopt a multipronged approach that tackles all the commonplace—whether in business processes or in

elements of the IPTOP framework. customer-facing products—operationalizing these

models will require continuous monitoring, integration,
Infrastructure versioning, and explainability. MLOps is still emerging
as a field and goes beyond just tools and frameworks
• Start investing in data discoverability — As data to adopt. The tooling space for operationalizing
infrastructure scales and the architecture of modern models in production is gaining steam and will allow
data infrastructure becomes standardized, a key organizations to scale deployment of data products
component of democratization and scaling data and machine learning models in production (INNOQ).
science will be ensuring that data is not only collected
and accessible, but trusted, reliable, actionable, and
easily discoverable. This is where data discovery tools
come in. Data discovery tools were born out of the
growing pains of data fluent organizations. For
example, Lyft developed its open source discovery
tool Amundsen, based on the challenges associated
with scaling the amount of data being collected.
Airbnb developed Dataportal in order to make sense,
and tackle the challenges with an increasingly
complex data landscape. Data discovery tools provide
Lyft’s Amundsen, and Uber’s Databook in action
Your Organization's Guide to Data Maturity 11
6. WHAT DATA PROGRESSIVE LOOKS LIKE

People
• Assess, measure, and track data skill competencies Tools
over time — A key component of scaling data literacy
• Develop frameworks to democratize working with
• Roll out organization-wide data training fit for all is tracking skill development and using assessments to
data — Providing access to tooling is only one part of
data personas — We’ve demonstrated that people are steer learning paths (Deloitte). Assessments provide
the equation in democratizing data. The next frontier
the cornerstone of data maturity across stages. individuals with insights into their data skills maturity,
is lowering the barrier to working with data with
Organizations looking to drive data fluency at this and will guide personalized learning paths for them
frameworks. For example, the data team at Airbnb
stage face the challenge of making data a default based on their strengths, weaknesses, and necessary
developed an R package conveniently named Rbnb,
consideration in most decisions and actions taken. data skills for their particular role. At an aggregate
which provides a streamlined way to move data
One of the main blockers organizations face is level, assessments will provide organizations an
between different places in their data infrastructure,
insufficient data literacy (Deloitte), which limits the overview of their data maturity skill matrix, which acts
branded visualization themes, R Markdown templates
ability of non-experts to become independent when as a balance sheet of the company’s data skills.
for different types of reports, and custom functions to
seeking and creating data insights. When rolling out
optimize different parts of data workflows. At
organization-wide training, the best practice is to
create personalized learning paths for different data DataCamp Signal™ helps DataCamp, we have R and Python packages that
organizations to pinpoint their abstract querying data from our data warehouse with
roles. For example, Airbnb created Data University, a
a few lines of code. Frameworks enable everyone to
program aimed at upskilling the entire organization strengths and skill gaps at scale.
adopt more advanced data tooling, and standardize
on data skills, with content curated for specific data
the data workflow for data experts throughout the
teams in their Data U Intensive program. Cohorts were
• Reward a culture of continuous learning — Data organization.
able to achieve a 30% sustained increase in SQL
usage. Bloomberg uses DataCamp as part of a science is not merely a skill to learn, but a

blended learning program, which has sustained a methodology for solving business problems. As a field,

561% increase in activities on their proprietary Jupyter data science is continuously evolving, with new

Notebook environment (DataCamp). In a data fluent techniques, frameworks, tools, and best practices

organization, learning paths run the gamut from being developed (Datafloq). Organizations should

general data literacy, to data science, machine reward a culture of continuous learning if they want to

learning, and data engineering, and more. This build and sustain data skills across data roles

ultimately enables the adoption of a common data (McKinsey). This not only includes creating a safe

language and reinforces data science as a habit and space for active learning and development, but

methodology for approaching business problems. incentivizing novel, data-driven approaches for solving
problems and promoting learning by doing.

Your Organization's Guide to Data Maturity 12

6. WHAT DATA PROGRESSIVE LOOKS LIKE

Organization
The Hybrid Model
• Consider moving to a hybrid organizational model — As data competencies mature and more
data experts are developed within the organization, the next step is to organize data talent under
a hybrid model. The hybrid model combines the centralized and decentralized data science team. FINANCE PRODUCT
In a hybrid model, a central data team functions as a center of excellence that drives data
democratization efforts like creating frameworks and standardizing the data tooling stack
DATA S C I E N C E
throughout the organization, all while embedding data scientists into business functions to create
better alignment between data science and business units. However, it’s important to note there
is a spectrum of hybrid models sitting between the two ends, and that organizations should find a
MARKETING ENGINEERING
sweet spot that suits their size, objectives, and data strategy (Altexsoft).

Benefits
Processes
Data science can function as a center of excellence.

• Create and reward processes that put collaboration and democratization at the center — A key Data science can drive common tech stack, tooling, frameworks, and standardization
Data science can collaborate and align on organizational goals.
mark of data fluent organizations is sharing tools, knowledge, and insights. These organizations
Better alignment between Data science and business units
have created processes that simplify access to data insights, making it easy to collaborate with
data experts and work with data. For example, LinkedIn developed DataHub, a data discovery Drawbacks
tool that also allows the organization to share data insights, charts, and dashboards. Airbnb Risk of mismatch of expectation leadership of Data science and business unit.

developed Knowledge Repo, a platform that makes it easy to publish reports and to surface data Everyone has at least two teams.

insights. Netflix created processes and templates for accessing and working with Notebooks
throughout depending on the type of problem being tackled. These processes enable easier
access to data knowledge, and are centered around simplifying work with data.

Your Organization's
Our Guide
Guide to
to Data
Data Maturity 13
Maturity 2
A blueprint for data fluency

Variety of ad-how legacy tools to work Any data work is done on ad-hoc basis
with data that are rarely leveraged and there are no processes for working
with data

Your Organization's Guide to Data Maturity 14

Data fluency is driven by people
Ultimately, a data fluent organization is where everyone knows how to access the data
they need to do their job. This could mean marketing managers accessing business
intelligence dashboards to make decisions around campaign spend, executives
surfacing data insights on a knowledge repository to make data-driven decisions,
business analysts leveraging SQL to streamline complex excel workflows, and data
experts leveraging the latest and most advanced data tools and techniques to further
democratize data and push innovation throughout the organization.

What is common between these individuals is a mindset and buy-in around the
importance of leveraging data for making decisions, and for solving problems. This is
why building data skills and culture is paramount for any data maturity initiative to
scale regardless of where the organization sits on the data maturity spectrum.

Your Organization's
Our Guide
Guide to
to Data
Data Maturity 15
Maturity 2
 Learn Track skills with skill matrix
How DataCamp supports DataCamp’s growing course library houses more than 350 Track the data skills your team has today and map a

organizations on their expert-led, hands-on courses across various technologies

and domains for all data skills and levels. Learners can hit
path to the skills they need tomorrow. Using the Skill
Matrix, admin users can easily filter to identify individuals
path to data fluency the ground running with our learn-by-doing approach— with the skills you need to take on specific projects or
our bite-sized videos and interactive coding exercises teams with low use or data skills gaps. They can then
allow them to start working with their preferred tool and create and assign custom tracks to help bridge these
topic right in the browser. gaps and report on skill development.

DataCamp’s proven learning methodology

provides a cyclical process for learning and Practice
retention. This learning methodology enables The next step in DataCamp’s proven learning
learners across the data fluency spectrum to methodology is to practice all the information retained in
assess their skills and identify gaps, develop a courses. Using practice mode, learners can practice what
learning plan based on these gaps, practice they’ve learned with short challenges to test critical
skills, and apply them in a real-world setting. concepts. With over 3,400 practice questions, learners can
Experienced data scientists can upskill on new practice their skills across various technologies and topics.

techniques in their target domain, and domain Our mobile app is the perfect way to practice and learn
on the go.
experts can learn the fundamentals of data
literacy and data science.
Apply
Assess Once skills have been assessed, cultivated through A robust training experience
Effective learning starts with understanding skill gaps and courses, and sharpened through practice, learners are We work with the largest brands in the world, including
strengths. With DataCamp Signal™, learners can ready to apply their skills in a project-based environment. PayPal, Uber, HSBC, and Google, to help them transform
understand specific skill gaps they have across various With DataCamp projects, learners can solve a variety of their data skills. Our experienced team provides you with
topics and tools. From data literacy assessments like real-world R and Python data science projects. Learners resources and guidance on everything from adoption
understanding and interpreting data to programming and can opt for guided projects, where they can follow step- best practices to SSO and LMS integrations—giving you
machine learning assessments in R, Python, and SQL, our by-step tasks and receive helpful feedback as they apply the tools you need to upskill your team with confidence.
10-minute adaptive evaluations provide learners with their newfound skills. They can also opt for unguided Join over 7 million learners around the world. Close your
personalized skill gaps and learning paths to address their projects, which are open-ended, offering a variety of talent gap. Visit datacamp.com.
skill gaps. possible solutions and a live-code-along video to follow
how an expert data scientist would approach a solution.

Your Organization's Guide to Data Maturity 16

More from DataCamp
Read our white papers and watch our own demand webinars
Learn how you can bridge your
team’s data fluency gap and
become more data-driven.
Visit DataCamp