Assignment 1 - Big Data in Big CompaniesA Newer Equal Part Linear Regression Model: A

Eduardo Octavio
Resendez Hernandez
Matricula: A00914060

Professor: Dr. Oscar Barzuna Hidalgo

 In the realm of business, the concept of big data has gained significant prominence,
particularly among startups and online companies. Nevertheless, this is not a new challenge for
larger corporations, as many of them have been grappling with the complexities of managing vast
amounts of data for a considerable period. While some managers recognize the innovative
potential of big data, for most, it is viewed as a natural progression or evolution towards working
with more data. These established firms have been incorporating various types of data into their
systems and models for years, making the idea of big data less revolutionary than it might initially
seem. To put it differently, they were already embracing the principles of big data before the term
gained widespread attention.

The fascination with big data for these managers is not primarily rooted in its sheer
volume. Instead, it revolves around three distinct aspects: the absence of inherent structure, the
numerous opportunities it presents, and the cost-effectiveness of the underlying technologies. This
sentiment aligns with the findings of a 2012 survey conducted by NewVantage Partners, which
examined over fifty major companies. The results unveiled a noteworthy trend: organizations
accustomed to handling massive data volumes were now exhibiting keen interest in effectively
managing a new class of data. This encompasses diverse forms such as voice, text, log files, images,
and videos. For instance, a retail bank gains insights into its multi-channel customer interactions by
analyzing log files, while a hotel chain utilizes video analytics to study customer behaviors. Even a
health insurance company leverages speech-to-text data from call center recordings to predict
customer dissatisfaction. By merging structured and unstructured data, these firms are attaining
comprehensive insights into their customers and operational processes.

Furthermore, advancements persist in the domain of structured data derived from sensors
and operational data collection devices. Companies like GE, UPS, and Schneider National are
progressively integrating sensors into moving or rotating objects, capturing resultant data to
optimize their operations. Although seemingly incremental, even minor enhancements yield
substantial benefits when applied on a large scale. For instance, GE estimates that a mere 1%
reduction in fuel consumption for aircraft engines using big data could save the commercial airline
industry an astounding $30 billion over 15 years. Similarly, a 1% efficiency increase in global gas-
fired power plant turbines could translate to a massive $66 billion in fuel savings. This notion is
exemplified by UPS, which achieved significant savings through enhanced vehicle routing methods,
a case explored in the "Big Data at UPS" study.

The multifaceted implications and applications of big data technologies are analogous to
other emerging information technologies. Big data holds the potential to result in significant cost
reductions, faster computation times, and the development of innovative products and services. It
can also be employed for internal business decision-making. The strategies that organizations
adopt when implementing big data solutions are contingent upon their objectives and priorities.

Cost Reduction through Big Data Technologies:

Some organizations concentrate on harnessing big data to decrease costs. They believe
that technologies like Hadoop clusters offer greater cost-effectiveness for data storage and
processing in comparison to traditional methods. For instance, storing a terabyte of data for a year
might cost $37,000 with a relational database, $5,000 with a database appliance, and only $2,000
with a Hadoop cluster. However, this shift necessitates considerations of reliability and data
security, as Hadoop clusters might not provide the same level of robustness in these aspects.
Organizations aiming for cost reduction frequently make decisions within their IT departments
based on technical and economic criteria.

Time Reduction and Real-Time Interaction:

Another objective revolves around time reduction, where big data technologies are
employed to expedite complex analytical processes. For instance, Macy's optimized its pricing
calculations from over 27 hours to slightly over 1 hour, enabling swifter adaptation to changes in
the retail market. This time reduction also supports real-time interactions with customers, allowing
companies to engage with customers while they are still actively involved. Rapid data capture,
aggregation, processing, and analysis are pivotal in this context.

Development of New Offerings from Big Data:

Big data serves as a foundational basis for creating novel product and service offerings,
particularly for online companies. LinkedIn and Google exemplify this approach. LinkedIn employs
data-driven features like "People You May Know" and "Jobs You May Be Interested In" to attract
new users. Google consistently refines its search and advertising algorithms, integrating big data
into products such as Gmail and Google Apps. Beyond online firms, GE is acknowledged for
devising new services based on big data, such as optimizing service contracts for industrial
products. This objective propels companies to formulate comprehensive programs for product and
service development.

Supporting Internal Business Decisions:

Traditionally, data analytics have underpinned internal business decisions. Big data extends
this concept by incorporating unstructured data to derive insights into customer satisfaction and
multi-channel customer journeys. For banks, comprehending intricate customer paths through
diverse touchpoints is pivotal for retention and product sales. This entails analyzing substantial
amounts of unstructured data, including website clicks, transaction records, and call center
interactions. Furthermore, external data sources play a role in augmenting analysis, such as
utilizing supplier data to evaluate supply chain risks.

In conclusion, big data technologies offer a myriad of opportunities and benefits for
organizations. The chosen objectives, whether cost reduction, time efficiency, product
development, or decision support, dictate the strategies implemented and the organizational
changes necessary to effectively harness the potential of big data.

The evolution of organizational structures and skillsets within large companies to

accommodate big data is a dynamic process, intertwined with existing frameworks rather than
being freshly established. Notably, none of the surveyed organizations have created entirely new
entities solely dedicated to big data endeavors. Instead, established analytics or technology
divisions have extended their mandates to encompass big data functions. Some entities have
remained unaffected, highlighting their history of managing massive data volumes. Others
emphasize the integration of data science capabilities into their existing portfolios.
 Regarding the organizational structures conducive to initiating or accommodating big data
technologies, two primary models emerge. Firstly, existing analytics groups, including those labeled
as "operations research," play a significant role in initiation. Secondly, innovation or architectural
factions within IT departments also wield influence. Often, these central services units align their
efforts in big data initiatives with analytically-oriented functions or business segments. For
example, the marketing domain or the online branches of banks and retailers demonstrate such
alignments. Certain business segments boast their own IT or analytics teams. Notably,
organizations demonstrating effectiveness in this realm establish close collaborations between the
business divisions focused on big data and the supporting IT units.

Addressing the scarcity of skills, most large firms strive to supplement their existing
analytical personnel with data scientists possessing enhanced IT proficiencies. These data scientists
adeptly navigate big data technologies, deviating from traditional quantitative analysts. Their
skillset encompasses a range of abilities, including natural language processing, text mining, video
and image analytics, and visual analytics. Proficiency in scripting languages like Python, Pig, and
Hive is also prevalent. The backgrounds of data scientists vary, encompassing those with Ph.D.s in
scientific fields to adept programmers with analytical acumen. It is noteworthy that assembling a
single data scientist with all the requisite skills is often questioned, leading to a team-oriented
approach.

A crucial skillset involves the ability to communicate big data findings to executives through
visual displays or verbal narratives. Even among data scientists, the significance of "telling a story
with data" and effectively communicating with decision-makers is emphasized. Additionally,
analytics personnel often shoulder substantial change management responsibilities. As analytical
models are integrated into crucial operational processes, someone must facilitate the necessary
alterations in roles, process designs, and proficiencies. This entails collaboration with front-line
employees and process owners to effectively implement these changes.

A Tech Business Process Outsourcing (BPO) industry involves providing various

technological services to businesses, often encompassing IT support, software development,
customer service, data management, and more. Its customers range from startups to established
enterprises seeking to optimize operations and enhance customer experiences through
technological solutions. These services enable businesses to focus on core functions while
entrusting their tech-related tasks to specialized providers.

In this context, Big Data for a Tech BPO company refers to the immense volumes of
structured and unstructured data generated from various sources, including customer interactions,
service requests, system logs, and market trends. The opportunities presented by Big Data to such
a company are substantial. It enables them to gain valuable insights into customer preferences,
identify emerging market trends, optimize service delivery, and predict potential technical issues
before they escalate. By harnessing Big Data, Tech BPO businesses can enhance their decision-
making processes, improve operational efficiency, and offer more personalized and proactive
services to their clients.
 Tech BPO companies should prioritize several value propositions of Big Data. Firstly, they
can refine their service offerings by tailoring solutions to specific client needs, leading to improved
customer satisfaction and retention. Secondly, predictive analytics driven by Big Data can help
anticipate client demands and dynamically allocate resources, optimizing service delivery. Lastly,
data-driven insights can be utilized to enhance internal processes, resulting in better resource
management, cost reduction, and overall business growth.

However, implementing Big Data initiatives also presents organizational challenges. Data
security and privacy concerns must be addressed, especially when handling sensitive client
information. Developing and maintaining robust data infrastructure, including storage, processing,
and analysis capabilities, requires significant investment. Skill gaps among existing staff may
necessitate hiring or training personnel proficient in data analytics and management. Additionally,
integrating diverse data sources and ensuring data quality can be complex tasks.

In conclusion, the Tech BPO industry plays a pivotal role in delivering technological
solutions to businesses, allowing them to focus on their core competencies. Embracing Big Data
offers substantial benefits, including enhanced decision-making, personalized services, and
optimized operations. To fully leverage these advantages, Tech BPO companies must prioritize
refining their service offerings, predictive analytics, and internal process improvements.
Nevertheless, challenges related to data security, infrastructure, skills, and integration must be
navigated effectively for successful implementation and long-term growth.