Benefits of data-driven decision making

1) Data allows for informed, smarter, and

faster decision-making
Accurate and high-speed data analytics allows businesses to
make decisions in real-time, and confidently. Being able to
make decisions on the go and efficiently helps save
organizations precious time.
Example:
Let’s think of an FMCG company that makes – let’s say –
pre-mix pancake batter. During the product process, a
certain metric – say the temperature of the batter measured
by a sensor being at a certain level may indicate that the
batter is ready for packaging. The production line would
then automatically move the batter to the packaging
section, post which it would be ready to be put in boxed and
subsequently shipped. With the sensor in place, as opposed
to a person manually checking the temperature, the
organization can reduce its lead time and make faster
decisions on the go.

2) The second benefit is that data-driven

decisions help an organization reduce its
costs
When an organization starts to analyze its data its actual
costs become evident. This visibility helps organizations get
a better understanding of how the business is performing
and make necessary adjustments to optimize costs. Often,
the biggest challenge for any business is cost.
Example:
We would take the help of a leading Business Analytics
company, which was working on a project to reduce costs
for a leading American airline. It started its analysis by
historically analyzing cost structures; which included
salaries, cost of parking slots, cost of food & beverages
provided onboard, cost of entertainment provided onboard,
and fuel among other heads. It then looked at the areas of
cost reduction, in combination with customer preferences
concerning flight timings & services they preferred onboard.
Many first-hand interviews and surveys were conducted and
after making sense of the data collected and analyzed
through analytics, a decision was taken to remove jalapenos
from in-flight meals. Yes, you heard that right. This decision
to remove jalapenos from in-flight meals was how the airline
company started its cost reduction exercise.

3) The third benefit is that making

decisions through data makes way for
improved efficiency
Data analytics enables organizations to understand
bottlenecks in the business and take action to simple
processes. Operational efficiency can also be improved by
analyzing customer behavior to identify behavior patterns
and optimize the customer experience.
E.g., a problem statement for a business analyst that can be
solved using data can be reducing wait time at the check-
out kiosk of a supermarket.

From a product development perspective,

making decisions through data helps to
identify market trends.
Organizations can use data analytics to analyze trends and
identify need gaps and areas of new product development
or innovation. This helps to stay competitive through
changing demand patterns and take necessary steps in
anticipation of what customers may need in the future, or
the direction the industry is headed towards. Do recall that
in the first module; we spoke about how the newer trend in
product development, which is to first study the market and
then move towards the product.
Let’s take the airline industry as a broad example here.
Often, you might have observed that different markets have
different tastes and preferences regarding what they want
in the sky. Middle Eastern airlines like Emirates, Etihad
Airways, and Qatar Airways have their core company
philosophies in luxury and hence, they account for a full-
service experience onboard. However, if you wanted to
launch an airline in an emerging market like India, would
you follow the same philosophy? The answer would most
likely be no. One can easily use historical data and current
industry patterns to analyze that in India, full-service
carriers do not work, largely due to the price sensitivity of
consumers, while booking tickets. A lot of full-service
airlines like Kingfisher, Jet Airways, and Air India among
others have struggled in the country. Access to their cost
and sales data, which are publicly available, coupled with
research on the insights that Indian fliers make their
purchase decisions, would be suggestive of the trend of how
low-cost budget carriers have been more successful in the
country. Hence, this could be a case of how data can help
you identify a market trend and enable better decision-
making; while you craft your offerings.

Robust data monitoring systems are useful

from a security perspective.
All businesses face security threats. Organizations can look
into and diagnose the causes of abnormal behavior by
analyzing the relevant data using diagnostic and predictive
analysis. This helps to identify and address vulnerabilities
and introduce monitoring systems that raise alerts in the
future to ensure immediate action.
A common example of this is net banking. Let’s say that you
are on an online shopping site and you decide to pay for
your order via net banking. When rerouted to your bank's
page, you mistakenly key in the wrong password while
logging in and hence, the login doesn’t go through. You then
go back to your cart and attempt the transaction again. This
time, your key in the correct login ID and password, and
now when you proceed ahead. You may observe that your
bank will ask you for additional security authentication,
maybe an OTP, or an answer to a security question. This is
because, when you attempted logging in earlier, there was
a deviation from your regular login pattern that was flagged
by the system, and hence, an additional security
authentication came in. Now this deviation could be
anything – it could be keying in on a wrong password, or
even perhaps forfeiting the login attempt when a wrong
password was keyed in. Banks have created such systems,
which monitor click data - to enhance their security.

Data Analytics allows for personalization.

Organizations have access to data across both online &
offline channels. Analytics allows them to understand their
customers better which helps to create extensive consumer
profiles and gain specific consumer insights to tailor a
personalized experience for each consumer. Using this
information, predictive analytics can be used to identify
product recommendations that are likely to boost sales.
An example of this could be an individual setting up a new
home who goes on to purchase household items like
drinking glasses, cutlery, and crockery from an online
shopping website. The website can access this customer’s
order history, cart items, searches, reviews as well as clicks
of products visited while on the website. Using analytics, the
website can analyze the individual’s behavior and decide
which specific recommendations to pitch to the individual.
For instance, the website may start recommending items
like coasters, table cloths, or even electronic kitchen items
to the individual.
A plethora of marketing benefits arise out
of data-driven decisions.
Marketing and analytics go hand in hand. Marketing teams
use data analytics regularly to gauge the success of
marketing efforts and improvise if needed. It helps
organizations get a better sense of their target audiences,
provides useful insights into their behavior, enables
methods like A/B testing, and even assists with pricing
setup.
To understand this more in detail, you can take the example
of an FMCG manufacturer, who has recently launched a new
variant of cream biscuits and wishes to evaluate two
website banner advertisement designs for the biscuit using
A/B testing. A/B testing is an experiment that allows you to
determine which option performs best and thus is likely to
lead to more conversions.
Having shown the two design options at random to two
different sets of viewers on Facebook, the manufacturer
finds that option A is more effective in garnering click-
throughs and eventual purchases.
Using this information, the manufacturer can now fully
launch the more successful banner across various
platforms.
Analytics allowed the FMCG player to gauge the success of
their banners before launch, aiding their future marketing
efforts.
For our first case on data-driven decision-making, we would
be looking at Amazon and its tryst with the grocery retail
space in the US.

Amazon Case Study:

In 2017, Amazon forayed into the grocery space in the US
after its acquisition of Whole Foods, a major offline retailer.
This rang alarm bells for all major offline grocers in the
country. Suddenly, these grocers were forced to bolster up
their online ordering and delivery services, in an attempt to
level the playing field. The acquisition allowed Amazon to
finally tap into a huge space they had their eye on – offline
grocery stores. Many speculated that this deal would allow
Amazon to dominate grocery sales both online, which was
via Amazon Fresh, and offline through Whole Foods. Some
industry experts also wondered how Amazon Fresh and the
newly acquired Whole Foods would stack up, in a way
competing against each other.

Whole Foods was an American multinational supermarket

chain started in 1980 and built its reputation as a brand
known only to stock products that are free from artificial
preservatives, colors, flavors, and even hydrogenated fats –
broadly organic & natural food. The supermarket chain had
massive success with stores opening up all over the United
States. However, by mid-2000, the mainstream soon caught
on. Conventional stores started to stock similar offerings,
but at a more affordable rate. Financial troubles started to
ensue, and investors were unhappy. Amazon’s proposal to
acquire the supermarket chain came as a huge relief.

A set of industry experts believed that for Amazon, the

Whole Foods acquisition was a way to boost sign-ups from
Prime memberships in return for discounts on groceries. The
second set of experts believed that the acquisition was to
help educate Amazon about the world of offline grocery
businesses, which would then lead to the company opening
up large mainstream grocery chains across the country.
This second set of experts was right; when in 2020, Amazon
Fresh, the e-commerce giant’s online retail offering,
commenced offline operations. When Amazon purchased
Whole Foods, many industry experts believed that Amazon
Fresh and Whole Foods would compete against each other.
However, research conducted across Amazon Fresh &
Whole Foods stores tells us that Amazon Fresh customers
represent a different demographic compared to that of
Whole Foods. The former is more popular with frugal, price-
sensitive, and diverse shoppers, while the latter is popular
among wealthier, health-focused shoppers. Whole Foods is
known for its broadly organic & natural food, which primarily
meant - high prices.

Therefore, with Whole Foods & Amazon Fresh, the e-

commerce giant is catering to a larger, more diverse market
and we’re pretty certain that a lot of data insights on
consumer demographics and buying patterns were analyzed
by the company; before acquiring Whole Foods and this has
been a primary reason for the success of the venture.
Through the Whole Foods example, we saw how Amazon
used data insights as part of its mergers & acquisitions
strategy, with great success, in the grocery retail space.
Now, let’s look at how a video-based data generation is a
form of unstructured data that enhances the experience of
the customer. If we compare Amazon Fresh and Whole
Foods as a store, the former comes across as smaller and
less lavish compared to the latter. The items available at
Amazon Fresh are more local, and the interiors are simpler,
which makes them cheaper, quicker to build, and easier to
run. Amazon Fresh stores have given rise to the use of
video analytics like never before to make processes efficient
and faster.

Amazon Fresh’s ‘Just Walk Out’ enabled stores are one such
example. In these stores, customers enter using their credit
cards. They can then proceed to go about searching for the
items they need, as they would in a normal grocery store –
but here there’s a big difference. As they pick up the items
they want and place the items in their shopping bags,
cameras and sensors detect these products taken or
returned and ultimately placed in the shopping carts and
keep track of them. When the customer is done shopping,
they can simply ‘just walk out and the total bill amount is
charged to their credit cards used upon entry. As evident
from this case, by eliminating checkout kiosks for a cashier-
less shopping experience and using sensors, Amazon has
helped save customers some precious time.

Over time, Amazon would have access to tremendous

amounts of data which would enable them to constantly
improve their computer vision systems and algorithms. As
the technology gets better, the ‘Just Walk Out’ system can
be used at several stores in their now large grocery
business. By simply modifying existing stores to fit this new
system, Amazon saves time as well as money that it would
need for additional space & personnel. Another way through
which Amazon is attempting to develop a cashier-less
checkout is Amazon’s Dash Carts. You log into the Dash
Cart using your Amazon app, and the cart has cameras and
sensors installed that detect and identify the items you drop
into your cart. You can then automatically pay from your
account without having to wait in line during check out.
Therefore, we can see how – through the help of video-
based data generation, Amazon is changing the way
consumers shop in-store, while also saving costs.

Data-driven-decision-making in
Manufacturing
McKinsey is a leading management consulting firm, which
does sector agnostic work and has offices worldwide.
Traditionally manufacturing processes were done by staff
manually checking and writing down production as well as
operation and maintenance histories. These methods were -
time-consuming, open to bias & lacked quality of Analysis.
The current manufacturing process involves monitoring as a
crucial aspect.

Applications of Data-driven-decision-
making in Manufacturing
Three Main Applications:
1) Predictive Maintenance: Analyzing the historical
performance data of machines to forecast when one is likely
to fail, limit the time it is out of service, and identify the root
cause of the problem.

Data can be used by manufacturers by maximizing the

operating time of machines by anticipating their failures and
determining breakage of machines even before it happens –
—or be ready to replace them when it does—thus
Minimizing Downtime. Predictive maintenance through data
typically reduces machine downtime by 30 to 50 percent
and increases machine life by 20 to 40 percent.
2) Yield-energy-throughput analysis: To ensure
individual machines are efficient while operating, helping to
increase their yields and throughput and reduce the amount
of energy they consume.

3) Profit per hour maximization: Scrutinizes parameters

that have an impact on the total profitability of an
integrated supply chain right from raw materials purchasing
to final sales – providing intelligence on how best to
capitalize on given conditions.
Profit-per-hour-maximization optimizes the interaction of
machines and processes to dynamically maximize profit
generation in production and supply chains. Encompassing
every step from purchasing - production – sales.
Dynamically maximize profit generation in production and
supply chains.

This is a typical case of how data-driven decision-making

helped a manufacturer maximize PPH by integrating sales
and production.