Databases Reference
In-Depth Information
mind. For example, say you are working with an e-commerce company and your area
of analysis is how to decrease the percentage of shopping cart abandonment. You have
several approaches available: you can decide to provide more discounts for returning
customers, establish a price comparison interface for the same products across a few
well- known e-commerce sites, etc. However, let's say that while analyzing the data,
you realized that a large percentage of the shopping carts are abandoned at the final
stages due to issues in the payment gateway; armed with this insight you recommended
enhancement of the payment processing application, and the resulting effect was huge
upswing in revenues!
Can the data scientist uncover business themes where a use case can unlock
disproportionate revenue-making potential for the organization? How would a data
scientist go about finding the business themes to move the needle? Which are the best
“impact zones” in a business process which are “ripe” for big data?
Test-4: “Sniff The Domain Out” Test:
Can the data scientist “sniff the domain out” by examining analytical
outputs and getting the business to put the numbers in context?
Data-driven domain knowledge can reduce the learning curve required to
understand domain and is deeper than theoretical knowledge. A data scientist can glean
far more knowledge about the nuances of a business by getting his/her hands dirty on
exploratory data analysis (EDA), and eyeballing univariate and bivariate results.
Can the data scientist “sniff the domain out” by examining EDA outputs and getting
the business to put the numbers in context?
Test-5: “Actionability” Test:
Is data scientist only generating insights or he/she is also crafting a
solution to put the insights into action?
Insights are important, but actionable insights are far more important. You can
develop a list of insights, prepare suave-looking presentations with lots of graphs and
numbers supporting your findings, and the result could be a feel-good effect; but if you
are not able to deliver what actions businesses need, your work is useless! For example,
say you are working on a use case to spot the high-value customers who are vulnerable to
churn. During your analysis, you were able to find out the factors for churning, you were
able to also develop predictive models to identify who is going to churn and when. These
are valuable insights, but if you are not delivering a solution to prevent the churn, then all
your hard work is wasted.
Besides the insight generation, you also proposed a solution where high-value
customers who are vulnerable to churn away are redirected in real time to high touch
contact center agents who would call them instantly and offer an instant rebate to woo
them back.
As a data scientist, you have a larger role to play in operational actions.
