Data Science & Insights:Overview
Real life business problem is a lot more complicated than what can be modelled or simulated. A broader business problem can be segmented or broken into different parts to understand linkages and interactions between different parts.
An Exploratory Data Science & Insights project can support business in understanding components & factors important for the business. DnI Consulting works with business managers to break the business problem into different parts, prioritize the components which are the most relevant, convert business problem into analytical problem, solve the business problem and provide actionable insights.
Objective Consulting Approach: Overall Approach to solve Un-defined Business Problems:-
Current context and historical perspective.
data and existing analysis and reports.
Create scenarios and approaches with advantages and challenges to help business get the full perspective.
Work with business stakeholders to agree on the final scope and approach
Leverage Data, Statistical & machine Learn techniques, and tools & technology to solve the business problem. Outcome is well defined actionable insights for business managers to act on.
Share & Recommend:
Share the final insights in business language and help business managers to deploy the insights.
Data Science and Insights Scenarios
Examples-Retail: Identifying products bought together and leverage them for promotions:-
- 1) Consolidating customer transactions data after defining analysis window.
- 2)Sales Basket Product Count Analysis and excluded baskets with product counts less than 4.
- 3) Considering luxury retailers has very specific ranges for Male and Female, we have segmented the baskets.
- 4) Product affinity analysis to find significant product combinations and their confidence level.
- 5) Select a few key combinations for business managers.
Example-Banking:Credit Card Add On:-
For a credit card provider, we want to understand purchase patterns on the cards. A credit card is approved to worthy customers based on their application scorecard value. And a customers who have been approved a credit card, can take an add-on (additional) card for his or her spouse. So, one credit card accounts can have multiple cards associated with.
Aim of analysis was to understand usage pattern at a card level. Who use the card and how spending patterns are different? Leveraged a structured approach to formulate problem, identify & prepare data for the analysis and generated insights which Marketing Managers can act on.
One of the key insights was that significantly higher % of add-on card holders with high spend levels were Female customers and main card holders were Male. Hence, marketing managers can communicate to the female users and with right retail purchase offers