In big data analytics, recommendation engine is used. A recommendation engine uses transaction data — how do customers actually behave. Its strength lies in using the actual data, and then layering it with social data.

Before using a recommendation engine, it is necessary to understand a business problem. Then we have to understand a client’s data, and how the data can be leveraged to solve the data problem. Lastly, right model is used to solve the problem. Though algorithms can solve a mathematical problem, they do not solve always a business problem. We have to use creativity to understand the problem the recommendation engine is solving. Recommendation engine is content based. Then there is collaborative filtering which is a general name used for algorithms which recommends ‘ if you like that you may like this too.’ It is more of retargeting. In Indian context, recommendation engines can be used to reduce churn, improve engagement, and in cases where products are not selling in spite of being good. Many companies do not have much transaction data. The result will refine over a period of time. A company should do a proof-of-concept study, and commit investment in recommendation engines after seeing the business outcome.