Putler Feb 6.docx

I agree to all strategies here are some of my strategies Sir : 1. We should add Sankay and chord charts they are very very important. I would like to know where exactly would the Sankay and Chord chart be used? As in what kind of data would better suit this type of charts according to you? I believe, these charts better to show flow of data. So customer journey could be a good fit. What are your ideas on it? 1 We can do churn or refund prediction through python and R Agreed. Churn and refund prediction are good. But with my experience of Time machine I don’t see it as a burning need because this is probable data which might never sum up to actual figures. Instead, renewal prediction based on current performance of the business could be of higher value. Say how many renewals are due on this date?

3. We can do propensity to buy, customer segmentation and customer buying behaviour. Elaborate on how will you perform this? We already do RFM so will you be further segmenting these RFM segments for deeper analysis? Will it also include ● Which channel a particular customer is more receptive too? ● Is he willing to take upsell/ cross-sell offers? ● What type of a customer he is - discount driven, price conscious, needs proof etc?

4. I have seen on first month 18000 was the sales then it increased to 23000 and then it decreased to 17000 This is a good trend which might be seen in many clients businesses. What kind of personalized actionable strategies will Putler prompt these clients?

Say: Your THIS CHANNEL got so much TRAFFIC which converted because of THIS MARKETING STRATEGY in first month.

5. I we can do detailed RFM analysis understand the customer segmentation. Using clustering, scikit learning ,PCI, decision tree analysis. What type of segments can you create? 6. It was mentioned in sheets if we get less sales then there are many factors contributing to this improper content, irrelevant and text, price high, no logical targeting. In such cases we shall mine the historical data and help people in dept visualization. Techniques and more elaborative graphs. Great idea if we execute it well.

7. We can do sunburst analysis for each Customer current location analysis. Customer journey can be done using Sunburst

8. Whenever we get sales high peaks we should try to know why it happened and on same strategies we shall implement new strategies Eg seasonal, some offer or discount or some Experimental success such as landing page improvisation, less profit margin on a launch of huge campaign. So revenue generated eventually shall contribute a great amount to Putler. Finding such anomalies and correctly pointing the reason out to Putler users will be really useful. 9. We shall use confusion matrix to draw correct decision and accuracy of model.