0
$\begingroup$

I'm working on a project to score my leads and customers based on the meeting our sales guys do with them. These are the fields-

  1. customer_id
  2. date (date when visit is completed)
  3. schedule_status (date when meeting was scheduled as they can )
  4. next_action_date( date if any next meeting scheduled on the present visit)
  5. status( 1- done 0 - missed,3 cancelled)

So I worked with decision tree but I'm not able to find numerical features from these. I thought of-

  1. days_since_last_visit
  2. total_visits
  3. total_done_visits
  4. total_missed_visits
  5. total_cancelled_visits

so basically from a visit, I need to score a lead. So how do I use feature engineering in this? How can I create numerical data so I can apply some algorithm on this? Or otherwise, I will need to set rules and then score accordingly.

$\endgroup$

closed as unclear what you're asking by Jan Kukacka, Michael Chernick, kjetil b halvorsen, mkt, Peter Flom Aug 15 at 11:52

Please clarify your specific problem or add additional details to highlight exactly what you need. As it's currently written, it’s hard to tell exactly what you're asking. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • $\begingroup$ When you say “score a lead”, what is the purpose of that score? Are you trying to predict something? $\endgroup$ – Joe Aug 14 at 11:44
  • $\begingroup$ yeah! 1. which customer is better to serve 2. lagging customers 3. leaderboard thru this data 4. also if some user is missing a customer basically I want to create a bunch of recommendation so that you can follow up this lead, or lead "a" is not good because of blah blah like this $\endgroup$ – Akshit jain Aug 14 at 11:48
  • $\begingroup$ I tried using RFM model for scoring, But I dont have monetary data. Only recency and frequency I can generate $\endgroup$ – Akshit jain Aug 14 at 11:51
  • $\begingroup$ Creating good features is an art of its own and I don't think this question can be easily answered by us. You (should) know the data better than anyone else and you should be able to figure out which useful pieces of information you can make out of your dates. $\endgroup$ – Jan Kukacka Aug 14 at 12:40