A field-sales company: more than 70,000 messy records to an 84-name call list

The job

Reps logged every visit in a lead app. Over time it held more than 70,000 records: the same address entered several times at different stages, and phone numbers typed into notes instead of the phone field. Booked jobs lived in a second system. The owner wanted one list: prospects where a rep had found a real need, but no job was ever booked.

What I built

A step-by-step cleanup, checked at every step:

1. Group the duplicate entries by location.
2. Remove anything already signed, booked or closed.
3. Remove anyone already in the job system, matched by address and by phone.
4. Recover phone numbers from notes and related records.
5. Check every number and remove duplicates.

What changed

• 943 entries flagged with a real need came down to 868 unique locations.
• 711 were not already signed, booked or closed. 568 were not already in the job system.
• 84 had a phone number that could be recovered and checked. 63% came from the phone field, 27% from reps' notes, and the rest from related records.

What stayed in human hands

The owner set the goal for the list. A person called it.

Honest limits

I do not have the number of jobs booked from the calls, so I am not going to guess at it. Most of the flagged entries had no phone number anyone could recover, which is a lesson in itself: the cleanup is only as good as what was recorded.

Written by Devan Borsboom. Client names and identifying details (city, team size and anything specific enough to place them) are left out on purpose.