CRMs at growing SaaS companies often turn into a swamp of duplicates and half-empty records. A nightly Make scenario merges dupes, enriches missing firmographics and flags bounced emails, and the first week of runs can clean around 12,000 records.
Typical results
1%
Duplicate rate, down from 19%
2%
Email bounce rate, down from 11%
9 hrs/wk
Manual cleanup eliminated
Figures reflect typical results for this kind of automation.
The challenge
Three years of imports, form fills, and rep shortcuts left the CRM with a 19% duplicate rate and thousands of contacts missing industry, company size, or a working email. Reps double-called the same prospects, segmentation was guesswork, and marketing emails bounced hard enough to threaten sender reputation. A manual cleanup was quoted at three months of admin work.
Before and after
Duplicate contact rate
−95%Email bounce rate
−82%Weekly manual cleanup
−94%The automation
We connected a nightly Make scenario that sweeps the full database at 2 AM. Fuzzy matching on email, name, and company domain merges duplicates while preserving the richest field values and full activity history. An enrichment API fills missing firmographics — industry, headcount, role — and an email verification step flags bounced or invalid addresses and suppresses them from active sequences. Every morning, a Slack digest reports exactly what was merged, filled, and flagged.
How it works
- 01
2 AM sweep
A nightly Make scenario runs over the entire contact database while the team sleeps.
- 02
Duplicate merge
Fuzzy matching on email, name, and domain merges dupes, keeping the richest record and full history.
- 03
Firmographic fill
Enrichment APIs add missing industry, headcount, and role to every incomplete contact.
- 04
Bounce flagging
Invalid and bounced addresses are flagged and suppressed from active sequences.
- 05
Morning report
A Slack digest lists everything merged, filled, and flagged before reps log in.
Starts from
- Messy CRM records
- Enrichment APIs
Delivers
- Merged contacts
- Verified emails
- Morning data report
The impact
The first week of nightly runs typically cleans around 12,000 records, and the duplicate rate falls from about 19% to under 1% and stays there. Email bounce rates drop from around 11% to 2%, segmentation works again, and reps stop spending Friday afternoons fixing data, about nine hours a week back per team.
Questions about this automation
How does the merge avoid combining two different people?
Matching uses several signals together, such as email, name and company domain, and only pairs above a confidence threshold are merged automatically. Borderline pairs, such as the same name at different companies, go to a review list instead. When records merge, the most complete field values and the full activity history are kept.
Where does the enrichment data come from?
An enrichment API looks up each incomplete contact by email or company domain and fills missing fields such as industry, headcount and role. Fields a rep has already filled are left as they are. A separate email verification step checks each address and suppresses bounced or invalid ones from active sequences.
What does the morning report include?
A Slack digest lists how many records were merged, which fields were filled and which addresses were flagged overnight, with links to anything waiting for review. This is a multi-step build, usually rolled out in stages, often starting with a one-time cleanup of the full database before the nightly run takes over.