Deduplication

Find every duplicate. Across the whole CRM. At £0.00.

Different spellings, different formatting, same person. Datuma surfaces every duplicate pair across your full CRM export, with confidence scores, ready for your team to merge or keep.

The 1× / 10× / 100× rule of bad data

SiriusDecisions (now part of Forrester) published a benchmark that's been re-validated for two decades. Costs scale by an order of magnitude at each stage:

to verify a record as it enters the CRM

10×

to clean it up after it's already in the CRM

100×

if it's left untreated and acted on by sales, marketing, or finance

Duplicates are the single biggest source of that 100× cost. The same contact in three records means three sequences, three calls, three sales reps thinking they own the relationship. Pipeline gets inflated. Forecasts get fabricated. And the customer gets contacted by people who don't know about each other.

Why your CRM's dedup tool misses them

Different spellings

"Alasdair Carter", "Ali Carter", "A. Carter". Same person, three records. Exact-match dedup catches none of these.

Different formatting

One record has the email in lowercase, one has trailing whitespace, one has the organisation as "Acme Ltd" vs "Acme Limited". Off-the-shelf dedup walks past these.

Job changes

Same person, different organisation, different email domain, no fuzzy match on first/last name alone. Without identity-aware matching, they live as two unrelated contacts forever.

Per-import scans only

Most CRMs dedup within an import. They don't run a full pairwise scan across the entire database. So your 50,000-row CRM has thousands of hidden duplicates that have been sitting there for years.

How Datuma finds them all

Full-database scan, not per-import

Export your CRM. Upload once. Datuma runs a complete pairwise scan across every contact in the file. Every duplicate pair, no matter how old, no matter when it was added.

Fuzzy matching with confidence scores

Different spellings, casing, whitespace, and abbreviation differences are all matched. Each pair comes back with a confidence score and the matching evidence, so your team can review the ambiguous ones and trust the clear ones.

Within-batch and cross-batch

New batches are deduplicated within themselves and against every prior batch you've ever uploaded. No more "we already enriched this person three months ago" surprises.

Your team decides: merge or keep

Datuma never auto-merges. Each pair is presented in a review page with the matching evidence. Your team picks merge or keep separate. Decisions are remembered for next time.

Zero enrichment credits used

Deduplication runs before a single enrichment credit is spent. The full-database scan is £0.00. You only pay credits for fresh enrichment on contacts that aren't already in your system.

See your duplicate rate in five minutes

Upload a CRM export. Datuma surfaces every duplicate pair with confidence scores. Free to find them. Your team decides what to do.