How to audit your data vendor spend
A data vendor audit inventories every contract, measures what each source uniquely contributes, and identifies overlap, unused capacity, and stranded credits. Most teams running four or more providers find meaningful duplication, because contracts are typically added over time without anyone testing marginal coverage.
Data spend accumulates rather than gets decided. A tool is added for one campaign, another for a region, a third because a rep asked. Nobody owns the total, and renewals land in different quarters so the full number never appears in one place. The audit's first job is simply making that number visible.
Step 1: Inventory every contract
For each provider, record annual cost, renewal date, notice period, what is contractually included, and who owns the relationship internally. Notice period matters more than people expect: many contracts auto renew with 60 or 90 days notice, so the window to act opens well before the renewal date.
Step 2: Measure unique contribution, not match rate
This is the step that finds the money, and the one most audits skip. Match rate on its own is misleading, because two providers can both match 70% of a list and be matching largely the same records.
Run one sample list through every provider you hold, then compare results record by record. For each provider ask: how many records did this source match that no other source matched? A provider with a high match rate and near zero unique contribution is duplicate coverage you are paying for twice.
Step 3: Find stranded capacity
Compare credits or records purchased against those actually consumed, per contract, over the last full term. Unused balance that expired is pure loss, and in credit based contracts it is common. If you are consistently using 60% of a plan, the plan is wrong or the commitment should be renegotiated downward.
Step 4: Check what you are re-buying
Take a sample of records acquired in the last quarter and check how many already existed in your CRM at the time of purchase. Without suppression against your own database, you are paying to reacquire contacts you already own. At volume this is often one of the larger single line items in the audit.
Step 5: Test quality, not just coverage
Coverage without accuracy is not worth anything. For contact data, track bounce rate on emails and connect rate on phone numbers, attributed back to the source that supplied the record. A provider matching 80% of your list at a 12% bounce rate is worse than one matching 65% cleanly.
What the audit typically finds
- Overlapping coverage. Two or three providers competing for the same records in the same segment.
- A regional gap nobody priced. Broad global providers thinning out in a market that matters, with no specialist filling it.
- Expired credits. Capacity purchased and never used.
- A source nobody can defend. Renewed because it always has been, contributing nothing measurable.
Consolidation follows from the findings rather than being the goal. Sometimes the right answer is fewer providers. Sometimes it is the same number, reordered, with one swapped out.
Running it on a schedule
Once is useful, annually is better, because coverage degrades and providers change. Align the review with your earliest renewal notice period so findings can actually be acted on rather than noted for next year.
While you have the contracts open, confirm each provider can supply current sourcing and lawful basis documentation. Under the GDPR that is a question you may have to answer at short notice, and renewal is the natural moment to ask.
Common questions
The contract inventory is quick. The coverage testing is the real work, since it requires running the same sample through every provider and comparing record by record. Expect a few weeks for a stack of four or more.
Then overlap is less likely and the audit is mostly about coverage gaps and pricing. It is still worth measuring unique contribution, because two providers can still be substantially duplicating each other.
Yes, and it is free, because it is also how we establish whether consolidating through our desk would actually save you anything. If it would not at your volume, we say so. See data buying.
We will run it against your actual contracts
Share what you currently hold and spend. You get the findings whether or not you go on to work with us.