A growing share of the software businesses buy arrives with contact data attached, or collects it on their behalf. Lead generation providers, CRM platforms with built-in enrichment, list vendors, marketing automation tools with contact databases, and the newer wave of AI-driven sales tools that source prospects automatically.
Buyers scrutinise all the usual things before signing. Price, features, contract length, support terms, whether the integration will work with what they already run. Very few ask how the contact data was verified, or whether it was verified at all.
That omission is worth correcting. Data quality is a procurement question rather than a technical one, and it is an entirely fair thing to raise with a supplier before money changes hands.
Where the data in your tools actually comes from
Most contact databases are assembled rather than collected. Providers compile records from public sources, from partner data, from user contributions, and from web crawling. Those records are then matched against each other, deduplicated, and formatted into the tidy rows a buyer eventually sees in an interface.
The tidiness is presentation. A neatly formatted email address sitting in a CRM tells you nothing about whether a message sent to it will reach a real person. Formatting and accuracy are separate qualities, and only one of them is visible to the buyer.
Serious providers deal with this by running their records through a verification layer before shipping them. That layer checks that each address is properly formed, that the domain can actually receive mail, and that the individual mailbox is likely to exist. Providers who take accuracy seriously build this into their pipeline and re-run it on a schedule as records age. That layer is usually a dedicated email verification api that a data provider integrates into its own systems, checking records at volume before they reach a customer’s account.
Not every provider does this. From the outside, there is usually no way to tell which ones do.
What poor data quality looks like once it reaches you
The symptoms arrive early and are easy to mistake for other problems.
The most obvious is a bounce rate on the first campaign to a purchased list that runs noticeably higher than on the company’s own data. Another is a large share of records coming back flagged as uncertain when run through an independent check, which suggests the supplier either never resolved them or chose not to disclose that they had not.
There are subtler versions. Contacts who have genuinely left the company the record says they work for, which is a currency problem rather than a validity one and will not show up as a bounce. Records that are technically deliverable but route into shared or unmonitored inboxes, so they never produce a reply and never explain why.
None of these are the buyer’s mistakes. All of them become the buyer’s problem, because the reputational damage from sending to bad addresses attaches to the buyer’s sending domain rather than the supplier’s.
Questions worth asking before you sign
None of the following requires technical knowledge, and all of them are reasonable to put to a salesperson.
- How is the data verified, and when? Ask specifically whether records are checked before delivery or only at the point they were first collected. The gap between those two answers can be several years.
- How often is the database re-verified? Contact data ages continuously as people change roles and companies restructure. A provider that verifies once at collection and never again is selling a snapshot of the past.
- What proportion of records are you confident about? A supplier who answers this precisely is one who measures it. A supplier who deflects usually is not measuring it at all.
- What happens to records the verification cannot resolve? Some providers remove them, some flag them clearly, and some ship them alongside everything else with no distinction. Which of those a supplier does matters a great deal.
- Can we test a sample first? Reasonable suppliers agree to this without much argument.
Testing a sample is the shortcut
The quickest way to assess a supplier is to take a few hundred records and run them through an independent check before committing at scale.
Two figures are worth looking at. The proportion that come back invalid is the obvious one. The proportion that come back uncertain is the one most buyers overlook, and it is often the more revealing number, because it shows how much of the file the supplier has handed over without resolving.
This takes very little time and costs very little. It also changes the conversation. A buyer holding a number can ask for a better price, a cleaner sample, or a different arrangement entirely, and the discussion stops being about promises.
The bottom line
Data quality has quietly become a supplier selection criterion in the same way uptime, support, and security already are. It simply has not made it onto most procurement checklists yet.
Businesses that ask these questions tend to get better data, partly because they end up with better suppliers and partly because suppliers behave differently with customers who are paying attention. Businesses that do not ask usually find out where they stand eventually, though normally after a campaign has already gone out.
Leave a Comment