How to identify inflated accuracy claims?

You're shopping for an email validation tool, and one vendor promises 99.9% accuracy. Another says 98%. A third just says "industry-leading." How do you know which claims are real and which are marketing fluff? Turns out, there are some pretty reliable tells.

The first thing to look for is whether the vendor ever returns an "unknown" or "catch-all" result. Real validation is honest about the addresses it can't fully verify. Catch-all servers accept every incoming email during the SMTP check, so no tool can confidently call those addresses valid or invalid. If a vendor marks every catch-all as "valid," they're either not running real SMTP checks or they're just telling you what you want to hear.

Speed is another giveaway. Genuine SMTP checks take time. The tool has to open a connection to the receiving mail server and wait for a response. If a vendor processes 100,000 addresses in a few seconds, they're not doing real SMTP checks. They're probably just running syntax checks and DNS lookups, then inflating the "valid" bucket to look impressive.

Here's what realistic results actually look like on a typical business list:

• Valid: somewhere around 70-85%
• Invalid: 5-15% (hard bounces waiting to happen)
• Catch-all or unknown: 10-20%, depending on how many corporate domains are in your list
• Disposable or role-based: a small percentage, but never zero on a real list

If a vendor hands back 99% valid with nearly nothing in the catch-all or unknown buckets, something's off. That's not accuracy. That's a vendor inflating their numbers to look better than the competition.

The other thing to ask about is transparency. Does the vendor explain what checks they actually run? Do they distinguish between syntax, DNS, and SMTP-level verification? Can they tell you how they handle role addresses (think info@, support@, admin@) and disposable domains? Vague answers like "proprietary AI" or "real-time verification" without specifics are a red flag. Good vendors are happy to explain their process because they know it holds up.

And if you want to test a vendor before committing, run a sample list with a healthy mix of corporate emails, free webmail addresses, and a few obvious typos. Then check how many unknowns and catch-alls come back. If the unknown bucket is suspiciously empty, you've learned something useful. (The previous question walks through exactly how to test a vendor's accuracy before you sign anything.)

One honest benchmark: the realistic accuracy ceiling for email validation is around 98-99%, and that's only for addresses that can actually be checked. Catch-alls will always sit in a grey zone. Any vendor claiming to solve that problem with a magic number is overselling.

If you want a second opinion on a tool you're already using, or you're not sure how to read the results you're getting, feel free to reach out through our SOS hotline. No pitch, just honest help.