Vetting phone lists before you buy: which checks actually catch the junk
If you buy lead lists or bulk CSVs from brokers, you already know the drill: the file looks clean in a spreadsheet, and the problems only show up after you pay to process it - or worse, after your first campaign underperforms or fails delivery checks.
Most "is this number valid?" questions collapse into five checks, and each one answers a different question:
1. Format - is the row even a phone number (E.164 syntax, correct country code)? Junk rows fail here first, and they are the easiest to kill before anything else runs.
2. Duplicates - is the same contact in the list twice (or three times)? If you pay per record, duplicates are pure waste.
3. Line type - mobile, landline, VoIP, or premium-rate? This decides whether SMS/WhatsApp are even possible for that number.
4. Carrier/portability - which carrier actually serves the number today? MNP means numbers move between carriers, and old data lies about who serves them.
5. Registration signals - is the number actually live on WhatsApp/Telegram/iMessage/RCS? A perfectly valid mobile number can be completely unregistered on the channel you planned to use.
Practical habit: run a preflight on a sample before you pay for the full list. If 15-20% of the sample fails format or duplicate checks, the rest of the file is not going to be any better - that is useful information before you negotiate or process.
Example: a 100k record file where 12% are unformatted, 8% are duplicates, and 20% are VoIP/landline rows when you only wanted mobiles - that is 40k records you would have paid to process or try to message. Catching them before processing changes the economics of the whole campaign.
If you are cleaning or verifying phone/email datasets before outreach, you can check the available verification workflows at
- but the habit matters more than the tool: check before you pay to process, not after.
What do you all do when a broker's list turns out to be mostly junk - refund request, re-process, or walk away?
Replies
No replies yet.
Please login to post a reply.