“The list is the message” is a line from Jordan Crawford, and it’s the reason how you build a cold email list matters more than what you write in it. I’ve heard it quoted by three different people who don’t know each other, and it keeps getting repeated because the people repeating it keep finding it true.
In practice it means this. If the list is the 400 companies with the exact problem you solve, the email can be plain and short. If the list is 40,000 companies that matched a filter, the copy has a lot more work to do, and usually it can’t do it.
Start with who already bought
Most small companies own a list they’ve never used.
One owner I worked with had decades of customer records that had never been exported. Another had years of trade show contacts in spreadsheets nobody had opened. Both had paid for tools to find new customers before anyone looked at the old ones.
Export every closed customer from the last five years and look for what they have in common. It often isn’t what the owner expects, which is the reason to look.
An agency owner showed me a way to do this without a data team. Three files: a questionnaire about the client’s best customers, a set of researched profiles for those companies, and the closed won records from the CRM. Then a model is asked what the closed ones share. For that client, the pattern turned out to be the size of the sales team at the customer. Industry and headcount predicted nothing, and nobody had guessed that beforehand.
Yours may show a different pattern, or none. The method is cheap enough to try in an afternoon with a spreadsheet.
Filters and signals
A filter is industry, headcount, state. Every competitor with a database subscription can pull the same list.
A signal is something that changed recently. A second location opened. A new product line. A booth booked at the trade show you also attend.
The people doing this well lean on signals. One caution from my own results: I tested “recently hired a salesperson” as a buying signal across three campaigns, and it predicted nothing. What the companies that replied had in common was someone who could answer the phone. Whether they’d hired anyone lately made no difference. Test a signal on a few hundred contacts before you build a campaign around it, because mine sounded plausible too.
Where the data comes from
For standard B2B, the databases everyone uses cover most of it. Where they run out is the interesting part.
Someone who sells into hospitals walked me through building a list when the databases come up empty: state licensing boards, procurement portals, compliance filings. Slower, and nobody else has that list.
Local businesses are the other gap. A plumber with three trucks is in no B2B database. He’s on the map, with a phone number and a review count, and that’s where the list comes from.
One agency owner described the list data built into his enrichment tool as a last resort, used after the primary sources have been tried. My experience with built in data has been similar.
Verify before anyone hits send
A list with bad addresses does two kinds of damage. The email doesn’t arrive, and each bounce tells the mailbox provider your domain sends to addresses that don’t exist.
Verify and deduplicate every list before it’s sent. One person described a campaign to me with a 30% bounce rate two days in. He stopped it that afternoon, and that was right. What counts as too high for your domain I can’t say precisely. A number like that is nowhere near it.
On size, the rules of thumb I hear from people who do this at volume are 800 to 1,000 contacts per angle and 1,000 to 2,000 per campaign before you can tell whether it’s working. My own campaigns run 900 to 1,900 contacts each. Below a few hundred you’re mostly reading noise.
Questions we get
Can I buy a list? You can. The ones sold cheaply are the ones everyone else bought. Lists built from your own customers, or from a source your competitors haven’t found, are the ones to start with.
How big does the list need to be? 1,000 to 2,000 per segment to learn anything. The total market matters more. If there are only a few thousand companies that could ever buy from you, cold email will run out of people before it pays for itself.
Is the data in my enrichment tool good enough on its own? Treat it as a supplement and verify everything regardless.
If you’re not sure whether the list you have is worth building on, we look at it on the first call before anything gets bought.
