Rahman publishes a nine-step lookalike list playbook
The process is specific, the evidence is absent, and that gap is the whole story.
Adam Rahman posted a nine-step "Lookalike Audiences Playbook" for building outbound lists off your existing customers rather than off filters. It starts with the customers you would sign again tomorrow, pulls the exact words their buyers used on calls, and ends with calling your best fits and automating the rest. There is no reply rate, send volume or meeting count anywhere in it.
What the nine steps actually say
Step one is the best customer set, defined by whether you would take them again rather than by revenue or logo. Step two is language, taking the words buyers used on calls instead of writing copy from scratch. Step three is working out what those customers share and testing that pattern on your own customer base first, before it ever touches a cold list.
The list build is where the numbers live. Rahman says to build lookalike lists from up to 10 seed websites each, then read every website before paying for a single contact, then find the right buyer only at the accounts that pass that read. Every email gets verified twice before anything sends. The copy instruction is one line, lead with a case study, which fits the earlier steps since the case study is the thing a lookalike account is most likely to recognise itself in.
The seed cap and the manual site read are the two choices a practitioner would notice. Ten seeds per list keeps each list narrow enough that one case study can carry it, and reading sites before buying contacts moves the qualification step ahead of the spend rather than after it. Double verification before send is the only deliverability control in the whole sequence.
What is missing
Nothing here has a result next to it. Rahman does not say how many lists he built this way, how many contacts survived the website read, what the verification pass rate was, or what any of it replied at. That makes this a process to copy if you like the logic, not evidence that lookalike seeding beats a well filtered list from the usual sources.
For people sending, the cost sits in step five. Reading every seed-matched website by hand is the slowest part of the build and the part most people skip or hand to a model, and Rahman's own framing puts it before the contact spend, which only pays back if your contact data is expensive enough to make the manual pass worth the hours. Treat the sequence as a hypothesis with a named author, and if you run it, run it next to your current list so you have the comparison Rahman did not publish.
Read every website before you pay for a single contact
Start with your best customers, the ones you'd sign again tomorrow
Know what is landing before you burn another domain
One email a day. Under three minutes. What the operators found out yesterday.