The step that decides what everything else costs
Sourcing is cheap and contact enrichment is not. That single fact drives the economics of an automated desk, and it is why the order of operations matters more than the choice of provider.
A search returns three hundred companies. Enriching all of them means paying to find contacts at the two hundred and forty that were never going to be your clients. Qualifying first is free, and it cuts the bill by roughly the proportion of the list that does not fit.
What you get back
Two lists, not one score. The companies that pass and the companies that do not, kept apart, so the qualified set is ready to hand straight to an enrichment play and the rejected set stays available to argue with.
A reason on every rejection. Which part of the profile the company failed, in a line. This is what makes the gate correctable: a filter that is throwing out companies you would have taken is visible in the reasons long before it is visible in your pipeline.
A tier and a one-line justification on every company that passes. The judgement pass says why this one resembles your clients, so the ranking is something you can check rather than a number to trust.
The bill you avoided. What contact enrichment would have cost on the whole list, against what it costs on the qualified subset. That number is the argument for running the gate at all.
Three funnels, narrowing
Hard filters run on data you already have. Size, industry, geography, exclusions. Free, and they remove most of the list.
Judgement runs on what survives. This is the part a filter cannot do: whether the company actually resembles the ones you win, which is often about their stage, their structure or how they buy rather than their industry code.
Verification runs only where the answer is genuinely unclear, and it is the only expensive part. It goes and looks at the company rather than guessing.
Each funnel is cheaper per company than the one after it and removes more of the list, which is what keeps the whole thing proportionate.
Why the profile has to be written down
An ideal client profile that lives in someone's head cannot be applied consistently and cannot be argued with. Written down, it becomes something you can point a gate at, correct when it drops a company you know is good, and hand to a colleague.
Derive it from the clients you actually make money on rather than the ones you would like to. The gate is only as good as that definition.
Where this goes wrong
Skipping it because the list looks fine. Lists always look fine. The rejects are what tells you whether it was.
Filtering on a currently open role. Whether a company is hiring today is timing, not fit. A company that matches your profile and has nothing open is still a target account.
Verifying everything. Verification is the paid pass. Running it across the full list instead of the borderline cases converts a nearly free gate into an expensive one for very little accuracy gained.
