The Client Acquisition OS

When outreach produces nothing, the useful question is not whether outreach works. It is which layer failed. No replies at all usually means the list or the opening line. Replies but no calls booked means the follow-up or the ask. Calls but no close means the offer or the price. Closes but no retention means delivery.

Without that separation every failure feels like the same failure, and the response is to send more messages, which multiplies the broken layer instead of fixing it.

A cold message has four jobs in a fixed order: prove you looked, name one specific problem, show the fix in a line, and ask for something small. Reverse that order and it reads as a pitch, which is what gets ignored.

The part that makes the opening line work is research you actually did. Not a claim about helping businesses grow, but the enquiry you sent that nobody answered, or the call you placed at a specific time that rang out, or the advertisement pointing at a page with no way to make contact. Specificity of that kind cannot be faked and is very hard to ignore.

AI is genuinely useful here, but as a research assistant per prospect rather than a message generator. Fully AI-written outreach all sounds the same, and the people receiving it recognise it immediately.

Pricing from how long the work takes you is the fastest way to underprice, because the work gets faster every time you do it while the value to the client stays the same.

The alternative is to price from what one additional customer is worth to that business, and how many the system should recover. When a prospect states that number themselves during discovery, the pricing conversation stops being an argument.