Paid ads for bootstrapped founders

Because it does not work for everyone, and taking the money anyway would be the easy version of this. Paid ads need an offer that already converts, some real customers to learn from, and enough margin that acquiring one profitably is arithmetically possible. A company still searching for its offer will burn the budget finding out, and no amount of campaign management changes that.

The application is a filter in both directions. If the constraint is your pricing, your landing page or your follow-up rather than your ads, that is what we will say — before there is an invoice, not after three months of it.

The five layers that determine paid performance sit in a causal order, and fixing one out of order wastes the work: whether the data can be trusted at all, whether demand exists and is being delivered to, whether you are competitive in the auction, whether the site converts what arrives, and whether the unit economics support any of it.

A layer further down cannot be assessed while a layer above it is broken. If tracking is undercounting, every efficiency number below it is wrong, and optimising against wrong numbers moves the account confidently in the wrong direction. So the first engagement is diagnostic, and the first change is usually not to a campaign.

Bootstrapped starts at £800 a month. It rises £200 a month, and only in a month where your profit grew by more than the step. A flat month costs what the previous one did. A bad month does not cost more than a flat one.

That is deliberately not a percentage of spend. Percentage-of-spend pays an agency more for spending more of your money, which is exactly backwards for a company funding growth out of its own margin. Self-serve at £19 exists as a way in, but it is a lead generator for this service rather than the service itself.

A founder-funded company usually cannot justify a senior paid-media hire, so the work lands on the founder, at the end of a day already full. What gets lost is not the campaign edits, it is the continuity: what was changed three weeks ago, what happened after, and whether this month's dip is the change or the season.

That continuity is the part the AI is genuinely good at. Every decision and its outcome is recorded, so a question like whether a bid change worked has an answer from the account's own history rather than a recollection.