Ad optimisation should not depend on who you can afford to hire

Why we exist

The gap between a well-run and a poorly-run ad account is enormous — often 30% or more of total spend. Yet the tools available to most advertisers are reporting dashboards that show you what happened, never what to do about it.

Agencies fill that gap, but they cost $2,000–5,000 a month and work on a human schedule. Freelancers are cheaper but cap out at a handful of accounts. Meanwhile the actual work — reading search terms, spotting fatigue, checking tracking, reallocating budget — is repetitive and rule-driven. It is exactly the kind of work software should be doing.

What we build

MyAd monitors Google Ads and Meta Ads continuously, generates prioritised recommendations with the evidence attached, and can apply them — either with your approval or autonomously within guardrails you set. Everything is logged and reversible.

We also expose the entire platform as an MCP server, so the work can happen inside Claude Code, Cursor, Windsurf or an agent you build yourself. A dashboard is one interface. It should not be the only one.

How we think about AI

AI in advertising has a trust problem, and it is deserved. Too much of it is a black box that spends your money and explains itself afterwards. Our position is simple: show the evidence, bound the blast radius, log everything, and make undo a first-class feature.