What problem does Roger solve?
Roger solves the gap between “AI can manage Google Ads” and “your AI is actually managing Google Ads tomorrow.” The first half is true today. Claude and GPT-4-class models, given the right tools, can read campaigns, spot wasted spend, draft negative keywords, and propose bid changes with the judgment of a senior PPC strategist. The second half is where every agency we’ve talked to gets stuck: OAuth flows, MCP servers, developer tokens, custom skills, guardrails, routines, multi-user permissions. Months of engineering before a single ad gets optimized.
Roger is the production-ready version of that infrastructure. You connect your Google Ads account, and within sixty seconds an AI agent has read the last thirty days of your campaigns, run eight checks in parallel, and handed you a list of things to fix. Read-only by default. No card, no contract.
How is Roger different from Optmyzr or other PPC tools?
The honest answer: Optmyzr and the rest are rule-based systems with a UI on top. They tell you “this keyword has zero conversions, consider pausing it.” Roger is an agent. It reasons about your account the way a strategist would, weighing budget pacing, conversion lag, seasonality, and intent. The output is the same shape (a recommendation), but the process is different, and the recommendations get better as the underlying models do.
The other difference is posture. Most PPC tools assume you’ve already given them write access to your account. Roger is read-only by default. Every change goes through an approval workflow. We built it that way because we’d never hand an unsupervised system the keys to an account we spent years building.
Is it safe to give an AI access to my ad account?
Short answer: yes, if the AI is read-only by default and changes go through an approval queue. That’s Roger. We grant scopes incrementally. The free audit is read-only. To let Roger apply changes you explicitly grant write access, and even then every action is logged and reversible. We’re a Belgian company, EU-based, GDPR-aligned, and the data residency story is short: your Google Ads data is fetched on demand and not warehoused.
The longer answer is that this question is about trust, not technology. We answer it by being founder-led, by publishing this blog under real names, and by giving you an account that can’t move money without your fingerprint on it.
Built by advertisers
Between Jean and me, we’ve spent hundreds of thousands of dollars on Google Ads, managing dozens of accounts and building several SaaS products along the way. We know what good PPC management looks like because we’ve done it, at scale, with our own money. We’ve felt the frustration of wasted spend, the tedium of weekly audits, the risk of a budget that runs unchecked overnight.
When Claude Code came out, we saw the potential immediately. AI could finally manage Google Ads the way a senior PPC strategist would. We also saw the problem. Every agency we talked to wanted to build their own AI agent, and every one of them got stuck on the same wall. Months of engineering before a single ad got optimized.
So we built Roger. Everything we wished we had on day one. Now agencies can skip the infrastructure layer and start managing campaigns better, today.
What’s next
The product is live, the free audit is free forever, and we’re adding skills (Roger’s term for “things it can do”) every week. The next post will walk through how Roger handles a multi-account agency (MCC) setup, which is where the boring infrastructure work pays off in interesting ways.
If you want to try it: run a free audit. Sixty seconds, read-only, no card.