A live, AI-powered SaaS product, designed and shipped solo.
CouncilAds is a working web app that tells advertisers whether their ad will land before they spend a dollar on it. A user submits their product, the app generates competing ad versions, and a council of twelve AI buyer personas scores and critiques each one, then hands back the winning ad, a review of the landing page it points to, and a plain-English campaign setup guide.
It is live, it takes real payments, and it was built end to end by one person.
Most small advertisers write an ad, launch it, and only find out it was wrong after they have already paid for the clicks. The only way to test a message is to spend real money and watch what happens. That is slow, expensive, and it teaches you almost nothing about why an ad failed.
CouncilAds moves that feedback to before the spend. Instead of guessing, you hear from twelve stand-ins for your actual buyers, each reacting the way a specific kind of customer would, and you find out which message works — and why — while it is still free to change.
The heart of it is the council. Twelve distinct AI personas, each modeled on a different type of buyer with their own priorities, doubts, and triggers, read your ad the way a real person would and react to it honestly. One might tell you the offer is unclear. Another might say the headline promised something the ad never delivered. You do not get a vague score — you get named reactions with reasons.
Around that engine sits a full product: the app writes two competing ads from different angles so you are comparing real options, runs both past the council, and returns the winner plus the reasoning behind it. It reviews the landing page the ad sends people to and flags where visitors would hesitate or drop off. And it produces a step-by-step guide for actually setting the campaign up. A user goes from a rough idea to a tested, ready-to-run ad in minutes.
Using AI to model how real customers react is not a passing idea — it is a fast-growing field, studied under names like synthetic audiences, synthetic respondents, and generative agents. The strongest evidence so far comes from a 2024 study by researchers at Stanford, Northwestern, and Google DeepMind. They interviewed 1,052 real Americans for two hours each, built an AI agent from each interview, and found the agents reproduced those people's own survey answers about 85 percent as consistently as the people themselves did when retested two weeks later (Park et al., "Generative Agent Simulations of 1,000 People," 2024).
The research is also honest about the limits — and that honesty is the point. Modeled audiences are strong at directional signal: which message lands, what objections come up, which version is weakest. That is precisely how CouncilAds is built to be used: a fast, low-cost read on your ad before you spend, so you carry the strongest version to real customers instead of paying to find out the weak one bombed.
That is what the council does. Instead of one generic AI opinion, it runs your ad past twelve distinct modeled buyers, so you get the range of reactions a real market would give you — not a single averaged guess.
This is not a mockup. Every piece runs against real, production services:
Payments are live through Stripe and were validated end to end with real transactions, not test mode. Sign-in uses real Google authentication, which I moved onto a dedicated domain (auth.councilads.com) after a tester told me the standard flow looked untrustworthy — fixing a real barrier that was costing signups. User accounts and data run on a proper database backend.
There was also a serious, invisible problem to solve: the site began as a JavaScript app that search engines could not read, meaning it was effectively invisible to Google. I rebuilt how the pages are served so search engines see full content, added automatic sitemap generation, and set it up to notify search engines the instant new content publishes. The site went from unreadable to fully indexable.
I built CouncilAds using an AI-accelerated workflow: I designed the product, made the architecture decisions, and directed the build in plain language, with AI writing and iterating the code under my direction. That is the same approach I bring to every WorkLeapAI project, and it is why I can deliver real, working software in a fraction of the usual time and cost.
Fast does not mean sloppy. The rule throughout was simple: nothing counts as done until it is proven live. Payments were tested with real money and refunded. The sign-in flow was checked against Google's real servers. Search-engine indexing was confirmed on the live site, not assumed from the code. That habit — verify instead of assume — is exactly why the product holds up under real use.
CouncilAds is proof of what I can build. If I can design and ship a live product that handles payments, secure sign-in, an AI decision engine, and real search visibility on my own, then the AI tool or automation your business needs is well within reach.
That might be an assistant that answers your customers around the clock, a system that takes the repetitive work off your plate, or a custom tool built around how your business actually runs. The method is the same every time: build the real thing, test it against reality, and put it to work.
Want to see what that could look like for your business? Take the free 3-minute assessment, or get in touch through the contact page.