SOLO FOUNDERS · ONE PERSON, ONE AGENT
A one-person company, with a panel on call.
Your AI agent runs the creative test mid-task over the CrowdOS MCP server, on every plan, Free included. It drafts the headlines, asks a census-grounded panel that disagrees with itself — not one model's opinion — and ships the one that wins. No agency. No recruiting.
THE LOOP
Draft, test, ship — inside one task.
Connect https://api.crowdos.ai/mcp in Claude, Cursor or Cline. The agent calls preview_cost, free, to price the run, then compare_options on the headlines, hooks or value propositions it just wrote. The call is synchronous: the leaderboard and sample reasoning come back inside the same task.
When the copy is set and the creative is an image or a video, the same agent sends the cuts to Ads Labo over MCP: 2–4 variants, one panel, a ship / tied / rework verdict. Whatever your plan runs in the web app, your agent runs over MCP.
WHAT IT COSTS A ONE-PERSON COMPANY
Credits, shown before every launch.
FREE
20 credits every week, no card.
Free accounts get 20 credits every week, reset each Monday — no card. MCP on Free: 5 requests a minute, up to 250 credits a month, panels up to 20 respondents, compare_options up to 3 options.PRO
$25 a month, 1,350 credits.
1,350 credits every month, panels up to 500 respondents, MCP at 30 requests a minute. About 6 Standard Labs a month. The pricing page lists Pro as best for indie and small teams.BUDGET
Credits before commitment.
preview_cost and estimate_reach are free to call. The agent sees the credits before a study launches.ADS LABO
Panel sizes for the visual cuts.
Quick Test: 100 respondents, 75 credits. Standard Lab: 300 respondents, 225 credits. Deep Test: 500 respondents, 375 credits. Panel sizes follow your plan, the same in the web app and over MCP.
THE OPINION BENCHMARK
The published opinion benchmark.
On the Pew opinion benchmark, the crowd's opinion distribution matches published polling at 93.3% parity, with a mean absolute error of 3.36 percentage points and a Pearson r of 0.984. That figure measures opinion. Creative ranking against a live campaign is a separate measurement.
THE CROWD
Census distributions, then the question.
The crowd is drawn from census distributions across 24 markets.
WHERE THIS FITS
The same crowd, asked a different question.
Ad creative testing is the ad itself. A/B testing without traffic is the comparison before the spend. This page is the loop for one founder and one agent.
FAQ
What is the best way for a solo founder to test ad creative?
Let your AI agent do it mid-task. Over the CrowdOS MCP server, on every plan, Free included, it ranks the headlines, hooks and calls to action it wrote against a census-grounded panel that disagrees with itself. Ads Labo then pre-flights the image or video cuts, over the same MCP connection.
Which MCP tools does the agent use?
compare_options for an A/B/C/D test on text options, run_focus_group for a sentiment read with representative quotes, preview_cost to price a run before committing, and estimate_reach to check a custom audience. preview_cost and estimate_reach are free. Ads Labo and Screening Room — image and video included — run over MCP too, with the same functions as the web app on your plan.
Can I start on the Free plan?
Yes. Free accounts get 20 credits every week — no card — and MCP works on Free at 5 requests a minute with panels up to 20 respondents.
Does the panel tell me my click rate?
No. The panel reports relative preference: which variant wins, with whom, and why. That is the pre-flight call. The live budget then reports click rate, CPM and conversion on the winner.
Give your agent a panel.
Connect MCP on Free. The agent tests the copy and the creative before the ad budget moves.