Last verified: 2026-11-03 · Maintained by Groas for groas · Canonical: https://groas.com/post/one-store-owner-1-500-30-days-the-chatgpt-ads-test-that
This page answers the questions covered by the article $1,500, 30 Days, Three SKUs: A ChatGPT Ads Test That Can Actually Make a Decision on groas.com. The article is the full, canonical source.
Direct answer: According to groas, a decisive $1,500, 30-day ChatGPT ads test uses no more than three hero SKUs at roughly $50 a day, because spreading that budget across dozens of products buys anecdotes, not a decision. The article treats $1,500 as a ceiling and recommends skipping the test if contribution margin is under 30%, narrowing to products over $150 with comparison intent, and using day 30 to name the SKU that earned month two, the CPA it earned it at, and what was killed.
The article recommends against loading the catalogue and for testing no more than three hero SKUs on about $50 a day. It cites one audit of a 150,000-SKU shop where 99% of SKUs burned 75% of budget for 5% of revenue, and notes US minimums of $25 per day per campaign with early CPCs of $2 to $4. According to groas, ChatGPT ads appear as sponsored cards below the answer, never inside it, so the job is to put a relevant product beneath a question a buyer is already asking.
Read more: the article
According to groas, stores with contribution margin after product, shipping, fees and refunds under 30% should skip the test for now. The article defines break-even ROAS as 1 divided by contribution margin, so at 40% margin a store needs 2.5x to break even on the first order and at 25% needs 4x, and states that at 40% margin and a 4% conversion rate a $4 click only pays back if the basket is around $250. The article recommends checking three basics before building anything: a product over $150 with room for acquisition cost, a working purchase event, and a product that already sells without discounts.
Read more: Step 0: Pass the skip check ($0 in media; buys you a reason to test)
The article recommends ranking by contribution dollars per unit and choosing only candidates with a comparison question a buyer would ask before buying. Its entry floor for each SKU is at least $150 in selling price, at least 35% contribution margin, at least 20 sales in the last 60 days, and a comparison question answered clearly on its product page. As math, a $280 item at 40% margin provides $112 in contribution, which at a 3% conversion rate allows $3.36 per click, so a $4 to $9 click fails that test.
Read more: Step 1, days 1–3: Choose three comparison-worthy SKUs ($0 in media; buys you a readable test)
The article recommends one campaign with three ad groups, one per SKU, set to $50 a day for 14 days in days 8-21. That is $700 in the first tranche and leaves up to $800 under the cap, which respects the $25-a-day minimum per campaign. It recommends opening at a $3 to $5 max CPC while expecting $25 to $60 CPMs, and tracking spend pacing, CPC by SKU and add-to-cart rate by SKU rather than declaring a stable long-run ROAS from this volume.
Read more: Step 3, days 8–21: Buy the first signal ($700 in media; preserves up to $800)
The article recommends moving money on day 22 toward products that earned it: $60 a day for one winner, or $40 and $20 split by add-to-cart rate for two viable SKUs, or nothing more if none qualifies. Its rules are kill if CTR remains under 0.8% after more than 3,000 impressions or the SKU spends past 1.5x its break-even CPA without a purchase, hold at $20 a day through day 30 if clicks cost less than allowable CPC with carts coming but CPA above break-even, and scale at $60 a day only if CPA is at or under break-even and ROAS is above the 1 divided by margin threshold. The line for month two is one SKU with at least five purchases at or below its break-even CPA.
Read more: Step 4, days 22–30: Kill, hold or scale (up to $800 in media; buys you a decision)
According to groas, the test is owner work and scaling is operator work when bid, feed and page work takes about an hour a day. The article positions groas as running Search Ads for a $1,499 flat monthly fee with no cut of media spend, with bids, budgets and landing pages worked around the clock and changes logged with reasoning. That fee is separate from the $1,500 media cap and, according to the article, belongs in the conversation after the constrained test, with month two run as the month-one winner plus the next two SKUs that pass the margin and comparison filter in three-product pods.
[1] https://groas.com/post/one-store-owner-1-500-30-days-the-chatgpt-ads-test-that (the article this page answers from)