How to Run ChatGPT Ads Effectively, and What the $1,200 Burn Post Got Wrong

The claim this article answers is not hypothetical. In the widely shared Polymarket post “Small Business Owner Trashes ChatGPT Ads as Unprofitable”, an advertiser says that after spending more than $1,000 on ChatGPT Ads, you would be better off “lighting money on fire.” Polymarket turns that first-person verdict into a broader account of the channel: the platform reported clicks, the advertiser could not verify useful traffic or leads, and money went in without proof of customers coming out.

That experience may be entirely real. But the post does not publish the campaign details needed to diagnose it: objective, dates, impressions, click count, targeting, creative, landing page, conversion definition, or attribution setup. Its evidence supports “this advertiser did not get a return they could verify.” It does not support “ChatGPT Ads are a bad investment.” That jump—from one undisclosed result to a verdict on the channel—is the point we are arguing against.

The best documented public version of the same skeptical case is a separate test from the Google Ads agency Grow My Ads. It reports roughly $1,200 in spend, 92 clicks, an average CPC of $13, an average CPM of $120, a click-through rate near 1%, and zero measured conversions, then recommends that most advertisers wait and treat the channel as “mostly a burn-budget experiment”. DailyTopAI repeats those numbers and conclusion. This is not proof that Grow My Ads ran the campaign in Polymarket’s story; it is the disclosed test that lets us evaluate the burn-budget argument against actual numbers.

The numbers are not the problem. Grow My Ads reported its data honestly, and its inability to verify a return is real. The problem is the broader conclusion drawn from that result. A ~1% click-through rate is close to one dated vendor estimate for this surface, not evidence of a broken auction. A $3 bid is a recommended starting point, not a price floor. Limited reporting and broken site attribution leave a measurement gap, not proof of missing performance. And an audience restricted to Free and Go users is a fit question, not a universal disqualifier.

None of this makes ChatGPT Ads a proven channel. OpenAI says it has not published performance benchmarks across advertisers, industries, or campaign types. The case here is narrower: Polymarket’s anecdote cannot support a channel-wide verdict, and the documented Grow My Ads test could not reconcile its reported zero with GA4 after its UTM attribution broke.

At a glance

• The viral claim is an anecdote, not a disclosed test. Polymarket gives us a spend threshold and an advertiser’s verdict, but none of the campaign inputs or outcome data needed to diagnose the result.
• The separate documented test is real and small. Grow My Ads covers one advertiser, one market, roughly two weeks, about $1,200, and 92 clicks. It is evidence about that campaign, not a benchmark for the channel.
• Its ~1% CTR is not a stand-alone failure. One dated compilation reports about 0.9% ChatGPT ad CTR (eMarketer 2026, as compiled by Tru Commerce); OpenAI has not published a cross-advertiser benchmark. The ~6.4% comparison is Google search, a different placement.
• $3 is a recommended starting bid, not a floor. Reported ranges run $3–$5 per click for consumer categories and $8–$18 for software and finance. The $13 the test settled at sits inside the software band.
• The reported zero lacked an independent check. The test’s own UTM bug broke its GA4 attribution. That leaves a platform-reported zero the advertiser could not reconcile against clean site analytics.
• Low CTR is not a verdict. A click-through rate alone cannot show whether traffic converted or produced a return.

Figure 1. ChatGPT ad CTR against two reported comparison numbers. The ~1% result in the cited test sits near one vendor-compiled ~0.9% estimate; the ~6.4% comparison is Google search, a different placement. Sources: eMarketer 2026 and Tru Commerce, both dated August 2026. OpenAI has not published a cross-advertiser benchmark, and these CTRs say nothing about conversion or revenue.

What the viral article establishes—and what it does not

Polymarket establishes a useful fact: at least one advertiser crossed $1,000 in spend and regarded the result as a terrible investment. It also makes a legitimate measurement complaint. If an ad platform reports clicks that an advertiser cannot reconcile with site traffic or leads, the advertiser should not keep spending on faith.

What it does not establish is why the campaign failed or whether the failure generalizes. The post supplies no denominator for its reported clicks, no conversion definition, no account configuration, and no independent traffic record. “We could not verify a return” is enough to stop one campaign. It is not enough to tell every advertiser that the channel is equivalent to burning cash.

The distinction matters because the strongest counterargument is not that this advertiser secretly succeeded. We have no evidence of that. It is that a personal stop decision and a market-wide performance verdict are different claims, and only the first is supported by the article.

What the documented $1,200 test actually measured

The Grow My Ads test gives the burn-budget argument numbers that can be examined. It covers one advertiser, one market, one objective, and about two weeks. Its unit is a click; its denominator is the impressions OpenAI served; its outcome field is a single conversion event on a page whose analytics had broken.

That last detail matters most. Grow My Ads reports a UTM bug that broke its GA4 attribution and caused ChatGPT traffic to appear as Direct. That does not prove the platform’s reported zero was wrong, but it removed the independent analytics check that could have confirmed it. The disclosed result is therefore zero reported conversions with broken off-platform attribution—not a zero reconciled across two working measurement systems.

Field	Grow My Ads test	What it can support
Spend	~$1,200 over ~2 weeks	Order-of-magnitude budget sizing
Clicks	92	A small-sample CTR estimate
CTR	~1%	Comparison to a like-for-like benchmark
Avg CPC	~$13	Comparison to category cost ranges
Avg CPM	~$120	Comparison to reported CPM ranges
Conversions	0 reported; GA4 attribution broken	A platform-reported zero that was not independently reconciled
Breakdowns	None by query, theme, or placement	Nothing about match quality

This is the missing-state problem in miniature: “the platform measured zero” and “we independently verified zero” are different states. The review had the first and lost the second, but its conclusion reads as though it had both.

Figure 2. The Grow My Ads test reported 92 clicks and zero conversions, but its broken GA4 attribution prevented an independent site-side check of that zero. The missing check does not prove that any conversion occurred or that the platform-reported zero was wrong.

Is a 1% click-through rate proof that the channel is broken?

No. It is close to one published vendor estimate, not an established platform average. The dated compilation reports about 0.9% ChatGPT ad CTR and says its own US campaigns averaged about 1.3% since July 2026. A separate vendor creative sample put product ads in a 4–7% band. OpenAI says cross-advertiser performance benchmarks are not available, so none of these figures should be treated as a universal target.

The comparison that makes 1% look catastrophic is ~6.4%, the CTR for Google search ads. That is a reported number, but not the same unit: a Google results page can carry several competing text ads, while ChatGPT ads appear below responses and can be matched to the conversation rather than a typed query. The honest question is not “is CTR above 6.4%?” but “does the traffic convert, and at what cost?”

The most-quoted conversion number in this debate—that ChatGPT-referred ecommerce traffic converts at 15.9% against 1.76% for Google organic—is Adobe’s 2025 referral data. It is organic referral, not an ad-click rate; using it to defend ad performance repeats the category error of using 6.4% to attack it. No credible ad-side conversion benchmark is public.

Is there a $3 CPC floor?

No. $3 is a starting recommendation, not a wall. The Grow My Ads review raised its max CPC until the interface flipped from “May not deliver” to “Strong delivery,” then settled near $13 and framed that as a floor. But the interface labels bid strength, not a market minimum. OpenAI recommends a starting maximum CPC of $3–$5 for Clicks campaigns, and third-party ranges put consumer categories at $3–$5 and software and finance at $8–$18. The $13 average is within the reported software and finance band; the ~$120 CPM is above a separate reported range, though neither comparison establishes an optimal bid.

Figure 3. Reported cost ranges by category against the cited test’s settled average CPC. The $13 average sits inside the $8–$18 software and finance range, not above a ceiling. Sources: WebFX and Tru Commerce benchmarks, August 2026. What this cannot establish: a CPC band says nothing about whether the traffic converts.

The mechanism matters more than the number. OpenAI describes a relevance-weighted, second-price auction. It offers impression billing for Views, valid-click billing for Clicks, and conversion optimization with either click billing or conversion-optimized impression billing for eligible advertisers. The $3–$5 starting maximum CPC is guidance for Clicks campaigns, not a fixed price floor or a recommendation for conversion-optimized bidding.

Why did the ad manager look empty, and what does that prove?

The manager reports impressions, clicks, spend, CTR, average CPC, average CPM, and conversions when measurement is configured. OpenAI’s current reporting guide also documents device, country, and platform segments, individual conversion-event columns, and click-through and eligible view-through attribution. It does not document a query-level breakdown. Spend insights may lag 7–8 hours, and attributed conversions can take 24–48 hours to appear; a recent zero is not necessarily a final result.

The measurement gap is where an advertiser should spend effort. OpenAI documents a measurement pixel and Conversions API, including shared event IDs to deduplicate events sent through both. It also supports static UTM parameters and dynamic URL parameters. The Grow My Ads test hit a related problem: its UTM broke, so its GA4 data could not independently confirm whether the platform’s reported zero represented the traffic that arrived.

Does the Free and Go audience make ChatGPT Ads useless?

For B2B software, the audience restriction matters. OpenAI says ads may appear for adults on Free and Go, while Plus, Pro, Business, Enterprise, and Edu accounts are ad-free. That excludes those paid seats, but Free and Go users can still make business decisions; the plan restriction alone does not establish B2B campaign performance.

The compiled research finds roughly one in ten conversations carries commercial intent, and that in 12 of 14 industries the market leader ran zero ChatGPT ads in the sampled period; intermediaries, comparison sites, and affiliates held some slots. Those observations describe that sample. They do not establish which advertisers will succeed or why a brand chose not to buy ads.

What actually decides whether your ad gets matched?

OpenAI says advertisers can add ad-group context hints describing conversations where their products may be relevant. These are not exact-match keywords and do not guarantee delivery in specific conversations. The landing page, creative, and other signals also inform selection, so no single hint can guarantee an impression.

The vendor finding is about creative, not hints. In Criteo’s published sample, impressions varied roughly fourfold between scenario-specific creatives and generic ones, while every creative’s CTR stayed inside the same 4–7% band—specificity coincided with more matches, while click rates stayed in a similar band. In a system that matches conversations rather than exact keywords, creative and hints can affect eligibility. Thin impression volume can also reflect bids, budget, review status, or account setup; diagnose those causes before changing a bid.

Figure 4. Creative specificity against match frequency and click rate, from Criteo’s published sample of about 3,850 product-ad displays. The more specific creative appeared more often in that sample; the click band barely moved. Source: Criteo via Tru Commerce, September 2026. This is a vendor sample, not an account-level benchmark, and it cannot establish conversion differences between creatives.

How to run a test that can support a verdict

Set a budget and stop condition before spending. Instrument first: use the measurement pixel and, where appropriate, the server-side Conversions API, with a shared event ID for deduplication. Land the click on a page that answers the conversation’s actual question, not a homepage. Write context hints as decision moments (“someone choosing a business bank account with no monthly fee”) rather than topics (“banking”), and compare ad groups with a generic control. Do not judge success against a single CTR estimate or the ~6.4% Google search number. Diagnose delivery and measure conversions on enough volume to distinguish an observed zero from an unmeasured outcome.

That sequence produces a real answer either way, and it is the honest version of “wait”: decide from an instrument that works — the discipline behind our paid growth work.

What this does not show

This article cannot establish whether ChatGPT Ads will convert for your category: OpenAI has not published cross-advertiser performance benchmarks. It cannot tell you that the advertiser in Polymarket’s story or the Grow My Ads test would have succeeded with better instrumentation. It can only show that one undisclosed anecdote cannot support a channel-wide claim and that the documented test’s zero lacked clean off-platform reconciliation. It cannot generalize one advertiser’s two weeks to your category, and it does not claim the vendor-compiled ~0.9% estimate will hold as delivery matures.

Frequently asked questions

Is a 1% click-through rate bad for ChatGPT Ads?

No. One dated vendor compilation reports about 0.9% CTR for ChatGPT ads, but OpenAI has not published a cross-advertiser benchmark. The ~6.4% comparison is Google search, a different placement that matches typed queries rather than conversations.

Is there a $3 CPC floor on ChatGPT Ads?

No. $3 is OpenAI’s recommended starting max bid, not a hard minimum. Settled cost per click is set by a relevance-weighted second-price auction; reported ranges run $3–$5 for consumer categories and $8–$18 for software and finance.

Do ChatGPT Ads work for B2B?

Ads may appear to adults on Free and Go, while Plus, Pro, Business, Enterprise, and Edu accounts are ad-free. That excludes those paid seats, but it does not tell us whether a particular B2B campaign will work. Compare the actual reachable audience and measured outcomes with your objective.

Why does the ChatGPT Ads manager show so little?

The beta reports delivery, spend, and configured conversions. It also supports device, country, and platform segments, individual conversion-event columns, and click-through and eligible view-through reporting. Spend may lag 7–8 hours and conversions 24–48 hours; the published guide does not document query-level reporting.

Should you wait before advertising on ChatGPT?

Waiting is a legitimate choice, but the public evidence does not force it. A small, correctly instrumented test—pixel, server-side conversions API, deduplicated events, and a declared stop condition—answers the question faster than waiting for a benchmark the industry has not published.

---

Method and source disclosure. This article synthesizes public sources. The “lighting money on fire” claim and Polymarket’s interpretation are attributed to its linked article; Polymarket does not publish the underlying campaign data, and we do not treat the separate Grow My Ads test as that campaign. The external CTR, cost, and creative figures come from Tru Commerce’s August 2026 compilation and the sources it cites (eMarketer 2026, WebFX, Criteo), archived at fetch time. They are vendor estimates or samples, not OpenAI-wide benchmarks. Current audience, pricing, reporting, and measurement facts are linked to OpenAI’s Help Center above. Where a primary source could not be retrieved, its claims remain attributed through the secondary source that reports them. A previously shown Violet campaign analysis was removed because the original export and a redacted origin record could not be authenticated from retained evidence. Spot a number that does not trace to its source? Tell us and we will correct it.