ChatGPT Ads Get 0.73% US CTR, Ecommerce Mostly Absent

A new Graphite report says ChatGPT ads earned a 0.73% click-through rate in the US, while retail and ecommerce advertisers have been largely absent from the platform.

If you run a high-ticket store through Ecommerce Paradise methods or your own, this is the first broad dataset on whether ChatGPT belongs in your Q4 ad budget. The answer from the data: the lane is open, the measurement is thin, and the top advertisers so far sell software, not $3,000 patio sets.

Below: what the report found, how ChatGPT ads got here, what the numbers mean for a store built on the high-ticket dropshipping model, and a capped test you can run before the holiday rush.

New ad channels change fast, so keep the paperwork side boring. Northwest Registered Agent has been doing this work for about 25 years, so your formation paperwork is one thing you never have to retest. See Northwest Registered Agent →

ChatGPT Ads Post a 0.73% US CTR, per Graphite

Graphite published “The State of ChatGPT Ads” this week, built on Similarweb panel data, according to Search Engine Journal’s September 29 write-up. The US click-through rate came in at 0.73% through mid-September. Four markets beat 1%: Brazil at 1.91%, Mexico at 1.55%, Japan at 1.53% and South Korea at 1.21%. Canada (0.75%), the UK (0.74%), Australia (0.67%) and New Zealand (0.60%) all sat below 1%, like the US.

One caveat from the report itself: the data covers between 5 and 25 weeks per market, and CTR is measured as clicks per ad-bearing exchange, meaning one user message plus ChatGPT’s response. That is not the same as clicks per impression on Google Search, so do not compare the two numbers directly.

The category mix matters more for you than the CTR. Per the same report, B2B SaaS made up 19.9% of US impressions, and retail and ecommerce advertisers “have been largely absent from the platform.” Search Engine Journal says Graphite attributes that to product feed compatibility issues with ChatGPT’s chat format. Only Jotform and Resume.io ranked in the top 10 advertisers across all 20 weeks analyzed, which tells you how much the leaderboard churns week to week.

A companion report from Search Engine Land, also dated September 29, describes the research base as more than 98,000 ads and 8,000 landing pages tracked between April 1 and August 16, 2026. It reports that 43% of ad-eligible US ChatGPT users have seen an ad within six months, that click-through rates have doubled since launch, and that the platform has attracted tens of thousands of advertisers.

The researchers also list what the best performers do. They write about 98% new copy for ChatGPT instead of recycling ads from other platforms, build messaging around the use cases people discuss in their conversations, keep landing pages to five or fewer, and see the strongest results from ads shown in the first turns of a conversation. Their advice, per Search Engine Land: treat ChatGPT as experimental spend, not a core acquisition channel.

Both sets of numbers can be true at once. Click-through rates can double from a low launch base and still sit at 0.73% in the US. A doubling tells you the ads are improving. It does not tell you the channel has caught up to Google Search, where a shopper typing a model number is usually ready to buy.

Pricing is still a wide spread. In an analysis published September 8, Brooke Osmundson, Director of Growth Marketing at Smith Micro Software, reported cost per click ranging from $3 to $22 across advertisers and markets. Common Thread Collective reported a $4.41 average CPC on $9,620 in spend, with estimated return on ad spend between 3.3x and 6.8x. Synter reported a $9.89 average CPC overall, from $5.10 in the UK to $22.89 in New Zealand. Osmundson notes OpenAI has not published benchmarks by industry or campaign type, and the $3 to $5 starting bid it suggests is bidding guidance, not a performance benchmark.

From February Pilot to $1B Run Rate: How ChatGPT Ads Got Here

OpenAI began piloting ads on February 9, 2026, then expanded them to all US users on the free and Go tiers, per EcomCrew’s March 24 coverage. Plus, Pro, Team and Enterprise subscribers do not see ads. EcomCrew reported at the time that early advertisers faced a $200,000 upfront commitment and could not yet show measurable business outcomes, and it cited estimates that only 2% to 3% of ChatGPT queries are commercial, against 10% to 20% for Google Search. Those March figures describe the launch phase, and the platform has moved since.

By August 17, PPC Land reported that Similarweb’s new AI Ads dataset showed about 26% of ChatGPT replies carrying sponsored content. Harel Amir, who oversees Similarweb’s advertising intelligence division, called it “the rare moment when a major ad channel is still wide open.” PPC Land also noted that no public ad library exists for ChatGPT, AI Mode or AI Overviews, so panel data is the only outside view.

The revenue story is more mixed. AdExchanger reported on September 3 that OpenAI’s ad business hit a $1 billion annualized run rate, which is short of the $2.5 billion the company had projected for 2026. Agency executives told AdExchanger why budgets are not moving faster. Victor Batista of Jellyfish said advertisers want tangible metrics and competitor benchmarks. Rajeev Nair of Lifesight called the conversion tracking and audience options “very rudimentary” for enterprise use. Scott Ensign of Butler/Till said agencies have no way to manage multiple advertisers in one dashboard.

There is a counterpoint to the gloom. AdExchanger notes OpenAI has since introduced a conversion API and conversion-optimized campaigns, and it cites evidence that ChatGPT queries drive long-term traffic and indirect conversions. Measurement vendors, according to the same story, need roughly six months of data to model cross-channel lift, so the honest read on ROI is still a few months away.

This is not the first AI-commerce shift I have tracked on the site. I broke down Shopify’s move to let AI agents submit orders on the checkout side. The organic side is in my post on answer engine optimization for AI shopping discovery, linked in the related reading below. Paid placement inside ChatGPT is the third piece, and it is the one with a price tag you control.

Why Ecommerce’s Absence From ChatGPT Ads Is Your Opening

My read is that the absence of retail advertisers is the most useful fact in this report. When B2B software fills a fifth of US impressions and stores fill almost none, the shopper-intent auctions are thin. Thin auctions usually mean cheaper clicks for whoever shows up first with a usable offer.

The feed problem is real, though, and it explains the gap. Google Shopping runs on structured product data: title, price, GTIN, availability, image. A chat window has no product grid. A high-ticket seller who has spent years perfecting a Merchant Center feed has no obvious way to move that asset into a conversational ad. The retailers who did try appear to have hit that wall.

There is also a targeting question nobody has answered yet. AdExchanger’s sources describe “context hint” targeting that gives advertisers little beyond keywords to avoid, and Google’s Auction Insights has no equivalent here, per the September 8 analysis. So you cannot see who you are bidding against or how often you lose. For a store owner used to reading impression share on Google, that blindness is the real cost of being early.

Here is the first-order math, and it is hypothetical, not reported. Say you sell a $3,000 outdoor set at a 25% gross margin, so $750 of gross profit per order. At the $4.41 CPC Common Thread Collective reported, you can afford about 170 clicks per sale, which means a 0.59% click-to-order rate breaks even. At the $9.89 Synter average, the break-even is about 76 clicks and a 1.3% rate. At $22, it is about 34 clicks and a 2.9% rate. For a $3,000 product with a considered purchase cycle, only the cheapest end of that range looks plausible on last-click math alone.

That is why I would treat ChatGPT as a research-stage channel, not a checkout-stage one. On Google, the shopper already knows the product. In a chat window, the shopper is still working out what to buy, and your ad is one voice in that conversation. Someone asking a chatbot which outdoor sets hold up in a coastal climate is three weeks from buying, and probably calls a phone number first. If your store cannot capture that visitor, the click is wasted at any CPC. My guide on high-ticket sales follow-up covers the capture side.

Three scenarios, with thresholds. If your first $1,500 gets clicks near $4 to $5 and you can show even a handful of phone calls or quote requests, keep testing at the same cap. If clicks land above $9 with no calls, stop, because the channel is not priced for your ticket size yet. If clicks are cheap but the landing page bounces, the problem is your page, and the same fix helps your Google Shopping traffic too.

Also weigh the environment you are testing in. The Fed hike and 81.9 consumer confidence reading I covered today mean big-ticket buyers are hesitating. Meanwhile the peak freight surcharges are already eating margin. A new channel with no benchmarks is a poor place to spend money you cannot afford to lose. That argues for a small, hard cap, not for skipping the test.

The operator complexity here is the point. A store owner now juggles Google Shopping, Meta, email, SEO, AI answer engines and a paid chatbot channel, each with its own reporting gaps. If you would rather have someone run that stack for you, my team builds and manages stores through the turnkey done-for-you service.

New ad channels are easier to test when the store already converts. Want my team to build and run your high-ticket store for you? See the turnkey done-for-you service →

Run a $1,500 ChatGPT Ads Test Without Wrecking Q4

Here are five moves, in order, for a store owner who wants data instead of opinions:

Set a hard cap before you open Ads Manager. A hypothetical $1,500 over four weeks buys roughly 340 clicks at $4.41. Nothing in the reported data supports a bigger commitment yet, and Graphite’s own advice is to treat the channel as experimental.
Write ChatGPT-only copy. The report’s top performers used about 98% new copy. Use a question your buyer would actually ask, like which product size fits a specific room or use, then answer it in the ad. A tool like Claude can draft 20 variants from your product spec sheets in minutes.
Pick five or fewer landing pages. Choose your highest-intent category pages and give each one a visible phone number and a quote form. Confirm your Shopify theme loads fast on mobile, since chat users often click from a phone.
Measure outside the ad platform. OpenAI has not published benchmarks, so track calls, quote requests and orders yourself. Tag every click with UTM parameters and route the leads into Klaviyo so you can see what happens over 30 days, not 30 minutes.
Research the questions first. Before you bid, ask ChatGPT the questions your buyers would ask, then check demand with Google Trends and a keyword tool such as Semrush. Book a free call through my discovery link if you want a second set of eyes on the plan.

Review the test at day 14 and day 28, not daily. With auctions this thin and traffic this small, one big-ticket sale or one dead week will swing every average, and you will be tempted to read a trend into noise. One quarter of data from one small budget is not a trend, so decide the pass or fail rules in writing before you spend the first dollar.

Keep your existing channels funded while you test. My rundown of the best ad platforms for high-ticket dropshipping shows where the proven dollars sit today, and none of this replaces them.

Frequently Asked Questions

Are ChatGPT ads worth it for a high-ticket store right now?
Only as a capped experiment. Reported CPCs run from $3 to $22 with no published benchmarks, so a small test with phone and quote tracking beats a big commitment. My guide to high-ticket Google Ads conversion systems shows the tracking setup to copy.

Who can see ChatGPT ads?
Per EcomCrew’s March coverage, ads appear to US users on the free and Go tiers, and Plus, Pro, Team and Enterprise subscribers do not see them. Search Engine Land reports 43% of ad-eligible US users have viewed an ad within six months.

Why are ecommerce advertisers missing from ChatGPT ads?
Search Engine Journal reports Graphite attributes it to product feed compatibility issues with ChatGPT’s chat format. Store product data is built for grids and shopping tabs, not conversations.

Does a 0.73% CTR mean the ads are bad?
Not by itself. Graphite measures clicks per ad-bearing exchange, not per impression, so it is not comparable to a Google Search CTR. Use it as a channel baseline, not a verdict.

Should I stop investing in SEO and Google Shopping?
No. ChatGPT ads are a small, unproven channel, and organic visibility in AI answers is a separate job. The best AI SEO tools for ecommerce stores cover the organic side. My free high-ticket niches list helps if you are still choosing what to sell.

What if I am not ready to test new ad channels yet?
Then fix the basics first: a fast store, a phone number, follow-up and chargeback protection. My guide to chargeback prevention for high-ticket stores is a good place to start. The free beginner guide walks through the full launch.

Want to work through new ad channels with other store owners and me inside the community? Join the Skool community →

Subscribe to the YouTube channel for daily breakdowns. More breaking news coming through the day. Investing and ad-spend decisions are yours to make, and I am not a financial or legal advisor, so treat the numbers above as reported information and run your own math before you commit budget.

If this was useful, these go deeper:

Trevor Fenner

Trevor Fenner is an ecommerce entrepreneur and the founder of Ecommerce Paradise, a platform focused on helping entrepreneurs build and scale profitable high-ticket ecommerce and dropshipping businesses. With over a decade of hands-on experience, Trevor specializes in high-ticket dropshipping strategy, niche and product selection, supplier recruiting and onboarding, Google & Bing Shopping ads, ecommerce SEO, and systems-driven automation and scaling. Through Ecommerce Paradise, he provides free education via in-depth guides like How to Start High-Ticket Dropshipping, advanced training through the High-Ticket Dropshipping Masterclass, and fully done-for-you turnkey ecommerce services for entrepreneurs who want a faster, more hands-off path to growth. Trevor is known for emphasizing sustainable, real-world ecommerce models over hype-driven tactics, helping store owners build scalable, sellable, and location-independent brands.