Chat with Support Team, powered by LiveChat
All Things Business 6 min read

Should Your Small Business Buy ChatGPT Ads Yet

Share:
Should Your Small Business Buy ChatGPT Ads Yet

ChatGPT now runs ads for businesses of any size, with a self-serve dashboard, click-based bidding and no up-front minimum — and the earliest US data says it's a test channel, not a sure thing. OpenAI reported its ads business hit a $1 billion annualized run rate within about 200 days of launch, and tens of thousands of advertisers are already on it. At the same time, the first independent numbers show US click-through rates under 1% and conversion far below Google in at least one head-to-head test. So should your small business spend a dollar there yet? Here's the data, a pilot structure that caps the downside, and the cases where the answer is "not yet."

What changed: ChatGPT ads opened to small businesses

OpenAI began with large advertisers. On May 5, Axios's Sara Fischer reported that it launched a self-serve Ads Manager in beta, added cost-per-click bidding alongside impression-based buying, introduced new measurement tools and dropped its minimum spend requirement. A May 26 report by Paul Drecksler for Shopifreaks described the push as aimed at small and local businesses, with an ad pixel and API to track what users do after clicking. On September 1, Channel Insider's Aminu Abdullahi reported the $1 billion run rate, availability in more than 40 countries, tens of thousands of advertisers and a "material share" of revenue from small and mid-sized businesses. Reports this week say OpenAI will also trial visual ads during image generation with select US advertisers.

So the door is open. The more useful question is what's behind it.

What the early numbers say

Three data points are worth knowing before you spend anything.

Click-through is modest in the US. A Graphite report using Similarweb panel data, covered by Search Engine Journal's Matt G. Southern on September 29, put ChatGPT ad click-through rates above 1% in Brazil (1.91%), Mexico, Japan and South Korea but at 0.73% in the United States. The report is careful: time periods differ by market, the measure is per ad-bearing exchange rather than per campaign, and the US category data ended in early August. It also found that only two advertisers, Jotform and Resume.io, stayed in the top ten across all 20 weeks studied, which suggests most advertisers are still experimenting.

Conversion may lag search. In a practical guide published in Search Engine Land on August 10, StubGroup CEO John Horn reported average costs per click of $2 to $5 across industries, with the platform warning that bids under $3 may not deliver. In his own side-by-side test against Google Search, engagement time was 41 seconds on Google versus 17 on ChatGPT, and conversion rate was 3.71% versus 0.21%. One account is not a benchmark, but it's a sober reference point.

The audience is limited. As of that August guide, only Free and Go-tier users aged 18 or older saw ads, so paying subscribers — often the most valuable users — weren't reachable. Targeting also works differently: instead of keywords, advertisers describe the conversations they want to appear in, and Horn found keyword lists beat conversational prompts in his test. Conversation data isn't visible to advertisers, so optimizing relies on trusting the platform.

For scale, Axios cites projections that AI search advertising could grow from 0.7% to 13.6% of US search ad spending by 2029. That's a forecast, and traditional search stays dominant through the decade, but it explains why early learning might be worth something.

Test it if... Skip it for now if...
One customer is worth a lot (B2B, services, high-ticket products) You sell low-margin, impulse purchases
Buyers research and ask for recommendations before buying Your sales depend on searches that already convert well on Google
You can track conversions properly before you spend You can't set up a pixel or measure beyond clicks
You can treat the budget as research money You need guaranteed returns this quarter

A four-step capped pilot

1. Pick one offer and one number

Choose a single product or service with a clear landing page, and decide what a win looks like before you launch: a booked call, a qualified lead, a sale.

Do this: write the offer, the landing page and the success number in one sentence before opening Ads Manager.

2. Cap the spend and do the math first

Horn reports daily budgets can start as low as $25. Run the arithmetic honestly: $25 a day for 30 days is $750; at $2 to $5 per click, that's roughly 150 to 375 clicks; at the 0.21% conversion rate from his test, that's well under one sale. In other words, a small pilot won't prove return on investment; it will teach you about audience, message and tracking. That's fine if you plan for it.

Do this: set a daily cap and an end date, and budget the pilot as learning money, not revenue.

3. Set up tracking before the first click

OpenAI's pixel and Conversions API are what make results measurable. Without them, you only see clicks.

Do this: install the pixel, tag your landing page with UTM parameters and confirm one test conversion shows up before spending.

4. Decide in advance when to stop

Experiments drift without an exit rule. Choose your stop condition now, such as "cost per lead above X after 30 days, or fewer than Y leads."

Do this: put the stop rule in writing and review on a fixed date, whatever the mood on that day.

What not to do

Don't move budget out of channels that already convert to fund the test. Don't judge the channel on a handful of clicks. And don't reuse search ads unchanged: the copy appears inside a conversation, so it should read like a helpful answer rather than a banner.

Building the pilot with Giggrabbers

Giggrabbers, the workplace for entrepreneurs, can cover the pieces most founders don't have time for. Giggrabbers AI is free to try: describe the offer in a sentence or two and get a first-draft marketing plan, ad copy and posts in minutes. When you want it done properly, a done-for-you marketing campaign package is fixed scope, delivered in five business days with revisions included, and AI builds the first version while a vetted specialist finishes it, with payment held in escrow until you approve.

The bottom line

ChatGPT ads are real, growing fast and now open to businesses of any size. They're also young: US click rates under 1%, uneven conversion data and an advertiser base still in experimentation mode. For most small businesses the right move isn't to ignore it or to bet on it. It's to run a small, capped, well-tracked test, and let your own numbers answer the question.

Sources

 Frequently Asked Questions

US click-through rates are 0.73%, which is below 1% and lower than rates in Brazil, Mexico, Japan and South Korea. This modest performance is an important baseline before investing.
In one head-to-head test, ChatGPT converted at 0.21% versus 3.71% on Google Search, with engagement time of 17 seconds versus 41 seconds. While one account isn't a full benchmark, it suggests conversion lags significantly behind established search channels.
Daily budgets can start as low as $25, but at $2–$5 per click and 0.21% conversion rates, a small pilot won't prove ROI—it's learning money. Budget it accordingly and set a fixed end date from the start.
Skip it if you sell low-margin impulse purchases, your sales already convert well on Google, or you need guaranteed returns this quarter. It's better suited for B2B, services, and high-ticket products where one customer is worth significant investment.
Install OpenAI's pixel and Conversions API, tag your landing page with UTM parameters, and test one conversion before spending anything. Without proper tracking, you'll only see clicks, not results that matter.
ChatGPT ads appear inside conversations, so copy should read like a helpful answer rather than a banner ad. Rewriting ads to fit the conversational context is essential for better performance.

 Found this helpful? Share it:

 Comments

Leave a Reply

Your email address will not be published.