How Many Ad Groups Should One ChatGPT Campaign Have?
By Ansh Khandelwal 8 min read
In brief
Calculate how many ChatGPT Ads ad groups your budget can support, using the daily spend each group needs to produce a readable result.
Last verified: 12 September 2026 | Version: 1.0 | Next scheduled review: 12 October 2026
Take your monthly budget, divide it by the spend one ad group needs before its numbers mean anything, and round down. At 100 USD a day that is usually two or three ad groups. At the 25 USD daily minimum it is one. At 500 USD a day it is three to five, not fifteen.
OpenAI's guidance is to keep each ad group focused on a single product, theme or intent area. It never defines focused, and it never mentions quantity. So the word does no work. What follows is an operational definition built from arithmetic instead, because the constraint that actually binds is budget, not conceptual tidiness.
Two readability floors, not one
An ad group needs different amounts of spend depending on what you are asking of it. Confusing these two is why budgets get spread across more groups than they can support.
The delivery floor answers whether the ad group serves at all, and roughly what it costs. Around 100 clicks gets you a usable read on CTR and average CPC, enough to tell a group that is working from one that is not serving.
The conversion floor answers what a conversion costs. If your click-to-conversion rate is around 2%, one conversion arrives every 50 clicks, and a cost-per-conversion figure you would act on needs somewhere near 20 conversions. That is roughly 1,000 clicks.
Both numbers are our reasoning, not OpenAI guidance. The 2% rate is an assumption you should replace with your own, since it is the input the whole calculation is most sensitive to. Halve it and the conversion floor doubles.
The arithmetic, at three budgets
Agency write-ups through 2026 report OpenAI guiding CPC bids into the three to five dollar range. That is reported rather than published as a floor. Take four dollars as the working figure and a 30 day test window.
| Daily budget | Monthly spend | Clicks at 4 USD CPC | Groups at delivery floor (100 clicks) | Groups at conversion floor (1,000 clicks) | Sensible answer |
|---|---|---|---|---|---|
| 25 USD | 750 USD | 188 | 1 | 0 | One ad group |
| 100 USD | 3,000 USD | 750 | 7 | 0 | Two or three |
| 500 USD | 15,000 USD | 3,750 | 37 | 3 | Three to five |
Three things fall out of that table, and none of them are obvious before you run the numbers.
At the platform minimum, you get one ad group. 25 USD a day buys roughly six clicks. Split that across three themes and each theme gets two clicks a day, which after a month is 60 clicks, below even the delivery floor. You will not learn which theme works. You will learn nothing, at full price.
At 100 USD a day, no ad group reaches the conversion floor in a month. This is the uncomfortable one, and it is true even with one single group. A B2B advertiser at this budget is reading delivery and CTR for the first month, and any cost-per-conversion figure they quote is built on two or three conversions. Two or three groups is defensible because the delivery floor allows it. Treating the resulting conversion numbers as decisions is not.
At 500 USD a day the delivery floor stops binding entirely. Thirty-seven groups clear it. The conversion floor allows three. This is the point where the limit stops being arithmetic and becomes the thing OpenAI's guidance was gesturing at: whether each group describes a genuinely separate buying situation.
Round down, not up
Two adjustments push the honest answer lower than the table suggests.
Budget pacing is not even across ad groups. OpenAI describes a daily budget as an average over a seven day period, with actual spend fluctuating above or below. Within a campaign, delivery concentrates where the auction finds matches, so the group you most want data on may be the one getting the least spend. Your quietest group gets less than budget divided by group count, sometimes much less.
And the test window is rarely a clean 30 days. Creative gets edited, bids get changed, someone pauses something on day nine. Every intervention restarts the clock on a readable sample without resetting the spend already committed.
So the rule: calculate the number, then subtract one. A campaign that can arithmetically support four readable ad groups should probably run three.
When to add a group rather than a hint
The corollary matters as much as the count. When volume is low, the instinct is to add hints to an existing group. That buys reach at the cost of attribution, because hints are not reported individually.
Adding a group instead keeps every result readable, but only if the budget supports it. Run the division before you build. If the answer is that you already have as many groups as your budget can read, the correct move when volume is low is not a fourth group and not a wider third one. It is a higher bid or a broader location, both of which are campaign or group-level settings you can reverse cleanly.
The reckoning: this contradicts the theme-first advice
Everything written about this topic, including our own piece on structuring an account, argues that each buying situation deserves its own ad group. The arithmetic above says that most advertisers cannot afford to act on that.
Both are true, and the resolution is not a compromise. It is a sequence. Consolidate first, at a budget that produces readable numbers. Split later, when budget growth funds it. An advertiser at 100 USD a day running one well-written ad group is in a better position than the same advertiser running four, because the first one will know something in six weeks and the second will not.
The cost of consolidating is real and should be named: you are choosing signal over attribution, and if the consolidated group underperforms you will not know which of its themes dragged. That is a worse position than perfect structure. It is a better position than no signal at all.
What we cannot tell you
- Whether a maximum number of ad groups per campaign exists. Not documented, and no advertiser has reported hitting one.
- How budget is distributed between ad groups inside a campaign. OpenAI documents pacing at campaign level and does not describe allocation across groups.
- What conversion rate to expect on this channel. No benchmarks are published by OpenAI, and no cross-advertiser data exists. The 2% used above is an assumption, not a figure.
- Whether ad groups in the same campaign suppress each other in the auction. Undocumented.
- Whether these floors hold in practice. They are derived from sampling logic, not from measured ChatGPT Ads accounts. InPromptAds has no first-party data.
Quick answers
How many ad groups should a ChatGPT Ads campaign have? As many as your budget can fund past a readable spend level, minus one. At 25 USD a day that is one group. At 100 USD a day, two or three. At 500 USD a day, three to five.
Does OpenAI publish a limit on ad groups? No. It advises keeping each ad group focused on a single product, theme or intent area, and never defines focused or states a number.
What is the minimum spend an ad group needs? Roughly 100 clicks to read delivery and CTR, roughly 1,000 clicks to read cost per conversion at a 2% conversion rate. Both figures are our reasoning, not platform guidance.
Can I run ten ad groups on a small budget? Arithmetically yes, informatively no. Each group would accumulate too few clicks to produce a number you could act on, and you would pay full price for the privilege.
Should I add an ad group or add hints when volume is low? Neither, if the budget is already fully committed. Raise the bid or widen the location first, since both are reversible and neither costs you attribution.
Does this mean structure advice is wrong? No, it means it is sequenced. Consolidate to get signal, split once budget growth funds readable groups.
Sources
| Claim | Source | Tier |
|---|---|---|
| Ad groups should stay focused on a single product, theme or intent area | OpenAI Help Center, Create Ad Groups for ChatGPT Ads, August 2026 | Confirmed, primary |
| Daily budget is an average over a seven day period, with daily spend fluctuating | OpenAI Help Center, Create Campaigns for ChatGPT Ads, 2026 | Confirmed, primary |
| Minimum daily budget of 25 USD | OpenAI Help Center, Create Campaigns for ChatGPT Ads, 2026 | Confirmed, primary |
| No published maximum number of ad groups per campaign | OpenAI documentation, by absence | Absent |
| Three to five dollar CPC starting bid guidance | Multiple agency write-ups citing OpenAI guidance, 2026 | Reported |
| Delivery floor of 100 clicks and conversion floor of 1,000 clicks | Newtation reasoning from sampling requirements | Inference, ours |
| 2% click-to-conversion assumption | Illustrative assumption, not a platform or benchmark figure | Inference, ours |
Related reading
- How to Structure a ChatGPT Ads Account When There Are No Keywords
- How Many Context Hints Should One Ad Group Have?
- Budget Calculator
- What ChatGPT Ads Cost
- Daily Budget or Total Campaign Budget on ChatGPT Ads
Changelog
12 September 2026, v1.0. First publication. Replaces the undefined word "focused" with two spend floors and works the arithmetic at 25, 100 and 500 USD a day.
Field kit
Tools
Site
OpenAI (primary)
Ansh works across GEO strategy, B2B research, and execution. At InPromptAds, he translates new AI advertising products into clear operating advice, tests, and measurement questions for marketing teams.
