The window is closing
Average CPC on ChatGPT Ads is 3-5 dollars today. On Google, the same-intent B2B searches can exceed 250 dollars. As competition grows, that gap will close. Early movers finish the learning curve before their rivals do.
When Google AdWords opened in 2000, the early movers dominated the market for years. ChatGPT Ads is at exactly that point today: CPCs are still 3-5 dollars, competition is thin, inventory is not yet saturated. Webtures runs the whole process from setup to feed, from campaigns to measurement, and builds the organic AI visibility that advertising cannot buy, all under one roof.
Users no longer scan ten blue links. They ask, get an answer, and decide. On that journey your brand is either inside the answer or nowhere. On 9 February 2026 OpenAI placed ads in ChatGPT and, in the first six weeks, reached 100 million dollars in annualised revenue and more than 600 advertisers. This is not an experiment; it is the birth of a channel.
Average CPC on ChatGPT Ads is 3-5 dollars today. On Google, the same-intent B2B searches can exceed 250 dollars. As competition grows, that gap will close. Early movers finish the learning curve before their rivals do.
OpenAI's Answer Independence principle is clear: advertisers cannot shape, rank or alter ChatGPT's answers. ChatGPT recommending your brand cannot be bought; that is entirely GEO/AEO work. Anyone who runs the two channels separately loses on both.
ChatGPT Ads is currently in the US, Canada, Australia, New Zealand, the UK, Japan and South Korea. Türkiye and the EU are excluded. This is not an obstacle; it is preparation time. When the channel opens, the brand whose feed, measurement and entity signals are ready scales on day one.
Both bars share one scale. The gap will close as competition grows; today's range is not permanent.
The last figure is deliberately negative. Because trust in AI answers is fragile, creative quality is critical; a well-written, honest ad card does more work here than spectacle.
ChatGPT Ads is not a classic PPC channel. There are no keywords, only context hints. There is no quality score, only a relevance-weighted second-price auction. There is no view-through measurement and no third-party verification. So the Google Ads playbook does not transfer one to one; you have to learn the channel by running it, and learning it on a client's budget gets expensive. What we sell is not a campaign setup; it is an integrated system spanning account, feed, creative, measurement, organic AI visibility and ad relevance signal.
Eligible legal-entity assessment, account opening, verification, billing profile, team authorisation and payment configuration. The country choice is permanent and cannot be undone; we make that decision together.
Purchase-intent map, context-hint clusters, campaign and ad-group hierarchy, negative-context management. We design, systematically, which conversation themes you want to appear in.
Ad-card copy, 128x128 favicon production, image variants, an A/B test matrix and landing-page matching. Brand name, favicon, headline, body, image and destination URL all work together in a single card.
A catalogue that matches the OpenAI Commerce feed specification, is_ads_eligible flag management, SFTP upload, GTIN, price and stock hygiene, and ChatGPT Shopping compliance.
Conversions API and pixel setup, UTM taxonomy, GA4 and data-warehouse integration, and geo-holdout tests. Since the platform does not yet offer incrementality, we measure incremental impact with our own test design.
The organic AI visibility ads cannot buy: entity building, fact-dense content, structured data, citation gains and multi-engine visibility tracking. For detail, see our Generative Engine Optimization and AI Visibility Web Design services.
We measure where you stand before spending on ads, and we never try to rescue a weak foundation with advertising. We start small, learn fast, and scale on data.
You need to know where you stand before spending on ads. We map 50-100 high-intent purchase prompts in your category and measure your multi-engine visibility.
Output: an AI Visibility Diagnostic Report with a current-state score, a competitor map and a 90-day priority list.
Advertising does not save a weak foundation; it just costs more. We get the feed, measurement stack and entity signals launch-ready.
is_ads_eligible flaggingOutput: a launch-ready feed, a live measurement line and an AI-readable site structure.
Start small, learn fast, scale on data. We start with a single high-intent theme; opening 10 campaigns at once makes learning impossible.
Output: validated winning context clusters, a creative library and real CPA and ROAS data.
We write the scaling threshold in advance so no emotional call is made later. If, at 90 days, your blended CPA is within 1.5 times your Google Search CPA, scale; if not, move budget to GEO/AEO.
Output: a scalable, repeatable campaign structure.
This platform changes weekly. Since February it has added a self-serve panel, CPC, oCPC, daily budgets, custom audiences and new countries. We track and adapt at that pace.
Output: quarterly strategy revisions and an operation that stays ahead of the channel.
| Week | Stage | Main activity | Concrete output |
|---|---|---|---|
| 1 | Diagnose | Prompt map, multi-engine scan | Visibility score and competitor map |
| 2 | Diagnose | Technical audit, eligibility assessment | Diagnostic report and roadmap sign-off |
| 3 | Foundation | Account opening, verification, billing profile | Active ad account |
| 4 | Foundation | Feed build and first upload | Approved product catalogue |
| 5 | Foundation | Measurement stack, schema, landing pages | Live Conversions API and pixel |
| 6 | Pilot | Campaign architecture and creative production | Launch-ready 1 campaign, 3 ad groups |
| 7 | Pilot | Go live: CPC, controlled budget | First impressions |
| 8-9 | Pilot | Context-hint testing, creative rotation | List of delivering and non-delivering hints |
| 10 | Pilot | First performance review | 30-day performance report |
| 11-12 | Pilot | Scaling winners, negative cleanup | CTR and CPC optimisation report |
| 13 | Pilot | Geo-holdout incrementality test | Incremental-impact measurement |
| 14 | Decision | Scaling-threshold assessment | Scale, hold or stop decision |
| Parallel: 1-13 | GEO/AEO | Continuous content, entity, citation work | Monthly AI visibility score |
This is the most critical point in understanding ChatGPT Ads. OpenAI's principle is firm: advertisers cannot shape, rank or alter ChatGPT's answers. The model is not even aware an ad exists unless you ask; when you do, it gives a fully neutral assessment and may even criticise the advertiser's product.
| What advertising gives you | What advertising cannot give you |
|---|---|
| A sponsored card below the answer | A brand recommendation inside the answer |
| Measurable clicks and conversions | ChatGPT recommending you |
| Instant visibility | Authority and perceived trust |
| Presence that toggles with budget | Lasting, compounding presence |
| A controlled message | A place in comparison lists |
The right column is entirely GEO/AEO work, and the two sides feed each other: clean structured data, strong entity signals and fact-dense content raise both organic mentions and your ad relevance score. As relevance rises you win more impressions on the same budget; a well-written, lower-bid ad can beat a weak, higher-bid one. Webtures runs both with the same team and the same data set. Split them across separate teams and the multiplier effect disappears.
Let us be plain: ChatGPT Ads is not currently open to advertisers in Türkiye. Available countries are the US, Canada, Australia, New Zealand, the UK, Japan and South Korea. There is no official date for the EU; OpenAI has said it will approach Europe country by country rather than as a bloc, and that ads will only be shown to users who opt in. If anyone tells you they are running ChatGPT ads in Türkiye, be careful.
If you have a legal entity in, or sales into, the US, UK, Canada, Australia, Japan or Korea, an account can be opened. Critical warning: the account country is permanent and cannot be undone. Make this decision with us.
You cannot advertise, but neither can your competitors. This market is won through content authority, entity strength and structured data. Turkish users already research on ChatGPT; the only path to visibility here is organic.
The feed, measurement stack, creative library and entity signals are prepared today. When Türkiye opens, you are scaling while your competitors are still learning how to open an account.
ChatGPT Ads still has no view-through attribution, brand lift, incrementality or third-party verification. This is the channel's least mature side. To avoid flying blind, we add our own measurement layer on top of the in-platform metrics.
| Scope | Starter | Growth · most popular | Enterprise |
|---|---|---|---|
| For whom | First-time AI-channel entrants | Actively advertising brands | Multi-market, multi-category |
| Diagnostic report | Yes | Yes | Quarterly refresh |
| Account setup | Yes | Yes | Multi-market |
| Campaign management | 1 campaign | Up to 5 campaigns | Unlimited |
| Creative production | 4 variants/month | 12 variants/month | Unlimited, video prep |
| Product feed | Basic setup | Full hygiene and optimisation | Multi-catalogue, automation |
| Measurement stack | Pixel | Pixel and Conversions API | Geo-holdout tests |
| GEO/AEO content | None | 4 pieces/month | 12 pieces/month, entity programme |
| Reporting | Monthly | Monthly, biweekly call | Weekly, custom dashboard |
| Dedicated team | Shared | Dedicated specialist | Dedicated team, strategy director |
Media spend is not included in the service fee and is paid directly to OpenAI. The model is postpay; ads are billed after they run. A temporary 100 dollar verification hold may apply when a card is added. For meaningful learning, the recommended minimum media budget is 5,000 dollars (B2B) to 15,000 dollars (e-commerce) over 90 days.
We have built search visibility since 2010. The channel moved from Google to ChatGPT; the problem stayed the same: is your brand inside the answer or not?
We were the first team to position Generative Engine Optimization and Answer Engine Optimization in the Turkish market. We are not learning this channel's rules after the fact; we are writing them.
Brantial is our own platform for measuring AI visibility. We test the methodology we sell to clients on our own product.
Local depth in Türkiye, eligible-market access in London. For our exporter clients this is an operational advantage. See the results in our success stories.
ChatGPT Ads is at exactly the point Google AdWords was at in 2000. But with one difference this time: advertising cannot buy ChatGPT recommending you. The recommendation has to be earned. A brand that does not do both loses on both.
Kaan Gülten · Founder and CEO, Webtures
No. ChatGPT Ads is currently open to advertisers only in the United States, Canada, Australia, New Zealand, the United Kingdom, Japan and South Korea. Türkiye and the European Union are not yet covered. However, Turkish brands with a legal entity in those countries, or selling into those markets, can open an account. The account country is permanent and cannot be changed later.
OpenAI recommends a maximum bid of 3 to 5 dollars for CPC. The default maximum CPM bid is 60 dollars; as inventory has grown, realised rates have settled into the 25 to 45 dollar range. Since May 2026 the self-serve panel has had no minimum spend requirement.
No. Under OpenAI's stated principle, advertisers cannot shape, rank or alter ChatGPT's answers. Ads run in systems separate from the answer, and the model is not aware an ad exists unless you ask. ChatGPT recommending your brand is earned only through organic AI visibility, that is GEO/AEO work.
Ads appear below the end of a ChatGPT answer, with a Sponsored label. The ad card contains a brand name, favicon, headline, body text, image and destination link. Typically a single ad is shown per conversation.
Only signed-in users aged 18 and over on the Free and Go plans. Plus, Pro, Business, Enterprise and Edu plans are ad-free. No ads are shown in temporary chats or the Atlas browser.
No. Instead of keywords it uses context hints: free-text phrases defined at ad-group level that describe the conversation themes where your product or service may be relevant. They are not exact matches and do not guarantee delivery.
No. Advertisers cannot access user conversations, chat history, memories, name, email, precise location or IP. They receive only aggregated, de-identified performance data. OpenAI does not sell user data to advertisers.
Under the launch policy, prohibited categories include adult content and dating, alcohol and tobacco, counterfeit goods, crypto and debt/credit services, gambling, prescription drugs and health claims, legal services, political content and unproven wellness claims. Financial services, healthcare and legal are open only to approved advertisers via case-by-case manual review.
No, at least not today. The channel's volume is far below Google Search. The right positioning treats ChatGPT Ads as a complementary, high-intent channel rather than a replacement. Early Criteo data suggests ChatGPT-referred users convert at roughly 1.5 times the rate of other channels.
Conversion tracking via the Conversions API and pixel, analytics integration through a UTM taxonomy, and incremental-impact measurement with geo-holdout tests. Because the platform does not yet offer view-through attribution, brand lift or third-party verification, we complete that layer with our own measurement design.
For meaningful learning we recommend a media budget of 5,000 to 15,000 dollars over 60 to 90 days. Below that, statistically reliable results become difficult. There is no minimum spend requirement on the platform side.
First impressions begin 6 to 7 weeks after setup. The first meaningful performance review comes after 30 days, and the scaling decision at the end of 90 days. Context hints take longer to learn than Google Ads; judging the channel on the first 30 days of data is a common mistake.
Two ways. First, organic AI visibility: through GEO/AEO work we increase how often you are mentioned in ChatGPT, Gemini, Perplexity and Copilot answers. Second, readiness mode: we build the feed, measurement stack, creative library and entity signals today, so you go live on day one when the channel opens in Türkiye.
Yes, we support that. The Enterprise package includes setup, training and a playbook handover; your team reaches a level to take over operations within 90 days. We can continue in a consulting model after handover.
Talk through your goals in a free 30-minute call. We review the opportunities in your search and AI visibility, then set the priority steps for your growth.
Book a strategy call30 minutes · free · no commitment Free AI visibility analysisYour readiness score in 60 seconds