Creating effective ChatGPT ads in 2026 means writing conversational, context-aware copy that feels like a helpful recommendation rather than a promotional interruption. ChatGPT’s ad system rewards relevance over budget, so the brands winning on the platform are those who understand the conversational moment their audience is in, not just the keywords they search for. This article covers the formats available, how targeting works, what makes copy perform, and how ChatGPT advertising compares to Google Ads.
What ad formats does ChatGPT actually support in 2026?
ChatGPT supports three confirmed ad formats in 2026:
- Sponsored answer cards: the primary format, clearly labeled “Sponsored,” appearing below the AI’s response (never inside it), containing a short headline, brief description, a square image, and a direct URL. Only one ad unit appears per response.
- Product spotlight ads: shopping-focused carousels designed for retail, showing product images, pricing, and direct checkout links.
- Contextual sidebar placements: desktop-only units supporting standard IAB display creative sizes.
Shopify merchants can sync their catalogs directly so ChatGPT surfaces specific products with accurate availability data. Retailers can also use the product feed feature in Ads Manager to upload an entire catalog and let ChatGPT generate ads for each item, matching products to relevant conversations automatically.
App install campaigns are also available through SearchGPT, with direct install buttons for mobile applications.
Ads appear only to logged-in adult users on the Free and Go tiers. Plus, Pro, Business, Enterprise, and Education accounts remain ad-free. Ads do not appear in temporary chats, after image generation, or in the ChatGPT Atlas browser. Richer formats like video and interactive conversational units have been discussed publicly but are not confirmed live as of mid-2026.
How does ChatGPT decide which ads to show users?
ChatGPT selects ads using a relevance-weighted second-price auction. The system scores relevance across four signals:
- The advertiser-written context hints
- The landing page content
- The ad title
- The ad copy
A more relevant, well-written ad can outrank a competitor with a higher bid. This means creative quality and targeting precision directly affect delivery, not just spend.
The primary targeting signal is the topic and intent of the current conversation. ChatGPT reads the entire conversation thread, not just the most recent message, to assess context. For users who opt into personalized ads, the system may also factor in past chat history, memory, prior ad interactions, and general location or language. Advertisers never receive user-level data; they see only aggregated performance metrics.
OpenAI partnered with Criteo as its first ad technology partner, connecting an initial network of advertisers to the platform. Criteo’s Commerce Media Platform handles ad serving, targeting, and measurement for many early campaigns, particularly in e-commerce and retail. Major agency partners have also integrated with the platform, including:
- Dentsu
- Omnicom
- Publicis
- Pacvue
- StackAdapt
OpenAI’s own published update confirmed early pilot results showed low dismissal rates and no measurable impact on consumer trust metrics, with ongoing improvements in ad relevance. Ads are filtered out of sensitive content contexts entirely, in line with OpenAI’s stated principles of answer independence and long-term user value.
What makes a ChatGPT ad effective versus ignored?
A ChatGPT ad is effective when it feels like it belongs in the conversation: specific, honest, and written like a knowledgeable recommendation rather than a promotional message. Ads that match the register of the conversation, use CTAs that read as logical next steps, and acknowledge the user’s context outperform ads that lead with aggressive sales language or superlative claims.
OpenAI enforces guidelines requiring ads to be useful and non-disruptive. Ads that overpromise lose relevance signals quickly and stop appearing. Honest capability claims consistently outperform exaggerated ones. Creative that imitates the ChatGPT interface, uses offensive language, or makes misleading claims is rejected outright under OpenAI’s ad policies.
The user’s mindset in a ChatGPT conversation is fundamentally different from that on a search results page. They are in a trust-based exchange, often mid-research or mid-decision. Copy optimized for search or social scrolling feels like an intrusion in that context. The ads that perform are ones where the product or service genuinely addresses the problem the user is working through at that moment.
Early data from Criteo’s work with US retailers suggests that users referred from conversational AI convert at roughly 1.5 times the rate of users from other referral channels, which points to the quality of intent these conversations carry. That quality is only captured when the ad itself matches the conversational context rather than disrupting it.
Would you like us to advertise for you in ChatGPT?
ChatGPT Ads are coming to WP SEO AI soon. With the new feature, you’ll be able to promote your services or products directly inside ChatGPT using the content you already have on your website, while managing campaigns, spend and conversions from the same WP SEO AI dashboard.
Join the waiting list and be among the first companies to start advertising in ChatGPT when the feature goes live.
How do you write ad copy that works inside a conversational AI?
Writing effective ChatGPT ad copy starts with mapping the conversations your customers are likely having, not building keyword lists. Before writing a single word of copy, identify:
- The specific questions your audience asks ChatGPT
- The problems they are solving
- The decisions they are weighing
That conversational map is your creative brief.
Headline and description constraints
The sponsored answer card format is tight. OpenAI recommends keeping the headline to around 16 characters and the description to around 32 characters. Every word must carry weight.
The principle is information over persuasion: explain what you offer specifically, not what makes you the best. Vague benefit statements like “transform your workflow” waste the limited space. Specific ones like “free 14-day trial” or “sync with Shopify” do not.
Image and creative approach
For image creatives, OpenAI’s guidance prioritizes clarity and specificity over attention-grabbing visuals. The image should answer “What is this?” within a second. Users have already been reading text, so a product shot or a clear visual representation of the offer works better than abstract brand imagery. Brands that develop new creative specifically for conversational context outperform those recycling assets from search or social campaigns.
Most brands starting on the platform begin with a monthly budget in the range of €10,000 to €25,000. That range gives the matching engine enough runway to learn how your context hints perform before you scale. Like any new search platform, ChatGPT Ads requires a warm-up period before optimization signals accumulate. Starting with a single well-defined campaign and a clear conversion event gives you cleaner data to work from.
What targeting options are available for ChatGPT advertisers?
ChatGPT Ads uses context hints instead of keywords. Advertisers write free-form, natural-language descriptions of the conversations where their ads should appear, set at the ad group level. The interface is a blank text box, not a keyword picker. These hints are thematic signals that guide matching but do not guarantee delivery in any specific conversation.
The most common mistake with context hints is describing the product rather than the conversation. A hint like “SaaS” tells the system almost nothing. A hint like “users comparing project management tools for a distributed team” describes the user’s situation, their decision stage, and their context. That specificity improves relevance scoring across all four auction signals.
Geo-targeting is currently available at the country level for a limited set of markets, including the US, Canada, Australia, and New Zealand, with expansion to additional markets announced but still rolling out on the advertiser side. Retargeting is not available on the self-serve platform. Advertisers cannot upload pixel audiences, session-based lists, or any form of website visitor data. Audience syncing capability is expected later in 2026 but is not confirmed live.
Campaigns support three bidding models:
- CPC (cost per click)
- CPM (cost per thousand impressions)
- oCPC (conversion-optimized CPC)
Starting CPC bids in the range of €3 to €5 are recommended. CPMs launched at a higher rate and have since settled into a range of roughly €25 to €65 depending on topic cluster and category.
Restricted categories under OpenAI’s current ad policies include gambling, adult content, cryptocurrency, and political content. Healthcare, financial services, and legal services have moved from blanket exclusion to case-by-case manual review as of the most recent policy update. Check OpenAI’s ad policies directly for the current version, as the policy is actively evolving.
How do you measure the performance of ChatGPT ads?
ChatGPT Ads performance is measured through OpenAI’s native Ads Manager, which reports seven metrics:
- Impressions
- Clicks
- Spend
- CTR
- Average CPC
- Average CPM
- Conversions
Reports are available at campaign, ad group, and ad level, with CSV export and an Insights API for programmatic access. Conversion tracking uses either the OAIQ JavaScript pixel deployed on confirmation pages or the server-side Conversions API (CAPI), which OpenAI considers the more reliable source.
The default attribution window is seven-day click and one-day view, configurable per campaign. Conversion-optimized campaigns using oCPC bidding allow eligible advertisers to pay for actual purchases or leads rather than clicks, provided a pixel or CAPI setup is in place beforehand.
A significant attribution pitfall affects many early advertisers. ChatGPT uses its own click identifier called oppref, not Google’s gclid. Mobile browsers often strip the referrer, causing ChatGPT traffic to appear as “Direct” in GA4. The fix is capturing the oppref click identifier alongside static UTM parameters. Without this, you will undercount ChatGPT’s contribution to conversions. The ChatGPT attribution setup guide from SureThing covers the technical implementation in detail.
OpenAI has not published cross-advertiser performance benchmarks by industry or objective, so there is no official standard to measure against. Early CTR figures from third-party sources range from under 1% to over 4% depending on the vertical and how CTR is measured. Treat all published benchmarks as directional planning figures, not guarantees, and build your own baseline from the first 30 to 60 days of your own campaign data.
How do ChatGPT ads differ from Google search ads?
ChatGPT ads and Google search ads target different moments in the buyer journey. Google search ads target explicit, high-intent queries where the user has already decided to look for something specific. ChatGPT ads target conversational intent: research conversations, comparisons, and decision-making discussions where the user is still forming their view. The two platforms are complements, not substitutes.
Google handles roughly five billion search queries per day worldwide. ChatGPT processes a significantly smaller volume of conversational queries, but the depth of context in those conversations is substantially higher. A user asking ChatGPT “what’s the best project management tool for a 20-person remote sales team” is further along in a considered decision than someone typing “project management software” into a search bar.
Targeting and audience control
Google Ads offers mature targeting including demographics, remarketing, in-market audiences, and cross-channel placement across Search, Display, YouTube, Shopping, and Gmail. ChatGPT Ads currently offers only contextual targeting via conversation topic, with no retargeting or demographic segmentation available to self-serve advertisers. For advertisers who rely on audience lists and remarketing, Google remains the more capable platform in 2026.
Cost and conversion benchmarks
The table below summarizes how ChatGPT Ads and Google Search Ads compare across key dimensions:
| Dimension | Google Search Ads | ChatGPT Ads |
|---|---|---|
| Query volume | ~5 billion per day | Significantly lower |
| Intent type | Explicit, high-intent keyword queries | Conversational, research-stage intent |
| Targeting options | Demographics, remarketing, in-market audiences, cross-channel | Contextual (conversation topic) only |
| Retargeting | Available | Not available (self-serve) |
| Starting CPC | Mid-single-digit range (varies by industry) | €3 to €5 recommended |
| Conversion rate data | Mature benchmarks by sector | Early stage, varies widely by vertical |
| Best fit | Direct-response, short-cycle e-commerce | B2B, SaaS, high-research categories |
Google Search’s average CPC sits in the mid-single-digit range across industries, with mature conversion rate benchmarks available by sector. ChatGPT’s recommended starting CPC is in a similar range, but conversion rate data is still early and varies widely by vertical. High-intent verticals report stronger conversion rates on ChatGPT; direct-response and short-cycle e-commerce purchases tend to perform better on Google. According to WebFX’s comparison of the two platforms, most full-funnel brands run both.
B2B companies with long sales cycles, SaaS businesses targeting technical buyers, and brands in categories where people do significant pre-purchase research have the strongest natural fit with ChatGPT Ads. Perplexity AI discontinued its advertising program entirely in early 2026, citing user trust concerns, leaving ChatGPT as the primary conversational AI advertising platform available to most advertisers. Google has integrated ads into AI Overviews and AI Mode, running through the existing Google Ads ecosystem, so advertisers already active on Google can reach AI-assisted search results without a separate platform.
The broader opportunity in generative engine advertising is significant. Industry analysts project US AI-driven search advertising will grow substantially over the next several years, and 2026 represents the earliest point at which advertisers can build expertise and establish baselines. If your audience does meaningful research before buying, starting now gives you a compounding advantage as the platform matures. For brands focused on appearing in both Google and AI-generated answers, a combined approach to generative engine optimization and paid advertising is the most complete strategy available today.
This content was generated with the help of AI and it may contain mistakes
