ChatGPT Ads moves to the performance side
OpenAI added conversion-optimized cost per click, location exclusions, average daily budgets spread across seven days, intraday spend pacing and bulk campaign management to ChatGPT Ads. Mobile measurement partnerships also bring install and in-app event attribution to app advertisers for the first time.
The ad platform is being shaped for performance marketers
The update package OpenAI sent to advertisers was reported by the trade press on July 24, 2026. It moves ChatGPT Ads from a channel worth testing towards a platform built for teams that expect measurement, automation and campaign management at scale. The headline items are conversion-optimized bidding, location exclusions, average daily budgets, intraday pacing, an API with bulk support, automatic advanced matching and refreshed product cards carrying price and rating information.
Bidding, budgets, exclusions and measurement
Conversion-focused campaigns are created by selecting the Conversions objective during setup and run on what OpenAI calls conversion-optimized cost per click: delivery steers towards clicks judged more likely to produce a conversion, while billing stays tied to valid clicks. Eligibility required a conversion event already flowing through the pixel or the Conversions API.
On budgets, fixed daily limits give way to average daily budgets: spend can vary from day to day while both the daily and the seven-day ceilings hold, and automatic pacing spreads the budget through the day. In targeting, specific locations can now be removed from a campaign, complementing the inclusion-based targeting added in May. For bulk work, campaigns, ad groups and ads can be created and updated asynchronously and at scale through the Ads API, with bulk editing also available inside the interface.
On measurement, integrations with the two largest mobile measurement partners went live, opening install and in-app event attribution for app advertisers. Those integrations are active in seven markets: the United States, Canada, Australia, New Zealand, the United Kingdom, Japan and South Korea, with Mexico and Brazil listed as coming soon. On the web side, automatic advanced matching uses hashed customer information to match additional conversions to ads. Product cards now show price, star ratings and the merchant name alongside the image.
A chat interface turns into a performance channel
For a channel to count as a performance channel it needs three things: the ability to bid towards an outcome, the ability to measure that outcome, and the ability to scale management. This package addresses all three at once. The conversions objective covers bidding, the measurement partnerships and advanced matching cover measurement, and the bulk API covers the teams running many account structures.
The second issue is comparability. If the same conversion event is measured with different definitions across Google, Meta and ChatGPT, cross-channel budget decisions get made on bad data. That is why the conversion definition and the attribution window need to be settled centrally before a new channel is opened. The third is that product cards now carry price and ratings: how shopping looks inside a conversation is tied more directly to the quality of your product data.
What it means for businesses in Türkiye
Türkiye is not on the announced market list, so a ChatGPT ad campaign is not on the agenda for most businesses here today. The preparation, however, can start now and none of it is wasted, because the same infrastructure strengthens the channels you already run. First, measurement: if you do not have a server-side conversion flow, build one, tie your conversion events to a single definition list and make sure the same event counts the same thing everywhere.
Second, product data. A catalogue where price, stock, imagery and ratings do not flow correctly falls behind both in conversational product cards and in search results; feed discipline is no longer one platform's concern. Third, content: visibility inside a chat interface is not only about ads, it is also about your brand being cited as the source that answers the question. Fourth, budget discipline: the most common mistake when a new channel opens is shifting budget before measurement exists. When Türkiye is added to the list, a prepared business starts testing with data rather than with setup.
The UNALSOFT take
When a new ad channel opens, our first question is not budget but measurement: which event counts as a conversion, who counts it and in which window. That is the order we keep in our ad management work; adding a channel is easy, building measurement that speaks the same language across channels is the hard part. When the ChatGPT side opens to Türkiye, businesses with measurement in place can compare from day one, while the others wait to accumulate data.
Measurement first, budget second.
Let's settle your conversion definitions and your cross-channel comparison together.