A product target URL is the destination web address an advertiser assigns to a ChatGPT ad, the page a user lands on after clicking the sponsored unit below a ChatGPT response. The ad creative and context hints decide where your ad shows; the product target URL decides what happens after the click.
Treat your product target URL as part of the ad you ship, because OpenAI weighs the destination page when it decides whether your ad is eligible to appear. Point it at a crawlable, relevant page and you earn delivery and conversions; point it at a blocked or generic homepage and OpenAI can reject the ad outright.
In ChatGPT Ads, the product target URL is the landing page an advertiser attaches to an ad so a click moves the user from the chat to a page the advertiser controls. In OpenAI's Ads API it is the creative.target_url field on a chat card; for product-feed ads, the destination comes from the url attribute of the matched catalog item instead.
The URL has to meet concrete requirements. OpenAI requires a valid HTTP or HTTPS address that its crawlers, OAI-AdsBot and OAI-SearchBot, can reach, and the destination must accurately represent the advertiser and the offer stated in the ad. Advertisers can also append UTM tags or dynamic macros such as {campaign_id} and {ad_id} that resolve at delivery, within OpenAI's published length limit.
The product target URL sits at the ad level of the account, campaign, ad group, and ad hierarchy, distinct from the context hints that live on the ad group. AirOps connects to a ChatGPT Ads account and surfaces the creative and landing page behind each observed ad, so you can see which destinations competitors point their spend toward.
Resources: how to send ChatGPT ad clicks to a page OpenAI's crawler can reach
A product target URL moves through the same pipeline every time a ChatGPT ad runs, from the moment you set it to the moment a click lands.
Assign: You set the destination on the ad, either as the target_url on a chat card or from the url field of a feed item for product ads.
Review: OpenAI crawls the page with OAI-AdsBot to confirm it loads, matches the ad, and follows ad policy before the ad is approved to serve.
Match: When a conversation fits, OpenAI weighs the destination alongside the creative and context hints in a relevance-weighted auction to decide eligibility.
Resolve: At delivery, any dynamic macros and UTM parameters in the URL are populated so the final link carries campaign and ad identifiers.
Click: The user taps the sponsored unit and arrives on your page, where UTM tags let GA4 or your analytics attribute the session.
The click data tells you which destinations earn traffic and conversions. It does not tell you which prompt or conversation triggered the ad, since OpenAI reports performance without exposing the chat behind it.
Where you send the click decides whether ChatGPT ad spend turns into pipeline or evaporates. The product target URL is the one creative element you fully control after the auction, so it carries the weight of the whole campaign's return on the money you commit.
Delivery depends on it: OpenAI can refuse to serve an ad whose destination its crawlers cannot load, so a page that blocks OAI-AdsBot stalls the launch before a single impression runs.
Relevance is scored here: OpenAI weighs the destination against the conversation, so a target URL that matches the specific intent wins eligibility more often than one pointed at a broad homepage.
Attribution lives here: UTM tags and dynamic macros on the URL are what let GA4 and your CRM tie a closed deal back to the ad, turning aggregated clicks into revenue you can defend to leadership.
SEO managers use product target URLs to send ChatGPT ad clicks to the same intent-matched pages they already optimize for organic answer visibility.
Content strategists use product target URLs to align the landing page's headline and offer with the ad copy so the promise stays consistent end to end.
Growth marketers use product target URLs to tag every destination with UTM parameters and route conversions into GA4 and the CRM for payback analysis.
Your destination must let OpenAI's crawlers, OAI-AdsBot and OAI-SearchBot, fetch the page, because an ad whose landing page returns a block, a 403, or a firewall error is not eligible to run and the whole campaign stalls at review.
The target URL should resolve to a page whose offer, price, and claims match the ad creative, since OpenAI reviews the ad and the destination together and rejects combinations where the promise and the page describe different things.
UTM tags and OpenAI's dynamic macros such as {campaign_id} and {ad_id} are appended to the URL and populated at delivery, which is how a raw click becomes an attributable session your analytics and CRM can report on.
Earn ad delivery by giving OpenAI a crawlable, policy-compliant destination it can review.
Lift conversion rates by matching each ChatGPT ad to a dedicated page instead of a homepage.
Attribute revenue by tagging the URL with UTMs that flow into GA4 and your CRM.
Support destinations up to 2,048 characters, enough room for full UTM and macro strings.
Reuse one product-feed URL across many catalog items pulled from the feed.
Send each ad group to a dedicated page that matches its specific intent, because a page built for the exact offer converts better than a generic homepage.
Allow OAI-AdsBot and OAI-SearchBot in robots.txt and your firewall rules, so OpenAI can crawl and approve the destination without delay.
Keep the page consistent with the ad's headline, image, and claims, since OpenAI reviews the creative and destination together.
Tag every URL with UTM parameters before launch, so no click arrives as unattributed traffic you cannot analyze.
Remove redirect chains and point the URL at the final page, because each hop adds latency that costs you conversions.
Confirm price, availability, and identifiers on product-feed destinations match the catalog item the ad renders from.
Avoid pointing every ad at your homepage to save setup time. It is the most common rejection reason and the fastest way to waste a click, because the visitor lands on a page that answers none of the intent the conversation carried.
AirOps: connects your ChatGPT Ads account and surfaces the creative and landing page behind each tracked ad, so you can see which product target URLs you and competitors are pointing spend toward.
Google Analytics 4: reads the UTM parameters on your target URL to attribute ChatGPT ad sessions, conversions, and revenue alongside your other channels.
Google Search Console: confirms the destination page is indexable and crawlable, which helps the same page earn organic answer visibility beyond the paid click.
Pick the page: Choose the single page that answers the intent you plan to advertise. Skip the homepage. You can do this this week with pages you already own, so it needs no budget approval.
Open the crawlers: Update robots.txt and firewall rules to allow OAI-AdsBot and OAI-SearchBot, then confirm the page returns a 200 to those agents.
Align the message: Edit the headline, offer, and proof on the page so they match the ad copy and image you intend to run, since OpenAI checks both together.
Build the tracking: Append UTM parameters and, where available, OpenAI's dynamic macros to the URL, then verify they pass into GA4 in a real-time test click.
Load and launch: Enter the finished URL as the ad's destination, submit the ad for OpenAI's review, and watch for crawl or policy flags before any spend begins, and clear any flags fast.
A product target URL is the destination page a ChatGPT ad sends a click to, set on the ad itself.
You configure it as the target_url on a chat card or let a product feed supply it per item.
It must stay crawlable by OAI-AdsBot and consistent with the ad, or OpenAI will not serve it.
Pointing it at a generic homepage is the quickest route to rejection and wasted spend.
The biggest gains come from matching each URL to one intent and tagging it for full attribution.
A product target URL and context hints do different jobs in the same ad. The context hints are plain-language descriptions of the conversations where you want the ad to appear, and they live at the ad group level. The product target URL is the page the click resolves to, and it lives on the ad itself. Hints shape when and where OpenAI considers showing your ad; the URL shapes what the user experiences once they tap it. Both feed the relevance check, since OpenAI weighs the destination and the hints together when it decides eligibility, but you edit them in separate places and for separate reasons. A useful way to keep them straight: hints influence the match before the click, and the target URL carries the whole burden of converting the visit after it. Getting one right does not rescue a weak version of the other.
Update the product target URL whenever the destination changes materially or the page stops matching the ad. Common triggers are a landing page move or redirect, a promotion ending, an offer or price change, or a page that starts failing the crawler check. Because OpenAI resends a changed ad for review, treat each URL edit as a relaunch: the ad can drop out of serving until the new destination is approved. On the API, sending a creative update means sending the full creative object, so you include the title, body, and image along with the new target_url, never the URL on its own. For product-feed ads there is usually no manual edit, because the destination comes from the feed item, so you update the catalog row instead and let the next feed refresh carry it through. Keep a record of each change with its date, so you can tie any dip in delivery back to a URL swap.
The same product target URL can perform differently because performance depends on more than the page itself. Each ad group carries its own context hints, creative, and audience, so the same destination meets a different conversation and a different intent in each one. A URL that matches the intent of one ad group tightly can feel generic to another whose hints describe a different need. Bids and competition vary too: the relevance-weighted auction may surface the ad more often in one group, changing the volume and quality of clicks the page receives. Seasonality, geography, and the plans and devices that see the ad all shift the mix. The practical read is that a target URL is only as strong as the match between the page and the specific intent of the ad group feeding it, so judge each pairing on its own instead of expecting one page to carry every group.
Yes, you control the product target URL directly for standard chat-card ads. You set it as the target_url when you create or update the ad, and you can change it at any time, subject to OpenAI re-reviewing the new destination before it serves. What you do not control is the conversation the ad lands in or the exact moment the auction chooses your ad, since OpenAI matches on live context. For product-feed ads the control is indirect: the destination comes from the url field of each catalog item, so you influence it by editing your feed and the ad template stays untouched. Tracking is also yours to set, through UTM parameters and OpenAI's dynamic macros applied to the URL or the campaign's query-string template. So the destination and its tracking sit fully in your hands, while the targeting and timing of the click are shaped by OpenAI's system, and you steer those through hints and bids instead.
A good product target URL is a fast, crawlable page that matches the ad's intent and makes the next action obvious. Start with the non-negotiables: it loads for OAI-AdsBot, follows OpenAI's ad policy, and represents the advertiser and offer accurately. Beyond eligibility, quality shows in the fit between the page and the conversation that triggered the ad, so a dedicated page built for one offer clears the bar that a homepage rarely does. Good destinations also carry clean UTM parameters so every click is attributable, resolve without redirect chains that add latency, and keep price, availability, and claims consistent with the creative. There is no single published benchmark number OpenAI promises, so judge quality by your own outcomes: approval on first review, a low bounce rate, and a conversion rate that holds up against the ad group's cost per click. If those hold, the URL is doing its job.