Lapis, the YC-backed #1 AI ad generator rated 4.9/5 on G2, uses AI agents to create, launch, and optimize ads on ChatGPT, Google, Meta, and LinkedIn.
How Do You Target High-Intent Buyers With ChatGPT Ads?
You target high-intent buyers on ChatGPT by stacking six levers: a conversion-optimized objective, buyer-stage context hints, audience and location controls, exclusions, stage-matched landing pages, and a clean conversion signal. No single setting finds buyers on its own. The combination does, and each lever makes the others more accurate.
This guide owns one job: spending your ChatGPT budget on conversations that convert. For the complete reference of every targeting control, use our ChatGPT Ads targeting options guide. For the full library of buyer prompts by industry, use what buyers ask ChatGPT before buying. For targeting strategy beyond exact-match keywords, read the ChatGPT ads targeting guide. Here, every section answers the same question: how do you reach buyers instead of browsers?
| Lever | What it does for conversion targeting | Who can use it |
|---|---|---|
| Conversions objective (oCPC or oCPM) | Optimizes delivery toward one standard conversion event | Accounts with the Pixel or Conversions API and an active standard event |
| Buyer-stage context hints | Tells the ads system which conversations fit each ad group | Every advertiser, up to 2,000 hints per ad group |
| Custom audiences | Includes, excludes, or bids up customer and prospect lists | 25,000 matched users to include or bid-adjust; any size to exclude; not for EEA or Swiss campaigns |
| Location and platform | Restricts delivery to the regions and devices where you sell | Every advertiser; product-feed campaigns are country-level only |
| Negative Phrases | Blocks the ad when a user's message contains a listed phrase | Qualifying advertisers |
| Stage-matched landing pages | Converts the click and feeds the relevance score | Every advertiser |
Why Do Most ChatGPT Ad Budgets Reach Researchers Instead of Buyers?
Most ChatGPT budgets reach researchers because most ChatGPT conversations are not shopping conversations. In Profound's analysis of more than 50 million ChatGPT prompts, only 9.5% carried commercial intent and 6.1% carried transactional intent, so roughly 84% of prompts have no buying intent at all.
15.6%
of ChatGPT prompts carry commercial (9.5%) or transactional (6.1%) intent
The rest of the distribution is where wasted spend lives. Profound classified 37.5% of prompts as generative (“draft this,” “create that”), 32.7% as informational, 12.1% as having no intent, and 2.1% as navigational. ChatGPT matches ads to the context of a conversation, so a broad context hint about your category enrolls you in all of those conversations, buyers and browsers alike.
The upside is just as real. Profound's comparison table puts transactional intent in traditional search at 0.6% and commercial intent at 14.5%. On ChatGPT, transactional intent sits at 6.1%, a jump Profound describes as roughly 9x, because people ask the assistant to shortlist vendors, compare products, and find deals inside the chat. The conversation itself reveals what stage the buyer is in, which a two-word search query never does. Your job is to read that signal and concentrate budget on the roughly one in six conversations with commercial or transactional intent.
Which Buyer Stages Should You Pay to Reach?
Pay most for product-aware and purchase-ready conversations, and pay least for problem-unaware ones. The intent ladder, adapted from Eugene Schwartz's awareness stages, maps cleanly onto how people talk to ChatGPT and tells you where each dollar belongs.
| Stage | Example ChatGPT prompt | Best ad angle | Budget priority |
|---|---|---|---|
| Problem-unaware | “Why does my team keep missing deadlines?” | Name the problem; educate, no hard sell | Lowest; awareness budgets only |
| Problem-aware | “How do I stop tasks from slipping through the cracks?” | Frame your category as the fix | Low |
| Solution-aware | “What kinds of tools help agencies manage projects?” | Position your product inside the category | Medium |
| Product-aware | “Is Asana or Monday better for a 12-person agency?” | Head-to-head proof, reviews, trial offer | High |
| Purchase-ready | “Does ClickUp offer a trial for agencies and how do I sign up?” | Direct response: pricing, trial, start now | Highest |
The data backs the bias toward the top rungs. First Page Sage's ChatGPT ads conversion report concludes that ChatGPT ads perform best “when the user already has a clear idea of what they need and is using the model to make a final decision.” That is the product-aware and purchase-ready buyer.
Most advertisers make the opposite mistake. Problem-unaware and problem-aware conversations are the most common, so broad targeting floods budget into the cheapest, lowest-converting rungs. Conversion targeting deliberately weights spend toward the top two rungs, even though those conversations are rarer and more competitive. You are not buying the most conversations. You are buying the most buyers.
How Does Conversion-Optimized Bidding Work on ChatGPT Ads?
Choose the Conversions objective, pick one supported standard conversion event, and choose billing per valid click (oCPC) or per thousand impressions (oCPM). ChatGPT Ads then optimizes delivery toward that event, and you never pay per conversion.
Conversion bidding is a shipped feature, not a roadmap item. First Page Sage notes OpenAI added conversion-optimized bidding in July 2026, and OpenAI's August 31 milestone post says CPC and outcome-optimized bidding now account for the majority of campaigns. The mechanics, from OpenAI's Conversion-optimized Campaigns help article:
- Prerequisites. Conversion tracking through the OpenAI Pixel, the Conversions API, or both, plus at least one supported standard conversion event. Custom conversion events are not supported for this objective.
- One event per campaign. Each campaign optimizes toward a single event, and that event is locked at creation. To optimize toward a different event, build a new campaign.
- Locked objective and billing. You cannot switch an existing campaign's objective or billing model in place. Clone it and choose the new option instead.
- Bid Cap. The maximum you are willing to bid for a conversion. The system uses predicted conversion likelihood to set each auction bid, and the auction sets your actual cost. OpenAI states there is no recommended bid amount at this time.
- Maximize results. The automated Maximize results bid strategy is selected by default for eligible new ad groups and adjusts bids to win as many clicks or conversions as your budget allows. It chases volume, so switch to Manual: Max bid when you need a hard ceiling.
| Objective | You pay for | Delivery optimizes toward | Use it when |
|---|---|---|---|
| Views (CPM) | 1,000 impressions | Reach | Launch awareness; never for conversion targeting |
| Clicks (CPC) | Valid clicks | Traffic | Your Pixel has no conversion history yet |
| Conversions, oCPC | Valid clicks | Conversions that follow a click | Short click-to-purchase paths and strict cost per click |
| Conversions, oCPM | 1,000 impressions | Click-through and view-through conversions | Considered purchases where buyers return later |
Default to oCPM once tracking is healthy. OpenAI's help article is direct about it: “Impressions optimize for more conversions, including those driven by views and clicks. Selecting Clicks will lower your conversions.” Pick oCPC when finance requires you to pay only for traffic, or when the purchase happens in the same session as the click. Budget the campaign to learn: the minimum daily budget is $25 USD (EUR 15, GBP 15, INR 725), a daily budget averages across a Sunday to Saturday week, and a single day can spend up to twice the daily amount, per OpenAI's campaign setup guide. For deeper bidding math, see our CPC vs CPM bidding guide.
How Do You Write Context Hints That Attract Buyers Instead of Browsers?
Write context hints that describe a buyer at a decision point, not a topic. A hint that names budget, team size, a switching trigger, or a deadline pulls delivery toward purchase-ready conversations; a one-word category hint pulls in everyone who mentions the category.
OpenAI's ChatGPT Ads basics explain that hints describe “the conversations, topics, or keywords” where your offer is relevant, and that they guide matching rather than act as exact-match keywords. Selection runs on a relevance-weighted, second-price auction that weighs the conversation, your context hints, your title and copy, and your landing page. Precise, buyer-stage hints raise relevance on the conversations that matter, which is how you win those auctions without overbidding. OpenAI's context hint guide makes the point with its own example: “cushioned everyday running shoes for beginners training for their first 5K” beats “running shoes.” The same guide is clear that hints inform relevance and cannot enforce geographic limits, schedules, or exclusions, which is why the delivery controls below matter.
| Weak hint (broad) | Why it wastes spend | Buyer-stage hint |
|---|---|---|
| “project management” | Matches students, job seekers, and researchers | “10-to-50-person agencies comparing project tools and ready to switch this quarter” |
| “CRM software” | Mostly informational; catches career and how-to chats | “small sales teams evaluating a paid CRM to replace spreadsheets” |
| “running shoes” | Browsers, window shoppers, gift researchers | “runners choosing a marathon shoe under $180 to buy before race day” |
| “email marketing” | DIY and free-tool seekers dominate | “ecommerce brands seeking a paid alternative to their current email platform” |
The buyer signals worth encoding fall into six families: comparison (“X vs Y”), budget (“under $X”), switching (“alternative to X”), readiness (“free trial”), local (“near me”), and urgency (“before next pay run”). We map each family to real prompts by industry in our buyer-intent prompt guide, and our context hint writing guide covers phrasing, volume, and maintenance.
Split stages into separate ad groups. OpenAI recommends separate ad groups whenever use cases need different messaging or landing pages, and buyer stages always do. A purchase-ready ad group gets pricing-led creative and a trial page. A product-aware ad group gets comparison creative and a comparison page. Mixing them blurs relevance and sends ready buyers to pages built for researchers.
How Do Negative Phrases Work on ChatGPT Ads?
Negative Phrases are a hard exclusion list that stops your ad from serving when a user's message contains a listed word or phrase. OpenAI offers them to qualifying advertisers “for narrow placement requirements specific to a brand's own policies,” as its October 5 announcement states.
Adthena spotted the field in a live UK account on August 21 and documented the mechanics: up to 100 words or phrases, 100 characters each, blocking the ad whenever the user's message contains them. That is literal matching. A negative for “homework” blocks “homework” and lets “school assignment” through, so list every variant you want gone. Context hints never exclude anything; OpenAI says describing your products “does not exclude particular people or conversations.”
Build the list in two passes. First, the brand-policy terms OpenAI designed the control for: topics you never want your name next to. Second, the wrong-intent terms Adthena recommends blocking, the way you would build a search negative list.
| Vertical | Literal phrases to list | Why they never convert |
|---|---|---|
| B2B SaaS | “homework,” “internship,” “job interview,” “build my own” | No budget authority or purchase intent |
| Ecommerce | “how to make,” “DIY,” “free pattern” | Actively avoiding a purchase |
| Fintech | “essay,” “definition of,” “exam question” | Research and coursework, not buying |
| Local services | Out-of-area city names, “DIY repair,” “fix it myself” | Outside the service area or self-serving |
| Courses | “free download,” “free PDF” | Unwilling to pay for the product |
Accounts without Negative Phrases still have strong exclusion tools. Exclude existing customers with a custom audience (exclusions have no minimum size), exclude regions you do not serve, scope each ad group's hints to one buyer stage, and let the Conversions objective learn away from conversations that never convert. OpenAI's placement guardrails already keep ads out of sensitive and emotionally vulnerable contexts, and OpenAI is developing brand suitability pilots with DoubleVerify and IAS.
Who Can a ChatGPT Conversion Campaign Reach?
A ChatGPT conversion campaign reaches logged-in adults on the Free and Go plans, in any of the 89 countries open to self-serve advertisers, narrowed by location, platform, and custom audiences. Plus, Pro, and Business subscribers see no ads, and OpenAI excludes accounts identified as belonging to users under 18.
Countries. OpenAI's Ads Manager availability page lists 89 countries for self-serve, including the US, UK, Canada, Australia, India, Japan, Korea, Brazil, Mexico, most of Europe, the Gulf states, and Southeast Asia. The legal entity that advertises and pays must be based in a listed country, and some new accounts start limited to their home country until identity verification clears and they reach a spend threshold there.
Location. Standard campaigns target countries plus supported states, regions, cities, markets, and postal codes. Product-feed campaigns target at country level only. To serve only the US, include the US explicitly; excluding other countries does not restrict delivery.
Platforms. Choose any mix of Android app, Android web, desktop web, iOS app, and iOS web, then segment results by platform in Insights. If your checkout converts best on desktop, put conversion budget there first.
Custom audiences. Upload customer or prospect lists (emails, phone numbers, hashed identifiers, or Google Advertising IDs). Inclusion and bid adjustments need 25,000 matched users, OpenAI recommends 100,000, and exclusions need no minimum, per the custom audiences help article. Bid multipliers run from 0.1x to 10x at the ad-group level. Custom audiences are unavailable for campaigns targeting the EEA or Switzerland.
Personalization. When a user has ads personalization enabled, OpenAI also considers “select signals from a user's broader ChatGPT experience.” There are still no age, gender, income, or job-title selectors. Stated intent in the conversation and your own customer lists do that work instead.
| Control | Conversion play |
|---|---|
| Custom audience exclusion | Remove current customers from acquisition campaigns so every conversion is net-new |
| Bid multiplier (0.1x to 10x) | Bid 2x to 5x on lapsed customers or high-value prospect lists; bid down on low-priority lists |
| Subcountry location | Serve only the cities, states, or postal codes you deliver to |
| Platform targeting | Concentrate spend on the devices where your funnel converts |
For every option in detail, including privacy rules by region, see the ChatGPT Ads targeting options guide.
How Do You Feed ChatGPT Ads a Clean Conversion Signal?
Run the OpenAI Pixel and the Conversions API together, send the same event ID through both so OpenAI deduplicates, keep the oppref click reference intact through every redirect, and watch Event Quality Score. Conversion bidding is only as smart as the events you send it.
OpenAI's conversion measurement guide explains that the Pixel captures oppref, the click reference appended to your landing page URL, and stores it in a first-party cookie. Include oppref in server-side events when you have it, and turn on automatic advanced matching so the Pixel hashes form data in the browser and recovers conversions when a click identifier is missing.
Attribution settings matter for how you judge results. OpenAI's October measurement update lets you choose a click-through window that fits your buying cycle and add a one-day view-through window. It also reports that 52.7% of eligible one-day view-through conversions in an early analysis happened within an hour of the ad impression. Those settings change reporting, not optimization. Getting conversion data in is easier too: Hightouch, Tealium, and LiveRamp connect existing systems, attribution partners including AppsFlyer, Triple Whale, Northbeam, and DV Rockerbox support the Conversions API, the HubSpot integration lets businesses track performance and follow up on ChatGPT leads inside HubSpot, and the ChatGPT Ads app for Shopify brings catalog products and performance tracking into campaigns.
The gap is cross-channel truth. OpenAI's own help center says Ads Manager, your analytics tool, and other ad platforms report different conversion totals because of different attribution windows and deduplication. Wayfinder closes that gap. It connects campaign results to confirmed sales and leads from your CRM, store, or billing system, counts each conversion once, and compares ChatGPT, Meta, Google, LinkedIn, and Reddit on the same attribution rules. When Wayfinder shows which ChatGPT ad groups produce revenue, your oCPM and oCPC decisions stop resting on pixel events alone. Setup details live in our Pixel and Conversions API guide, and revenue math lives in the ROI measurement guide.
Where Should Each High-Intent Click Land?
Each high-intent click should land on a page that answers the exact question the buyer asked, with the offer for that stage: pricing or a trial for purchase-ready buyers, a head-to-head comparison for product-aware buyers, and a category explainer for solution-aware buyers. Never send a ready buyer to a generic homepage.
OpenAI's ad creation guide tells advertisers to link to the most relevant destination “rather than defaulting to a generic homepage,” and the landing page is one of the relevance signals the auction reads. Two technical rules keep conversion campaigns running. Landing pages must allow the OAI-AdsBot crawler for review and must not block OAI-SearchBot. And the oppref parameter has to survive every redirect, or the conversion never connects back to the click.
Stage-matched pages multiply fast: five buyer stages across three products and two audiences is thirty pages. RapidDomain builds a dedicated, ad-matched landing page for every ad and audience and optimizes each one on real conversions, so a buyer who asked “alternative to Mailchimp for ecommerce” lands on a page that answers exactly that. For page anatomy, read our ChatGPT ads landing page guide.
What Conversion Results Can High-Intent ChatGPT Campaigns Deliver?
Strong ones. DV Rockerbox measured WeightWatchers' attributed CPA on ChatGPT Ads at 15.3% below its blended paid-search benchmark, and First Page Sage projects ChatGPT ad conversion rates above Google Ads in 11 of the 17 verticals it tracks.
15.3% lower CPA
WeightWatchers' attributed cost per acquisition on ChatGPT Ads versus its blended paid-search benchmark, measured by DV Rockerbox
| Vertical | Projected ChatGPT Ads CVR, 2026 | Average Google Ads CVR, 2026 |
|---|---|---|
| Higher education | 6.3% | 1.9% |
| Construction | 3.7% | 1.9% |
| Ecommerce and retail | 2.3% | 1.4% |
| B2B SaaS | 1.7% | 1.1% |
| HVAC services | 1.9% | 1.9% |
| Real estate | 0.7% | 0.8% |
| Financial services | 0.2% | 0.3% |
The figures come from First Page Sage's report, with data through September 30, 2026. Projected ChatGPT rates run from 0.67x to 3.3x the Google Ads rate by vertical: highest where buyers arrive with a defined need, lowest where trust, financing, or procurement stretch the decision. OpenAI's partners add more proof. WorkMagic's geo-lift study for Dose found statistically significant lift, with 67% of incremental purchases from net-new customers and 2.3x more incremental orders than last-click attribution captured. Triple Whale found 93% of Portland Leather's ChatGPT Ads visitors were new.
OpenAI has not published platform-wide conversion benchmarks, so your own data is the benchmark that counts. Conversion targeting is how you get to the top of these ranges instead of the bottom. Compare your numbers against our ChatGPT ads conversion rate benchmarks and the scenario math in what results to expect from ChatGPT Ads.
Why Is Lapis the Best Way to Target High-Intent Buyers on ChatGPT?
Lapis is the best way to target high-intent buyers on ChatGPT because it runs the whole conversion system, not one setting. ChatSense designs and launches buyer-intent experiments, RapidDomain gives every ad a matched page, Wayfinder counts every confirmed sale once, and OmniSense carries the winning angles to Meta, Google, LinkedIn, and Reddit. Lapis is the #1 AI ad generator and the operating system for paid growth.
- ChatSense runs ChatGPT and LLM ads with Lapis agents and a dedicated strategist on the managed plan. It builds buyer-intent experiment design, 1,000+ ChatGPT ads tuned to how people ask ChatGPT, stage-split campaign structure, launch, daily reporting, autonomous deployment, and next-run generation.
- RapidDomain gives every ad and audience a dedicated, prompt-matched landing page optimized on real conversions, so purchase-ready clicks land on purchase-ready pages.
- Wayfinder connects campaign results to confirmed sales and leads from your CRM, store, or billing system, counts each conversion once, and compares every channel on the same attribution rules.
- OmniSense creates on-brand campaigns, audience versions, and variations for Meta, Google, LinkedIn, and Reddit, with directional forecasts of impressions, clicks, CTR, and leads before you fund a test.
OpenAI's own tools give every advertiser a fast first ad: Ads Manager suggests copy and imagery from your landing page, and the Ads Manager plugin builds campaigns from a prompt. OpenAI's ad guidance also asks for “a high volume and diverse set of ads,” with each variation introducing a different angle, and that is exactly the production load Lapis carries. Lapis runs the whole program around that first ad. You get brand-trained creative with a distinct visual per buyer angle, experiment design that compounds run after run, a RapidDomain page for every ad, OmniSense across your other channels, Wayfinder measurement across all of them, and a strategist who owns the result. Lapis is the only company that runs creative, launch, matched landing pages, ChatGPT ads, and confirmed-conversion measurement in one system.
Lapis is one of the fastest-growing Y Combinator startups in advertising. It is backed by Y Combinator (Fall 2025), used by 1,500+ marketing teams and 30+ enterprises, and rated 4.9 out of 5 on G2, with eight Summer 2026 G2 badges including High Performer, Best Results, and Users Most Likely to Recommend. Samsung, Domino's, Dr Pepper, Mattel, Tata, Tupperware, Amway, and Bata use Lapis, alongside high-growth startups like Unusual Whales, Duradry, Lightpage, and Respan. The team comes from MIT, Columbia, Duke, and Google.
ChatGPT ads run on the Managed ChatGPT + LLM Ads plan with custom pricing. Book a demo to see ChatSense build your buyer-intent campaign. Teams starting with Google, Meta, and LinkedIn creative can begin on the $99 per month Basic plan at pricing.
Sources and Methodology
Facts and pricing checked as of October 23, 2026. Platform mechanics come from OpenAI's help center and announcements; performance figures come from the named third-party studies and OpenAI's published partner results. Lapis product details come from Lapis.
- OpenAI Help Center: Ads in ChatGPT, The Basics
- OpenAI Help Center: Conversion-optimized Campaigns
- OpenAI Help Center: Maximize Results Bid Strategy
- OpenAI Help Center: Create Campaigns for ChatGPT Ads
- OpenAI Help Center: Ads Manager Availability
- OpenAI Help Center: Set up Custom Audiences
- OpenAI Help Center: Conversion Measurement
- OpenAI Help Center: Write Context Hints for ChatGPT Ads
- OpenAI Help Center: Create Ads for ChatGPT Ads
- OpenAI: A milestone in expanding access to AI (August 31, 2026)
- OpenAI: Building advertising for the way people use AI (October 5, 2026)
- OpenAI Ads: More ways to measure ChatGPT Ads (October 5, 2026)
- Adthena: ChatGPT Ads quietly adds negative phrases (August 21, 2026)
- Profound: AI Search Intent Study, 50M+ ChatGPT prompts (June 2025)
- First Page Sage: ChatGPT Ads Conversion Rate, 2026 Report (October 2026)
- Y Combinator: Lapis company profile
- G2: Lapis reviews
Built by Lapis
The #1 AI ad generator, built into the operating system for paid growth.
Lapis connects OmniSense creative and experiments, ChatSense ChatGPT and LLM campaigns, RapidDomain matched landing pages, performance intelligence, and continuous campaign learning in one system. Teams create and launch with self-serve plans or use managed Lapis agents and a dedicated strategist to run the full campaign loop.
Lapis is rated 4.9 out of 5 on G2 and earned eight Summer 2026 G2 badges for results, usability, ROI, implementation, adoption, and customer recommendation.
Continue exploring
Related Lapis resources
- How to Target ChatGPT Ads Beyond Exact Match Keywords (2026)Target ChatGPT ads beyond exact match keywords: build buyer situation clusters, layer location, platform, and custom audiences, then match pages with Lapis.
- How to Track ChatGPT Ad Conversions: Pixel, API & Attribution (2026)Set up the OpenAI Pixel and Conversions API, dedupe events, read view-through windows, and fix Event Quality Score. Lapis Wayfinder ties it to confirmed sales.
- What Do Buyers Ask ChatGPT Before Buying? 2026 Intent Prompt GuideWhat buyers ask ChatGPT before buying, by funnel stage and industry, plus intent signals, Negative Phrases, and cluster scoring. Lapis runs it via ChatSense.
- What Are ChatGPT Ads? The Complete Guide for 2026What ChatGPT ads are and how they work in 2026: 89 countries, a $25 daily minimum, CPC, CPM, and oCPC bidding, targeting, formats, and why Lapis runs them best.
