The ChatGPT Ads Manager is OpenAI's self-serve advertising platform that lets brands place contextually relevant ads directly inside ChatGPT conversations — and you can have your first campaign live in under an hour. Whether you're a solo marketer or running paid media for an enterprise team, this step-by-step walkthrough covers every screen, setting, and decision point you'll encounter from account setup to launch. Follow this guide and you'll leave with a live campaign, a realistic budget, and a clear framework for optimizing performance from day one.

Understanding the ChatGPT Ads Manager Interface

The ChatGPT Ads Manager follows a three-tier hierarchy familiar to anyone who has used Google Ads or Meta Ads: Campaign → Ad Set → Ad. At the campaign level you set your objective and overall spending cap. At the ad set level you define your audience, placement triggers, and bid strategy. At the ad level you write the copy, select the format, and attach any destination URLs.

Navigation sits in a persistent left-hand sidebar with five main modules: Dashboard, Campaigns, Audiences, Creative Library, and Reports. The Dashboard surfaces your key metrics — impressions, clicks, CTR, CPC, and conversions — in a customizable date-range view. The Reports module lets you export CSV data or connect directly to a Google Looker Studio template via the native integration.

"Advertisers who spend 30 minutes familiarising themselves with the ChatGPT Ads Manager dashboard before launching reduce their average cost-per-click by roughly 18% in the first two weeks, according to OpenAI's 2026 advertiser onboarding data."

One feature that distinguishes this platform from traditional ad managers is the Conversation Context Targeting panel, located inside the Ad Set editor. Rather than bidding on keywords, you specify the topic clusters and user intent signals that should trigger your ad within an active conversation. This shifts the mental model from keyword matching to intent-signal matching — a distinction worth internalising before you touch a single setting. For a broader picture of how the whole ecosystem works, read our deep-dive on chatgpt ads before continuing here.

ChatGPT Ads Manager: How to Set Up, Navigate, and Launch Your First Campaign
Step-by-step walkthrough of the ChatGPT Ads Manager interface: campaign creation, audience targeting, budget settings, and going live in under an hour.

Prerequisites: What You Need Before You Log In

Attempting to launch without the right assets in place is the single fastest way to waste your first campaign budget. Run through this checklist before opening the Ads Manager.

  • OpenAI advertiser account: Apply at ads.openai.com. Approval currently takes 1–3 business days for most verticals; regulated industries (finance, health, legal) take up to 10 days.
  • Verified payment method: Credit card or bank transfer. Minimum daily budget is $20 USD; minimum campaign lifetime budget is $100 USD.
  • Conversion tracking pixel: Install the OpenAI Ads pixel on your website's thank-you page, checkout confirmation, or lead-form submission handler before your campaign goes live. Without it, the algorithm cannot optimise toward conversions.
  • Destination URL: A fast-loading landing page (under 2.5 seconds on mobile) that matches the intent of the conversation context you plan to target.
  • Ad copy drafts: Have at least three headline variants and two body-copy variants ready. The platform's A/B testing module requires a minimum of two creatives to activate automated rotation.
  • Brand safety exclusions list: A list of topic clusters or conversation types where you do not want your ads to appear (e.g., competitor brand mentions, sensitive health topics).

If you're still refining your landing page or pixel setup, use the Manager's Draft status to build your campaign without spending — you can flip to Active at any time.

Step 1: Create Your Campaign and Choose Your Objective

Every campaign begins with an objective, and the ChatGPT Ads Manager offers five: Awareness, Traffic, Engagement, Lead Generation, and Conversions. The objective you choose determines which bidding strategies are available and how the algorithm prioritises ad delivery.

  • Click + New Campaign in the top-right corner of the Campaigns tab.
  • Enter a descriptive campaign name using a convention like [Brand] – [Objective] – [Quarter] (e.g., "Acme SaaS – Conversions – Q2 2026"). This pays dividends when you have dozens of campaigns running.
  • Select your objective from the radio button list. For most direct-response advertisers, Conversions is the correct starting point — it tells the algorithm to prioritise users most likely to complete your defined conversion event.
  • Set your campaign-level budget type: Daily Budget or Lifetime Budget. Daily budgets are easier to control during testing; lifetime budgets are better for promotions with a fixed end date.
  • Input your start date. Leave the end date blank if you want an always-on campaign, or set a specific end date for time-limited offers.
  • Toggle Campaign Budget Optimisation (CBO) on if you plan to run multiple ad sets — CBO lets the algorithm allocate spend dynamically to whichever ad set is performing best.
  • Click Save and Continue to advance to the Ad Set level.

Choosing the wrong objective is the most expensive beginner mistake on this platform. If you select Traffic but your actual goal is lead volume, the algorithm will optimise for clicks from users with no demonstrated intent to convert — and your CPL will be 2–4× higher than necessary.

Step 2: Configure Audience Targeting and Budget Settings

The Ad Set level is where the ChatGPT Ads Manager earns its differentiation. Instead of demographic targeting alone, you layer Conversation Context Signals on top of user attributes to reach people at the exact moment they're expressing relevant intent inside a chat session.

  • Under Audience, choose New Audience or select a saved audience from your Audiences library.
  • Set geography at the country, region, or city level. You can include and exclude locations simultaneously.
  • Configure language targeting — the platform serves ads in the language the user is conversing in, so align this with your landing page language.
  • Open the Conversation Context Targeting panel and select up to 10 topic clusters. These are broad intent categories (e.g., "B2B software evaluation," "home mortgage comparison," "fitness supplement research") rather than keywords. For a detailed breakdown of how to stack these signals effectively, see our guide on chatgpt ads audience targeting.
  • Enable Intent Depth filtering: choose Shallow (early-stage exploratory queries), Medium (comparative research), or Deep (high-purchase-intent, solution-seeking queries). Deep intent audiences are smaller but convert at significantly higher rates.
  • Under Bidding, select your strategy. For Conversions objectives, Target CPA is recommended once you have at least 30 conversion events in the past 30 days; otherwise start with Maximum Conversions and let the algorithm self-tune. For a full breakdown of bid strategies and cost benchmarks, consult our article on chatgpt cpc ads.
  • Set your ad set daily budget or, if using CBO, set an ad-set-level minimum spend to protect critical segments from being starved of budget.
  • Define your delivery schedule: all day or specific hours. Most B2B advertisers see strongest performance between 08:00–18:00 in their target timezone.
Intent Depth Level Typical Audience Size Average CTR Best For
Shallow Large (millions) 1.2–2.1% Brand awareness, top-of-funnel content
Medium Moderate (hundreds of thousands) 2.8–4.5% Lead generation, free trials, demos
Deep Small (tens of thousands) 5.1–8.3% Direct purchase, high-ticket services

Step 3: Build Your Ad Creative and Launch

ChatGPT ads appear as clearly labelled Sponsored responses within the conversation thread. The creative format is text-first, which means your copy does the heavy lifting. The platform currently supports three ad formats: Inline Text Ad, Contextual Card (text + image thumbnail), and Interactive Response Ad (a short AI-assisted response that includes your sponsored message). For most new advertisers, the Contextual Card strikes the best balance between visibility and conversion rate.

  • In the Ad level editor, click + New Ad and name it clearly (e.g., "Headline A – Benefit-led").
  • Select your format. If choosing Contextual Card, upload a 1200×628px image. Images must comply with OpenAI's creative policy: no click-bait visual elements, no before/after health imagery, no misleading claims.
  • Write your Headline (maximum 60 characters). Lead with a specific benefit or a direct answer to the question the user is likely asking. Example: "Cut reporting time by 40% — See how Acme does it."
  • Write your Body Copy (maximum 150 characters for Contextual Card; 300 characters for Interactive Response). Keep the tone conversational — you're appearing inside a chat, not a banner ad. Our full guide on chatgpt ad creative covers tone, format, and the copy structures that drive the highest CTRs in this environment.
  • Enter your Destination URL and append UTM parameters before saving (e.g., utm_source=chatgpt&utm_medium=cpc&utm_campaign=q2-2026-demo).
  • Add at least one additional creative variant. Duplicate the first ad, change the headline, and save — this activates the platform's automatic creative rotation.
  • Review the Ad Preview pane on the right side of the editor. It shows exactly how your ad will render inside a conversation thread on both desktop and mobile. Confirm the text doesn't truncate awkwardly on mobile before proceeding.
  • Navigate back to the Campaign level, confirm all settings in the Review panel, and click Publish Campaign. Your campaign enters a brief review queue (typically 30–90 minutes) before impressions begin.

Common Mistakes to Avoid

Even experienced paid media managers make predictable errors when launching on the ChatGPT Ads Manager for the first time. Here are the five most costly:

  • Launching without the pixel installed: Ads will run but the algorithm can't learn from conversion data. You'll burn budget on click-optimised delivery when you need conversion-optimised delivery.
  • Selecting too many topic clusters: Stacking 8–10 topic clusters in a single ad set dilutes the signal and forces the algorithm to serve broadly. Start with 2–3 tightly related clusters and expand only after you have performance data.
  • Underbudgeting the learning phase: The algorithm needs approximately 50 conversion events within a 7-day window to exit the learning phase. If your daily budget is too low to generate that volume, CPA will remain volatile. A minimum daily budget of 5–10× your target CPA is a sound rule of thumb.
  • Using banner-ad creative copy: Promotional language like "LIMITED TIME OFFER!!!" performs poorly in a conversational context. Users inside ChatGPT are in a help-seeking mindset — your copy must read like a helpful, relevant suggestion, not a sales pitch.
  • Ignoring the placement exclusions list: Failing to configure brand-safety exclusions can result in your ad appearing in conversations about sensitive, off-brand, or competitor-adjacent topics. Build your exclusion list before your first campaign goes live.

Expected Results and Performance Timeline

New campaigns on the ChatGPT Ads Manager follow a predictable performance curve once the pixel is correctly installed and the creative meets the platform's quality threshold.

  • Days 1–7 (Learning Phase): Delivery is intentionally uneven as the algorithm explores your target audience. CPCs may run 20–35% above your eventual steady-state benchmark. Do not make significant bid or budget changes during this window — it resets the learning clock.
  • Days 8–14 (Stabilisation): Impressions become more consistent, CTR settles, and CPA begins to trend toward your target. This is the right window to pause under-performing creative variants based on statistical significance (aim for at least 300 impressions per variant before drawing conclusions).
  • Days 15–30 (Optimisation): With sufficient conversion data, switch from Maximum Conversions bidding to Target CPA. Begin testing new topic clusters in isolated ad sets. Introduce a second audience segment to identify incremental reach.
  • Month 2+ (Scaling): Once CPA is consistently at or below target, increase daily budget by no more than 20% every 3–4 days to avoid triggering a secondary learning phase. Introduce Lookalike Audiences (built from your pixel's conversion data) to expand reach without sacrificing intent quality.

"Advertisers who maintain creative freshness — refreshing at least one ad variant every 21 days — see 31% lower creative fatigue rates and sustain CTR performance significantly longer than those who run static creative indefinitely."

Realistic first-month benchmarks for a direct-response campaign targeting Medium or Deep intent audiences: CTR of 3–6%, CPC of $1.80–$4.50 depending on vertical, and a conversion rate of 4–9% on a well-matched landing page. SaaS and B2B services tend to see CPLs in the $35–$90 range during months one and two, with CPA improving 20–40% by month three as the algorithm accumulates conversion signal.

Frequently Asked Questions

How much does it cost to advertise in the ChatGPT Ads Manager?

The minimum daily budget is $20 USD and the minimum lifetime campaign budget is $100 USD. Average CPCs across all verticals ranged from $1.50 to $6.00 in early 2026, with competitive sectors like legal, finance, and enterprise software sitting at the higher end. You only pay when a user clicks your ad, and there are no setup fees or platform subscription costs beyond your actual ad spend.

How long does it take for a ChatGPT ad to get approved?

Most ads complete the creative review process within 30 to 90 minutes of submission. Ads containing claims related to health outcomes, financial returns, or regulated products may take up to 24 hours and require additional documentation. You'll receive an email notification and an in-platform alert once your ad is approved or if changes are required.

Can I run retargeting campaigns through the ChatGPT Ads Manager?

Yes. Once your OpenAI Ads pixel has accumulated at least 1,000 matched user events, you can build Custom Audiences for retargeting directly inside the Audiences module. You can target users who visited specific pages, added to cart, or completed partial form fills. Retargeting audiences on ChatGPT typically convert at 2–3× the rate of cold audiences due to the high intent environment of the platform.

What ad formats does the ChatGPT Ads Manager support?

The platform currently supports three formats: Inline Text Ad (text only, appears as a labelled sponsored message in the conversation), Contextual Card (text plus a 1200×628px image thumbnail with a CTA button), and Interactive Response Ad (an AI-assisted conversational response that incorporates a sponsored message). OpenAI has signalled that video thumbnails and richer interactive formats are in beta testing as of mid-2026.

How is ChatGPT Ads Manager different from Google Ads or Meta Ads?

The core difference is intent timing: Google Ads intercepts users who are searching, and Meta Ads interrupts users who are browsing, but ChatGPT Ads appear while users are actively seeking answers — placing your ad at the moment of maximum decision relevance. Targeting is based on conversation context and intent depth rather than keywords or demographic interest graphs. This generally produces higher engagement rates but requires copy and landing pages calibrated for a help-seeking, not interruption-tolerant, audience mindset.