Most Shopify founders launching their first ChatGPT Ads campaign in 2026 lose the first two weeks to avoidable mistakes: bad category mapping, default merchant descriptions, budget set too high before the system has learned anything. This walkthrough is the sequence we use with new clients to skip those weeks.
Pre-launch checklist (do this before you install anything)
- Your Shopify store has at least 90 days of trading history
- You have an active Shopify Payments account (required for the native feed integration)
- You can identify your top 20 to 40 SKUs by revenue
- You have $1,500 to $3,000 budget for the first 30 days
- You have 4 to 8 hours of focused time in the first week for setup and creative
If any of these are missing, fix them first. Launching ChatGPT Ads without these conditions wastes spend in week one.
Step 1: Install the Shopify app and authenticate the feed
Search the Shopify app store for "ChatGPT Ads" (it appears as the OpenAI-built official app). Install, authenticate, and grant the product catalog read permission. The feed sync runs immediately and typically completes within 30 minutes for catalogs under 500 SKUs.
Verify the feed pulled correctly by checking the dashboard for product count, image presence, and price accuracy. If anything looks off, fix it in your Shopify admin and re-sync; the ChatGPT Ads dashboard always reflects the most recent sync.
Step 2: Identify and prioritize your top 20 SKUs
Open your Shopify analytics and pull top SKUs by revenue over the past 90 days. ChatGPT Ads works best when concentrated on proven sellers, not spread across your full catalog. Mark these 20 products as your initial campaign focus.
For each of the 20 SKUs, verify:
- The product has at least one clean square hero image (1:1 aspect, 1000x1000 minimum)
- The product description is at least 60 words of meaningful copy
- The price is competitive within the category (within 15 percent of category median)
- Inventory is healthy (no risk of stockout in the next 14 days)
SKUs missing any of these get fixed before they enter the campaign. Running a campaign on a SKU with a thin description or missing hero image just wastes engagement budget.
Step 3: Write merchant descriptions
For each of the 20 prioritized SKUs, write a fresh merchant description specifically for ChatGPT Ads placement. The format is 40 to 80 words, conversational tone, written as a knowledgeable recommendation rather than a marketing pitch.
What works:
- Open with the specific use case the product solves ("for runners who want...")
- Mention one or two specific attributes that differentiate it from category alternatives
- Reference the buyer pattern (who buys it, why)
- Avoid superlatives ("best," "amazing," "incredible") unless backed by specific data
What does not work: rewritten product page copy, generic marketing claims, bullet-point feature lists, anything that reads like an ad rather than a recommendation.
Step 4: Configure category mapping
In the ChatGPT Ads dashboard, map each of the 20 SKUs to the most specific applicable category. ChatGPT Ads category taxonomy mirrors Google Shopping with some additions for AI-relevant query types. Being too broad ("Home Goods" instead of "Home Lighting > LED Bulbs") is the most common cause of wasted spend in the first 30 days.
For products that fit multiple categories, pick the one most aligned with the buyer journey, not the broadest. A standing desk is "Home Office > Standing Desks," not "Furniture > Desks."
Step 5: Set bid strategy and launch budget
For the first 30 days, use "Maximize Engagements" bid strategy at the platform default bid. Resist the urge to bid above floor; the system learns better when you let it find the natural clearing price.
Launch budget: $1,500 to $3,000 for month one, distributed evenly across the 20 SKUs (not concentrated on one or two). This gives the system enough volume to learn while keeping spend on individual SKUs low enough that early mistakes do not blow the budget.
Step 6: Set up conversion tracking
Install the ChatGPT Ads pixel in your Shopify checkout. The Shopify app handles this automatically if you grant the required permission during install. Verify the pixel fires by completing a test order and checking the dashboard for the conversion within 15 minutes.
If you run multi-touch attribution (Triple Whale, North Beam, Rockerbox), add ChatGPT Ads as a tracked channel in your attribution platform now so the data builds correctly from day one.
Step 7: Launch and resist the urge to optimize
Launch the campaign and check the dashboard daily for the first three days, then every 2 to 3 days for the rest of the first 21 days. Do not make optimization changes before day 14. The system genuinely needs that time to learn, and changes made early reset the learning clock.
What to watch for in week one: any SKU getting zero engagements (likely wrong category mapping), any SKU burning more than 3x its category benchmark CPE (bid too aggressive or category mismatch), the overall pacing trend (under-pacing means bids too low; over-pacing means bids too high).
What week one mistakes look like
The mistakes we see most often in week one of new accounts:
- Launching with the default Shopify product description instead of a written merchant description
- Mapping all products to overly broad categories
- Starting at $500 monthly budget (too low to generate meaningful data)
- Starting at $10,000 monthly budget (too high before the system has learned)
- Pausing SKUs after day 5 because they have not converted yet (the conversion cycle on ChatGPT Ads is 5 to 15 days)
- Making bid changes daily based on noise rather than signal
Our team at ScaleWise VA runs ChatGPT Ads for Shopify clients from setup through scale. If you would rather have an experienced team handle the launch than learn the platform from zero, book a free 30-minute discovery call and we will scope the engagement specifically.