Find the photos your product page is missing
Tolstoy reviews the images on one product page and returns up to five missing shots, in priority order, each tied to a question shoppers cannot answer yet.
What you need
- One product from your connected catalog, with its current images and product details
What you get
- The purchase questions your images answer, and the ones they leave open
- Up to five missing shots in priority order, with view, framing, input and placement
- One recommended shot to make next
How it runs
Tolstoy runs this playbook in 4 steps.
-
Select one product
You pick the product page to review, and the agent reads its current images and catalog details.
-
Map the shopper questions
The agent lists the purchase questions your images answer, such as size, material or use, and the important ones they leave open.
-
Rank the missing shots
You get up to five missing shots, most useful first, each with its view, framing, required input and place on the page.
-
Pick the next shot
The audit ends with the single shot to make next, and keeps what the images show apart from what is uncertain.
Check the result
- Each recommended shot answers a question your current images leave open, not one they already answer.
- Each shot names a view, framing and input specific enough to brief a photographer or the agent.
- Statements about the product come from your images or product data, and anything uncertain is marked as uncertain.
Know the limits
- It returns a shot list, not images. Make the shots in the same chat or with another playbook.
- It does not read store analytics or conversion data. It judges what a shopper can see, not how the page performs.
- One product per run.
When to use this playbook
Use it before you spend money on new product content, to decide which image to make first. It helps most on pages that get many sizing or “how does it look in use” questions, and on pages that still show only supplier packshots. For the other gaps that cost a product page sales, read five common ecommerce product page mistakes.
Get better results
- Complete the product description, materials and sizing first. Thin product data means more open questions.
- Tell the agent who buys the product. A gift buyer and a repeat buyer ask different questions.
- Make the top recommended shot, add it to the page, and run the audit again.
The prompt
“Run this playbook” opens Tolstoy with this prompt as a draft. Pick your products, edit anything, then send it.
Review the current images and product information for my selected product. List the purchase questions those images answer and the important questions they leave open. Recommend up to five specific missing shots, ordered by usefulness, with the view, framing, required input and intended placement for each. Separate what the images show from what is uncertain. Do not generate a whole campaign or invent conversion data. End with the single most useful shot to make next. If no product is selected, ask only for that. Keep the real product shape, proportions, construction, print, color, logos and readable labels intact. Do not invent product features, claims or reviews.
FAQ
What should I consider before using AI to update product page imagery?
Start with what shoppers cannot see yet, not with what AI can make. List the purchase questions your current images leave open, such as size, material or use, and make the shot that answers the most important one first. Every new image must show the product exactly as you ship it: shape, color, print and label text.
Why only five shots?
A short ranked list tells you where to start. When you have made the top shots, run the audit again to see which questions are still open.
Can I audit a whole collection?
The playbook reviews one product per run, because each product raises different questions. Start with the products that sell most, then check whether the same gap appears across the collection.
Run it on your own catalog
Start free, connect your store, and Tolstoy opens this playbook for you. You review every result before anything goes live.