TikTok Shop now lists up to ten reasons a product's sales fell. Give each kind an owner before Black Friday
Your best seller on TikTok Shop dropped last week, and Seller Center can now tell you why in up to ten reasons. Some of them are a stock problem, some are a creator who stopped posting, and some are your product page. Each lands on a different desk.

TikTok Shop's Product Diagnosis, documented for U.S. sellers on November 12, 2026, flags products whose sales are falling and lists up to ten ranked reasons, split into blocking risks, changes against your own previous period, and gaps against top products in the category. It works from offline data that can lag, and TikTok calls the AI output reference only. Before peak, decide who owns each kind of reason and how often someone reads the list.
What TikTok Shop added
TikTok Shop has added Product Diagnosis to Product Analytics in its U.S. Seller Center. PPC Land reported on it on November 12, quoting a TikTok Shop Academy guide updated that day. The guide says the feature launched in November and is available in the U.S.
The guide frames it as three questions: which product needs attention, why its sales moved, and what to do next. In practice that means four changes.
Products whose sales are falling get a Need attention flag. Each product gets a diagnosis view showing how its sales moved, which sales channels moved them, and the factors that mattered most, with a "Chat with AI" assistant. Every key metric gets a benchmark against both your own previous period and the top-performing products in the same category. And each factor links to the Seller Center page where you'd fix it: listing, stock, price, ratings, the affiliate program and account health among them.
No new data feeds it. TikTok's guide says it's the same product analytics data, organized around a diagnosis.
How to read the ten cards
The diagnosis summary shows up to ten factor cards. TikTok's guide sorts them into three types.
Risk cards are blocking issues: listing status, stock and violations. The guide says to address these first, and it's right. A product that's out of stock or under review can't be rescued by better videos.
AA cards are changes against your own previous period of the same length. Fewer creators posting, fewer of your own lives or videos, a price change.
AB cards are gaps against the benchmark: where top products in the category convert or sell better than yours. Weak click-through points at the product card. Weak click-to-order points at the product page and checkout. A weak average order points at the mix of variants.
Each product's sales are also split across five channels: Creator LIVE, Creator Video, Seller LIVE, Seller Video, and Shop Tab and Search. That split is the most useful part. A product whose sales fell because two creators stopped posting looks different from one whose search conversion slipped, even when the headline drop is the same.
Decision guide
Three kinds of reason, three different desks
| Consideration | What the card says | Who should pick it up |
|---|---|---|
| Risk | Blocking issues: listing status, stock, violations | Whoever runs listings and inventory, same day |
| Change vs your last period | Fewer creator posts, fewer of your own lives or videos, a price change | Creator or content lead for posting; whoever set the price for price |
| Gap vs category leaders | Weaker click-through, click-to-order or product mix than top products | Whoever owns the product page, images and variants |
| No flag at all | No confirmed decline, which TikTok says isn't the same as verified good | Nobody yet; check again on your usual day |
A diagnosis that nobody owns is just a more detailed way of watching sales fall.
Give every kind of reason an owner
Here's the trap. A small TikTok Shop team opens the diagnosis, sees ten sensible cards, agrees they all matter, and nobody does anything by Friday. A diagnosis that nobody owns is just a more detailed way of watching sales fall.
So before peak, decide who picks up each kind of card.
Risk goes to whoever runs listings and inventory, the same day. Stock and listing status are usually one person's job already. Make it explicit that a Risk card on a top product is today's work.
Changes in creator activity go to whoever manages creators and affiliates. If a product sold mostly through Creator Video and that channel fell, the fix is a conversation with creators, samples or the affiliate program, not a new product photo.
Changes in your own content go to whoever runs your lives and videos. Posting cadence first, then which products you feature.
Price changes go to whoever set the price. If a promotion ended and sales dropped, that may be expected. The person who ended it should say so.
Benchmark gaps go to whoever owns the product page. Title, images, variants, price display. These are slower fixes, so they belong on a list with dates, not in a panic.
Here's how that might play out for one illustrative product. A best-selling lip oil drops. The diagnosis shows a Risk card for low stock in one shade, a period card showing Creator Video sales fell, and a benchmark card showing click-to-order trails the category. Inventory reorders the shade the same day. The creator lead finds that two affiliates stopped posting and sends samples. The person who owns the listing adds the shade swatches the page was missing. Three people, three fixes, one product, and nobody waiting for anybody else.
Our note on TikTok Shop's score knowing which product hurt you makes the same point about another TikTok metric: the tool can point; someone still has to go.
What to treat with care
TikTok's guide is candid about the limits, and they're worth taking seriously during the busiest weeks of the year.
The data can lag. The guide says diagnosis relies on offline data, so it can trail the real-time numbers elsewhere in Seller Center. It doesn't say by how much. With Black Friday on November 27, a delayed signal matters more than usual. Check a surprising diagnosis against your live numbers before acting.
The benchmark isn't yours to inspect. The comparison group is platform-defined: top-performing products in the same category, with no way to change it. The guide doesn't say how "top" is chosen. Use it as a direction, not a target.
The AI is reference only. In TikTok's own words, AI-generated content is for reference only and should be checked against page metrics and your own business context before action.
No flag doesn't mean fine. The guide says a product without a flag has no confirmed decline, which isn't the same as verified good. A product can be slipping slowly without tripping the two-part test.
Not everything has a fix page. Some factors describe a condition, like how long a product was out of stock, rather than a page to edit.
If you're wondering whether to keep paying for a separate analytics app, our note on checking TikTok's Compass reports first is the companion to this one.
How the work moves
A twice-weekly read through peak
- Open Need attention
Start with the product with the biggest drop, which is where Product Diagnosis opens by default.
- Clear Risk first
Stock, listing status and violations block everything else.
- Hand out the rest
Each remaining card goes to its owner with a date.
- Check against your own numbers
Offline data can lag; compare before acting on a surprise.
- Mark what was wrong
Use the not-useful feedback when attribution looks off, and note it for next time.
Before Black Friday
Write the owners down: one name per kind of card. Pick two fixed times a week through peak to open Need attention, starting with the biggest drop. Clear Risk the same day. Hand out the rest with dates. When a diagnosis turns out wrong, mark it not useful and note why, so the team learns which cards to trust.
If the real problem is that stock, listings and creator programs live in different tools and nobody sees them together, that's the operating side of selling on a marketplace, and it's our marketplaces work. Talk to us.
Related work: Paddle8
We build alongside the people doing the work.
Bring us the problem you’re thinking about.