TikTok Shop search volume: why nobody can tell you what it is
5 min read
If you have used a TikTok Shop research tool, you have seen a search volume column: "collagen powder — 74,000/mo". It is worth knowing where that number comes from, because the honest answer is that it is a model output, not a measurement.
What TikTok actually publishes
For TikTok Shop search specifically, the surfaces available to anyone outside ByteDance publish no volume figure at all:
- Related searches return phrases. Names, not numbers.
- Search autocomplete returns suggested strings. No counts attached.
- The results page returns listings, positions and product metadata. Again, no query volume.
TikTok's advertising products expose audience and interest sizing, and the Creative Center exposes hashtag and video trends — but neither is TikTok Shop search volume. They are different surfaces measuring different behaviour, and mapping one onto the other is a modelling exercise.
So where do the numbers come from?
Three routes, roughly in order of how commonly they are used:
- Extrapolation from a panel. Observe a sample of behaviour, scale it up to an assumed total. The scaling factor is the whole answer, and it is a guess.
- Borrowing from another platform. Take a keyword's known volume elsewhere and apply a TikTok multiplier. This assumes people search the same things in the same proportions on a video app as they do in a web search box, which is not obviously true.
- Inference from product-level signals. Work backwards from sales, sold counts and engagement on the listings that rank for a phrase. Reasonable, and still an estimate with a wide interval.
None of these are frauds. All of them are estimates presented with more decimal places than they have earned. A number written as "74,000" implies a precision that a panel-scaled figure simply does not have.
Why we do not show one
Vantage does not display a search volume column, and that is a deliberate omission rather than a missing feature. The product's whole claim is that its numbers are measured — where your listing sat, when, and next to whom. Putting an estimated figure in the same table, in the same typeface, next to measured positions would quietly teach you to trust both equally.
If a reliable figure becomes available, it belongs in the product. Until then, an absent column is more useful than a confident guess.
How to prioritise searches without volume
The good news is that volume was never the most useful input anyway. What actually decides where your attention goes:
Where you already sit. A search where you are #11 is worth more attention than one where you are #40, regardless of how many people type each one. Moving from #11 to #8 is achievable and lands you inside the band shoppers actually read. Moving from #40 to #8 usually is not.
How specific the phrase is. "Protein powder" is enormous and brutally contested. "Chocolate whey isolate" is smaller and converts better, because somebody typing it has already decided what they want. Long, specific phrases are the ones a small seller can win, and their lower volume is the reason they are winnable.
Whether the phrase describes your actual product. The cheapest ranking gain available to most sellers is dropping the phrases their listing was never really an answer to, and concentrating on the ones it is.
What your customers tell you. Read your own reviews, questions and comments. The words buyers use for your product are the words they search for, and no tool has access to that better than you do.
Your own conversion data. Which searches send traffic that buys is a fact you can observe, and it beats any external estimate of how many people typed the phrase.
The one number worth watching
Rather than volume, watch where you sit across the searches you have chosen, over time. That is measurable, it is specific to your shop, and it moves when something real happens.
Vantage checks the searches you pick once a day and records your position on each, so the trend is a record of what actually happened rather than a projection of what might be happening.