2026-08-04 · 7 min read

TikTok says 10,000 visits, your store says 900

Short answer: the two numbers measure different things. The platform counts clicks; your store counts visits that actually arrived and loaded. Four things live in the gap: clicks that never landed, clicks that were not people, visits that landed but were never recorded, and attribution windows that credit people who never clicked at all. A gap of 10 to 30% is normal. A gap of ten times is a fault you can find today.

How merchants describe it

In May 2026 a Saudi practitioner posted on X, and it drew 111 likes and ten reposts: “If you ever open your TikTok ad account and find the platform recording 10,000 visits against a chunk of your daily budget, but your store actually has under 1,000 visits from every platform combined, then you have been defrauded outright by the platform.”

He went on to describe a pattern many advertisers recognise: on certain days of the month, minutes after midnight, most of the daily budget drains, the clicks column shows a large number, and the sales column shows zero.

The first reply came from an agency: they see it constantly, and it puts them in an awkward position in front of their client. The second reply matters more, because it comes from an ordinary shopper describing the same thing from the other side of the screen. She scrolls TikTok, an ad appears, she does not tap it, and suddenly she is in an online store she never chose to visit, so she leaves immediately.

The thread’s author also claimed roughly a quarter of platform traffic is fake. That is his estimate, not an independently verified figure, and we do not build on it. But the description itself, the gap between the platform’s number and the store’s number, recurs often enough to deserve a precise explanation.

The four causes, largest first

A click is not a visit. The platform records the click the moment a finger lifts. The visit is recorded when your page actually loads. Between those two moments sit weak connections, people leaving before the page renders, and accidental taps on an ad placed mid-content. That alone accounts for 10 to 20% in a perfectly healthy account.
Invalid traffic. Some share of clicks does not come from people who buy. This is an industry-wide problem, every platform carries some of it, and the platforms themselves acknowledge it and credit part of it back later. But the credit arrives after you have already read the report and made your decision.
The browser blocked the recording.The visit genuinely arrived but was never logged on your side: an ad blocker stopped the script, the user declined tracking, or the browser shortened the cookie’s life. The visit is real and your number is short. This one makes your store look worse than it is.
The attribution window.The platform credits itself for people who saw the ad without clicking, if they buy within the set period. The second reply above describes exactly that from the shopper’s side: she arrived without tapping, and left.

The first three you can check today. The fourth is a settings decision that belongs to you.

The normal gap, and the gap that should stop you

Under 10%. A healthy setup.
10 to 30%. Normal, and where most accounts sit.
30 to 60%. A measurement fault. Start with cause three.
Over 60%. Do not read the report at all until it is fixed.

The number that actually matters is neither clicks nor visits. It is orders against spend, and that one is untouched by all four causes above, because it comes from your store.

Where MIQAS comes in

MIQAS measures both sides and puts the two numbers next to each other. On your store’s side, the first script records the visit and the click identifier that arrived with it, so you know how many visits genuinely landed, rather than how many clicks were claimed.

On the platform’s side, purchase events are sent to the TikTok Events API from the server directly, carrying the same order ID you hold. Server-side sending is unaffected by an ad blocker or a browser setting, so cause three is removed at the root, and the order ID stops the sale being counted twice.

In the dashboard you see spend and real orders per channel in one place, so the question changes from “who is lying, the platform or the store?” to “how many orders did TikTok bring me this week, and at what cost?”. TikTok is available on Starter alongside Meta. Snapchat, Google and X arrive with Growth and above.

Check it yourself, in ten minutes

Open the platform’s report for the last seven days and note the clicks.
Open your store analytics for the same period and note visits from that source.
Divide the second by the first. That is your ratio.
Open Events Manager and look at the event source column. If everything says “Browser” only, cause three is running in your account right now.

Common questions

So the platform is lying? No. The platform is answering a different question. It counts clicks on its ad; you want to know how many people arrived and bought. Both numbers are correct in their own place. The error is comparing them as though they were the same number.

Can I claim budget back for the bad days? Some advertisers say they asked platform support for a review and recovered amounts. Those are individual published accounts, not a stated policy, and we promise no outcome.

Should I switch TikTok off? Do not kill a channel over a measurement gap. Fix the measurement first, then judge the channel on orders rather than clicks. Plenty of channels look better than they did once measurement settles.

The gap itself is not evidence of fraud, and it is not evidence of success either. It is evidence that you are reading two different numbers as one. The first step is seeing them apart.