TikTok analytics explained: where each number lives and when to trust it

Where is TikTok Analytics on a phone and on a computer?

Analytics has three working routes and one dead end. The phone app and two desktop addresses reach it. Mobile web has no Analytics.

The order matters for a laptop user. Turn on happens only in the mobile app, so a creator who plans to read numbers at tiktok.com/analytics still starts on the phone. Data then begins with the next post: switch on at noon on Monday, and the 7 p.m. and 9 p.m. posts below get numbers while Friday’s clip stays at No data.

What does TikTok Analytics show?

Analytics has two levels. TikTok describes the account overview as weekly and monthly views, follower growth and trending videos. A single post’s analytics go further: total play time, average watch time, watched full video, a retention chart and traffic sources. Those are TikTok Studio’s own labels, and this page uses them as written.

Start with the post. The overview tells you a week went well; the post page tells you which video carried it and the second where viewers left. A retention chart that falls hard in the opening seconds is a different problem from one that sags halfway, and the guide on how to read a retention curve names the edit for each shape.

TikTok dates Analytics in UTC and needs about two days to settle

Analytics shows every figure except Follower activity in UTC, and TikTok says the data takes about 1 to 2 days to update. TikTok’s help page for the in-app route adds that data may be delayed depending on the post, and suggests waiting a few days. Two effects follow.

Take a creator in New York in October, when the city runs four hours behind UTC (UTC−4). Two posts go up on the same Monday:

• 7 p.m. local: 19:00 + 4 h = 23:00 UTC, still Monday.
• 9 p.m. local: 21:00 + 4 h = 25:00, which rolls past midnight to 01:00 UTC on Tuesday.

The 9 p.m. post is reported under Tuesday, and every view from its first New York evening, 9 p.m. to midnight, is dated Tuesday in UTC (01:00 to 04:00). The 7 p.m. post gets one hour on Monday, from 23:00 to midnight UTC, and the rest of its evening on Tuesday. Two posts the creator thinks of as Monday posts end up filed under two different days.

Then add the delay. The 9 p.m. post’s Tuesday figures need about 1 to 2 days to update, so read them on Wednesday or Thursday. A check on Tuesday morning sees part of the evening, or No data.

Follower activity is the exception: the only figure in local time. In this example its hours are New York hours and need no conversion. If Follower activity peaks at 9 p.m., that means 9 p.m. in New York, the same clock the creator posts by.

Each question you bring to Analytics has one number that answers it

Five questions cover most visits, and the fifth is answered outside Analytics. Read the row before opening the app.

Row five is different. TikTok’s help pages put restrictions in Account check, a separate screen under More tools, and in the “Ineligible for recommendation” notice below the username. A falling For You share in traffic sources can hint at one. Only those two screens confirm it.

Go Viral estimates the retention shape before Analytics can

Go Viral gives an estimate for the first row of that table before the post exists. Its Engagement Heat Map marks likely drop-offs, replays and retention trends on the clip, and Hook Score rates the first 3 seconds from 0 to 100. TikTok’s own retention chart for the 9 p.m. Monday post arrives on Wednesday at the earliest, once its 1 to 2 days of updating have passed.

Import the file or paste a TikTok, Reels or Shorts link. The app works on the video alone, with no login to TikTok and no access to Analytics. Keep both readings. On Wednesday, put the Monday estimate beside the TikTok Studio chart.