TikTok Studio and YouTube Studio analytics
Short version: both are free, both are first-party, and for your own account both are better than anything you can buy. Use them. The wall is not their quality, it is their scope: they cover your account and nobody else’s. And on TikTok specifically, the data about other people’s videos that most tools imply they have does not exist in any form a commercial product can legally obtain.
This is the sibling of the same question for Instagram. Same answer, different platform.
Start with the free tools, genuinely
TikTok Studio and YouTube Studio are where your own numbers live, and they see things no third party can. Retention curves for your own videos, traffic sources, the drop-off shape. Nothing you pay for improves on first-party access to your own account.
Anyone telling you their dashboard shows more about your own account than the platform’s own tool is describing a nicer layout, not more data.
What YouTube documents, and one number that changed
YouTube Analytics exposes estimatedMinutesWatched, averageViewDuration, averageViewPercentage, subscribersGained, shares, likes and comments, all defined in Google’s docs and all measured “for the specified channel”, which is to say yours.
One worth knowing: since 13 December 2021, both averageViewDuration and averageViewPercentage exclude looping clips traffic. If a number looks lower than you remember, or lower than a third-party report claims, that exclusion is often why. Two measurements, one label.
Nobody has TikTok retention for an account they do not own
This is the part worth reading twice, because it decides what you should believe when you shop.
TikTok’s Research API returns view_count, like_count, comment_count, share_count, favorites_count and video_duration, along with descriptive fields like hashtags and effects. It returns no watch time, no retention and no completion rate. Those numbers are simply not in the response.
And that API is not available to companies at all. Eligibility is limited to academic institutions in the U.S., EEA, UK, Canada or Switzerland, not-for-profit or independent research bodies in the EU, and academic or not-for-profit organisations in Brazil. Applicants must submit a research proposal, disclose funding, and show the work has passed an ethical review. Research has to be conducted on a not-for-profit or non-commercial basis.
So when a paid tool shows you a competitor’s retention curve, it is inferring it. That includes us. Silma reads public counts for accounts it does not own, the same as everyone else. Any product claiming measured retention on somebody else’s TikTok is claiming access that TikTok does not sell.
What to ask a TikTok analytics tool before you pay
- Where does this number come from? For your own account, first-party connection. For anyone else’s, public counts. There is no third option.
- Is retention on a competitor measured or modelled? It is modelled. A tool that will say so is being straight with you.
- What does it add that TikTok Studio does not? If the answer is a nicer chart of your own data, you already have the data.
What Silma adds, and why it is not more API access
Silma watches the video. Reading the frames and hearing the words needs no API and no permission, which is why it works identically on an account you do not own and on your own. That is the half the platform tools cannot do at any price, because they count outcomes rather than inspecting the thing that produced them.
On top of that it ranks posts by how far each beat that account’s own median view count, which is computable from the public counts that exist. A small account’s 4x post is a better lesson than a big account’s ordinary day, and raw view counts cannot tell those apart. That is what Discover and competitor analysis are built on.
What it will not do is show you a retention curve for a video it has no retention data for.
When the free tools are enough
- You post on one account and want to know how it did.
- You are not benchmarking against anyone else in a way that has to hold up.
- You are content to learn after publishing rather than before.
If that is you, open Studio. You will get more from an hour there than from any subscription.
When they stop being enough
- You need to explain why a video worked, not just that it did.
- You are learning from other people’s posts, which the native tools do not cover at all.
- You need a read on a draft before it is public.
- Somebody is paying you, and “our own account went up” is not evidence.
What Silma costs, plainly
$9 a month for Lite, 200 credits, roughly 40 pre-checks or 16 full breakdowns of a new video. No free tier and no trial. TikTok Studio and YouTube Studio remain free and remain better at reporting your own account.
The native tools count. This one watches. Paste a TikTok or a Short, or upload a draft before it goes out. $9 a month, cancel any time. Check your first video →
Questions
What is the best free TikTok analytics tool?+
TikTok Studio, the one built into the app. It is first-party, so for your own account it sees retention and watch data that no third-party tool has access to. Every free third-party TikTok analytics tool is working from public counts, because that is all TikTok exposes. Start with Studio.
Can any tool show me retention on someone else’s TikTok?+
No, and be suspicious of any that claims to. TikTok’s Research API returns view, like, comment, share and favourite counts plus video duration. It returns no watch time, no retention and no completion rate. That API is also restricted to academic institutions and not-for-profit research bodies in eligible regions, and requires a research proposal, funding disclosure and ethical review, so commercial tools cannot use it at all. Anything you are shown about a competitor’s retention is inferred, not measured. That applies to Silma too.
Why does my YouTube average view duration look lower than expected?+
Since 13 December 2021, YouTube’s averageViewDuration and averageViewPercentage metrics exclude looping clips traffic, per Google’s own documentation. If you are comparing against numbers from before that date, or against a third-party report that does not apply the same exclusion, you are comparing two different measurements.
Does Silma have a free plan?+
No. There is no free tier and no trial. Lite is $9 a month. TikTok Studio and YouTube Studio cost nothing and are better than Silma at reporting your own account, so if that is the job, use them.
So what does Silma actually add on TikTok?+
It watches the video. Reading the frames and hearing the words needs no API access and no permission from the platform, which is why it works the same on an account you do not own as on your own. It also ranks posts by how far each beat that account’s own median view count, which is computable from the public counts that are available. What it does not do is pretend to know retention on somebody else’s post.
Sources
- TikTok Research API returned video fields, and the absence of watch time, retention or completion rate: TikTok for Developers, Research API query videos, read 9 September 2026.
- TikTok Research API eligibility, the exclusion of commercial entities, and the proposal, funding-disclosure and ethical-review requirements: TikTok for Developers, Research API, read 9 September 2026.
- YouTube Analytics metric definitions, per-channel scope, and the exclusion of looping clips traffic from
averageViewDurationandaverageViewPercentageas of 13 December 2021: Google, YouTube Analytics metrics, read 9 September 2026. - Silma plan limits and credit costs: silma.app/pricing.
Claims we could not source and therefore did not make: what TikTok Studio shows in-product beyond your own retention curve, a benchmark for a good completion rate, and any figure comparing paid tools to the native ones. Found something out of date? Email support@silma.app.