How the TikTok side works
What was deliberately not built
A feature list read backwards. Nine mechanisms that exist on the Instagram side, or that any competitor would ship, are absent from the TikTok side — each for a stated reason. The shape of what somebody chose not to build says more about their judgement than the shape of what they did.
Nine things that were not built
A follow exchange
The single most detectable exchange on TikTok and the one most clearly named in policy. Reciprocal-follow clusters are trivial to find in a bipartite graph, and the damage is durable in a way a bad comment is not — a follower who never watches is a permanent edge that depresses your Following-feed performance and skews what the platform thinks your audience is. PodSwap's own risk assessment puts the chance of a participant-visible penalty inside twelve months at 30–50%. It is the easiest exchange to build and it is not being built.
A like exchange
Likes are the most heavily instrumented fake-engagement surface on the platform and among the weakest ranking signals. The product plan describes an unpaid, optional liking step inside the watch flow — "like it if you genuinely enjoyed it" — with the resulting like rate deliberately kept realistic, because most real viewers of a video do not like it. That step is not in the shipped code either: there is no like anywhere in the TikTok flows today.
Search-and-find delivery
Sending members to find your video through TikTok search would be the highest-value action in the category if it worked. It cannot be built honestly: search is not a deep-linkable path on iOS, so there is no way to route someone there, no way to verify they arrived, and no way to know the video was found rather than pasted. The idea rests on an inference nobody has demonstrated.
Share-to-DM
The share sheet is not deep-linkable — the messages path opens an inbox, not a composer. DM eligibility is also age-gated and restricted by default to mutual follows, so a meaningful share of members could not perform the action at all.
Following-feed delivery
A common piece of advice is to have members engage through the Following tab rather than through links. There is no route to that tab, no way to verify it happened, and no way to measure it. It is advice, not a feature, and it is not on any roadmap here.
A sound or duet exchange
Using someone else's sound is a real distribution signal, but it asks a creator to make a whole video to someone else's brief. That is production work, not an engagement action, and pricing it against a save would be dishonest.
A TikTok health or shadowban checker
Specified in detail and deliberately not shipped. An honest version needs daily per-video snapshots over months and a cohort of comparable accounts to normalise against — without both, it produces confident nonsense. Instagram's scanner exists because that history exists on the Instagram side. It does not on the TikTok side yet, and a checker built without it would be the exact kind of tool these pages criticise.
A TikTok creator directory
The Instagram directory is built out of profiles that have been through the exchange. There are none on the TikTok side, so a directory would be an empty page or a fabricated one.
TikTok tools on the free side of the site
The Podscore rubric, the shadowban scanner, the hashtag density report and the viral predictor are all calibrated on Instagram data. Re-labelling them for TikTok would be a lie about what they measure.
The follow exchange is the one that matters
It is the easiest of them all to build, the easiest to verify, and the one every competitor offers. It is not being built.
Reciprocal-follow clusters are trivially detectable in a bipartite graph, the behaviour is named directly in TikTok's guidelines, and the damage is durable in a way a bad comment is not — a follower who never watches is a permanent edge that depresses your Following-feed performance and skews what the platform believes your audience to be.
The internal estimate is a 30–50% chance of a participant-visible penalty within twelve months. That is the highest figure attached to any of the six actions assessed, and it is why the answer to "why is there no follow option" is not "later".
The honest positioning is that on TikTok, a follower who does not watch is worse than no follower.
And the honest limit of the whole thing
The premise itself is unproven. It is genuinely unknown whether deep-linked, off-feed attention moves TikTok distribution the way in-feed attention does. Every source in this category assumes it. None demonstrates it.
That is on the record in the product's own risk assessment, and it is why this page exists on a public site rather than in an internal document. If the mechanism turns out not to work, these pages describe software that will not ship — which is a better outcome than pages describing something that shipped and did nothing.
Common questions
Will the follow exchange be added later?
There is no plan to add it. The reasons against it are structural rather than a matter of timing.
Why publish this at all?
Because "we do not do that, and here is the number attached to why" is checkable, and "we take safety seriously" is not.