Tracking TikTok Effects and Filter Trends in 2026

Effects and filters are how trends actually move on TikTok. A single AR effect can turn a small creator into a viral hit, push a product into the For You feed for weeks, or anchor an entire branded campaign. The catch for analysts, brand teams, and AR creators is simple: the public TikTok platform never shipped a clean "effects" endpoint, and most third party scraper APIs are no different.

Let's be honest up front. TikLiveAPI does not have a dedicated /effects/ endpoint. There is no single call that returns the top videos for the Bold Glamour filter, no endpoint that lists every effect a creator has published in Effect House, and no first class "effect ID to usage count" lookup. What we do have are 37 well documented endpoints for users, videos, comments, hashtags and search, and those endpoints are enough to build a real effects tracking workflow if you treat effects as a soft signal rather than a structured field.

This post walks through what TikTok effects are, what data is and is not exposed, and a practical pipeline that brand teams, AR creators and trend analysts can run today using TikLiveAPI. Where there are gaps, we'll point them out instead of pretending they don't exist.

What TikTok Effects and Filters Actually Are

TikTok's effects ecosystem has three layers worth understanding before you write a single line of code.

Native AR Effects

These are the face filters, background masks, transitions, and full screen overlays that TikTok ships in the camera. Examples include the Bold Glamour filter that went viral in 2023, the Inverted filter, time warp scans, and beauty mode variants. Each native effect has a stable name and an internal effect ID that TikTok surfaces inside the app when you tap "Effect" on a video.

Branded Effects

Brands can commission custom AR effects through TikTok's ad sales team or build them through Effect House. A typical product launch effect lets users virtually try on a lipstick shade, place a sneaker in a room, or stamp a logo on their face. Branded effects show up with the brand's name attached and usually run alongside a hashtag challenge.

Effect House Community Effects

Effect House is TikTok's free creator tool, similar to Spark AR for Instagram. Independent AR creators publish effects there, and any user can apply them. These effects carry the AR creator's username, and the most successful ones get millions of uses without any paid push.

From a tracking perspective, all three layers leak the same kind of signal into the public feed: an effect name that often appears in video captions, hashtags, comments and on the video player. That string is your hook.

What Is Not Exposed and Why

Before describing the workaround, here is the honest gap list:

• No endpoint accepts an effect ID and returns usage counts.
• No endpoint lists all videos that used a specific effect, sorted by date.
• No endpoint enumerates the effects a single creator has published in Effect House.
• The effect ID embedded in a video's metadata is not consistently exposed across TikTok's public web surfaces, so we cannot promise it on every /post-detail/ response.

If you need guaranteed effect level analytics for paid campaigns, the right tool is TikTok's own Effect House dashboard or a partnership with TikTok Ads. For everything else, public data plus careful string matching gets you most of the way there.

The Workaround: Treat Effect Names as Search Terms

The core idea is that creators, when they care about an effect, almost always name it. They type "using the bold glamour filter" in the caption, hashtag #boldglamour, or pin a comment that says "filter is called X". That gives you three observable surfaces, each backed by an existing TikLiveAPI endpoint.

Surface One: Video Search

Use the video search endpoint to fan out across an effect's known names and aliases. Every effect has a canonical name plus a handful of misspellings the community settles on.

GET https://api.tikliveapi.com/search-video/?keyword=bold%20glamour%20filter
X-Api-Key: YOUR_KEY

The response contains a videos array, a cursor, and a hasMore flag. Iterate the cursor and collect everything matching the effect name. Run the same loop for variants like bold glamour, boldglamour, and glamour filter. Dedupe by video ID. The full search contract lives at /documentation/posts/search/.

Surface Two: Challenge Lookup

If the effect launched alongside a hashtag (extremely common for branded effects), the /challenge-info-name/ endpoint returns the hashtag's profile: total views, video count, and a description blurb. That's your aggregate signal for how big the campaign is in absolute terms.

GET https://api.tikliveapi.com/challenge-info-name/?name=boldglamour
X-Api-Key: YOUR_KEY

Combine the challenge total view count with your video search counts and you get a rough "share of voice" calculation: how many of the hashtag's videos explicitly cite the effect name in the caption.

Surface Three: Post Detail Inspection

For any video you have already discovered, call /post-detail/ and inspect the response for effect related metadata. When the field is present, you can confirm an effect ID, the effect's display name, and sometimes the AR creator's handle. Treat this as best effort: if the field is missing for a particular video, fall back to the caption text.

GET https://api.tikliveapi.com/post-detail/?video_id=7340000000000000000
X-Api-Key: YOUR_KEY

A full list of post and search endpoints is at /documentation/posts/.

Worked Example: Tracking Bold Glamour Across a Quarter

Suppose you are a trend analyst and want a weekly trend line for the Bold Glamour filter over the last 12 weeks. Here is the pipeline.

1. Seed the name dictionary. Start with the canonical effect name, plus three to five common variants you find by skimming the first 100 search results manually.
2. Fan out the search. Hit /search-video/ once per variant per week of interest. Paginate via cursor until hasMore is false or you hit a sensible cap such as 2,000 videos per term.
3. Dedupe and timestamp. Use the video ID as the primary key. Parse the create time and bucket videos by ISO week.
4. Confirm with post detail. For a random 10 percent sample per bucket, call /post-detail/ and check whether the effect metadata is exposed. That gives you a confidence score for how reliable the caption matching is.
5. Augment with challenge totals. Pull /challenge-info-name/ for boldglamour once per week and store the cumulative view count. The week over week delta is your hashtag growth.
6. Plot. Three lines per week: caption mentions, sampled metadata confirms, and hashtag view delta. When the lines diverge sharply, that's where your interesting analytical story lives.

This is not a perfect census. It is, however, a defensible methodology that you can run repeatedly with the same code and explain to a client without overstating accuracy.

Branded Effect ROI: A Brand Team Playbook

For brand teams, the question is rarely "how big is this effect" and usually "did our branded effect earn its budget". Here is a workable framework.

Define the Effect Footprint

Before launch, agree on the effect's textual footprint: official hashtag, effect display name, and any spelling variants you expect creators to use. Lock that list as the campaign's tracking corpus.

Baseline the Hashtag

Two weeks before launch, snapshot /challenge-info-name/ for the campaign hashtag every day. You want a clean pre launch baseline so you can attribute the post launch lift.

Discover Top Videos Daily

During the campaign, run /search-video/ against your corpus daily, sort by view count, and store the top 50 videos plus their creators. These are your earned media wins. For each top video, call /user-posts/ against the creator's userid to confirm whether the effect post was a one off or part of a series, and call /post-comments/ to mine sentiment.

Calculate Earned Plus Paid

You should already know your paid impressions from TikTok Ads. Add the earned impressions estimate from your video search aggregate, divide by total spend, and you have an effective CPM that includes UGC amplification. Most brands stop at paid CPM and undercount the effect's real value by 30 to 60 percent.

Worth being honest again: the earned impressions number is an estimate built on public view counts, not a TikTok issued attribution. Document the methodology in the report so internal reviewers know exactly what they are looking at.

Building an Effect Discovery Pipeline

If you want to find effects rather than just track ones you already know about, comments and captions are your richest source. Creators routinely answer "what filter is this" in their comments, and ambitious AR creators promote their own effects in their video captions.

A discovery pipeline looks roughly like this:

1. Pull a daily sample of trending videos for your category via /search-video/ with broad category keywords.
2. Pull comments via /post-comments/ for the top videos.
3. Run a simple NER or regex pass over both captions and comments looking for the phrases "filter", "effect", "lens", and "using". Capture the surrounding noun phrase.
4. Cluster the extracted noun phrases by string similarity. Each cluster is a candidate effect name.
5. For clusters that pass a frequency threshold, feed the name back into the video search workflow described above and start tracking it.

This gives you a self refreshing watchlist of rising effects without needing any insider data.

Compliance and Effect House Developer Terms

Two compliance points are worth flagging.

First, TikTok's Effect House developer terms govern what you can do with effects you publish, not with public usage data about effects. Tracking how often a public effect appears in public videos through public search endpoints is the same legal posture as any other social listening work. Standard caveats apply: respect robots.txt, do not republish protected content, honour deletion requests promptly, and keep your retention windows defensible.

Second, branded effects commissioned through paid TikTok Ads packages sometimes come with NDA language. If you are tracking a competitor's branded effect and you happen to be inside their ad ecosystem under contract, that contract beats anything in this post. Check with legal before publishing analyses of effects you have privileged knowledge about.

FAQ

Can TikLiveAPI give me the effect ID for any video?

Not reliably. /post-detail/ sometimes exposes effect metadata and sometimes does not, depending on how TikTok itself surfaces the field in the public payload. Treat it as best effort and pair it with caption matching.

Is there a way to list every video that used an effect?

No. The closest substitute is a multi keyword video search across the effect's known names and a comparison against the campaign hashtag's video count.

How fresh is the data?

Search and post endpoints reflect TikTok's public surface in near real time at request time. Cache aggressively on your side to control cost and respect rate limits.

Do I need a paid plan to run this pipeline?

Volume drives cost. A daily snapshot of one effect with ten name variants and 2,000 videos per variant is well within mid tier plan budgets. Pricing details are at /pricing/.

Where do I start if I have never used the API before?

Open /documentation/, get an API key, and run a few searches against /search-video/ with a known effect name to see the response shape. The /playground/ lets you fire requests without writing code.

What if I need something this workflow cannot deliver?

Tell us. We have shipped endpoints in response to specific brand and analyst requests before, and effect related metadata is on the radar. Reach out via /contact/ with the exact field you need and the use case behind it.

Wrap Up

TikTok effects are a real trend driver and a real blind spot for most data tools. TikLiveAPI does not pretend to close that gap with a single endpoint, but the public search, post and challenge endpoints we do offer are enough to build a credible tracking pipeline if you accept the limits and document them. Start with one effect, build the name dictionary, baseline the hashtag, run the search loop, and pair captions with sampled post detail confirmation. That is how trend analysts and brand teams operate today without insider access, and it is how you can too. More guides like this on the blog.