logoPandorLabs
The fastest trend cycle online

TikTok Data, Delivered

Public TikTok content and engagement metrics, normalised to the same schema as every other platform we cover and pushed into your stack. Nothing to build, nothing to maintain.

source specification

Source
tiktok.com
Objects
Videos, Public creator profiles, Hashtag and sound feeds, Comment threads, View and engagement counts
Formats
JSON · CSV · Parquet
Refresh
Continuous to daily
Scope
Public content only

about the platform

What makes TikTok different

TikTok compresses the trend cycle further than any platform before it — a format can go from nothing to saturated inside a fortnight, which makes monthly reporting on it close to useless. Its distinguishing structure is the sound: audio is a first-class object that videos attach to, so a trend can be tracked through the sound it is built on rather than only through hashtags or captions. Because distribution is recommendation-driven rather than follower-driven, view counts on a small account can dwarf those on a large one, and follower count is a much weaker proxy for reach here than anywhere else.

why teams track it

Why TikTok data is worth having

TikTok is where consumer demand is now manufactured, and the gap between a trend appearing and it reaching your sales data is often the whole commercial opportunity. Brand and merchandising teams use it to catch demand spikes early enough to move inventory, and agencies use per-video reach data to price creator work on actual distribution instead of audience size, which on this platform barely predicts it.

Visit tiktok.com

applications

What teams build on TikTok data

Catching a trend while it is still early

Posting velocity on a hashtag or sound tells you a format is accelerating days before it reaches the volume that makes it obvious. On a platform where the cycle runs in weeks, days of lead time is the entire advantage.

Sound-led trend tracking

Tracking the sound rather than the caption catches a trend whose text framing keeps mutating. It is the most reliable identifier TikTok offers, and it has no equivalent on any other network.

Creator selection on reach, not followers

Because distribution is algorithmic, median views per video is a far better predictor of what a partnership will deliver than follower count. Pulling a creator’s recent video history gives you that number directly.

Demand forecasting for inventory

A product format trending here shows up in search and then in orders, in that sequence. Teams that watch the first step get to reorder before the category sells out.

Campaign and hashtag challenge reporting

Every public video carrying a branded hashtag, with views and engagement attached, aggregated into one table instead of assembled from creator screenshots.

Competitive format analysis

Which video formats, hooks, and lengths are working for competitors right now, measured across their whole recent output rather than the three posts they chose to amplify.

what you get back

Fields in the TikTok feed

Platform-specific detail on top of the unified social schema, so a dashboard built on one platform works on the next without a rewrite.

  • Video description, hashtags, and mentions
  • View, like, comment, share, and save counts
  • Attached sound or music identifier, with its own usage count
  • Public creator profile — follower, following, and total like counts
  • Comment threads with author handle and reply nesting
  • Video duration and posting timestamp

Public surfaces only

We collect what a logged-out visitor can see. No authentication is bypassed, no private content is touched, and no attempt is made to deanonymise anyone.

One schema across platforms

A post from any network arrives with the same field names, so adding a platform does not mean rewriting anything downstream.

Deletion propagates

When content is removed at source it drops out of subsequent deliveries, and retention windows on delivered data are configurable per engagement.

10+
Platforms covered
<60s
Detection latency
3
Delivery formats
99.9%
Uptime SLA

questions

Frequently Asked Questions

Everything you need to know before you send us your first request.

Detection latency on tracked hashtags, sounds, and handles is under a minute in the standard configuration, and the aggregate velocity signal is recomputed continuously. That matters more here than on other platforms: a daily batch on TikTok can miss the entire useful window of a fast trend.
Yes, and for trend work it is usually the better identifier. Videos carry the sound they were built on as a first-class field, so a format can be followed through its audio even as captions, hashtags, and framings mutate around it. Both can be tracked together in the same engagement.
By default we deliver metadata, engagement counts, and text — descriptions and comments — rather than media binaries. Media collection is available where the intended use supports it, but it is scoped explicitly in the engagement rather than enabled by default, because storage, cost, and licensing all change materially once video is involved.
It is the number the platform publishes, and we deliver it as observed with the timestamp of observation attached. Because views keep accruing long after posting, a single snapshot understates a video’s eventual reach — so for anything comparative we recommend repeated observation, which is why the record carries an observation time rather than being overwritten in place.
Into the systems you already run. Normalised records land as JSON, CSV, or Parquet in S3, GCS, or Azure Blob, or straight into Snowflake or BigQuery, on whatever cadence you set. Webhooks push matches to your endpoints as they are detected, which is how most customers wire alerting. A solutions engineer sets the schema, cadence, and destination with you during onboarding.
Keeping the collector working is our responsibility. Extractors are monitored continuously, and when the platform changes its structure we patch upstream while the schema we deliver to you stays fixed. If a change causes a coverage gap we tell you which window was affected rather than quietly returning fewer records.
Pricing is scoped per engagement — the number of tracked terms, handles, or communities, the refresh cadence, and the delivery destinations involved. Historical backfill is quoted separately. Every engagement starts with a free sample run against your own brand terms, so you can judge the data before committing.

Still have questions?

Talk to an engineer

Ready to Get Started?

Talk to us about your sources and volume. We'll return a sample dataset from your target sites before you commit to anything.

SOC 2 Type II
GDPR & CCPA compliant
99.9% uptime SLA

© 2026 PandorLabs, Inc. All rights reserved.