TikTok Profile Video Signals

Get current metrics for videos visible in the public embeds of profiles you specify. Supply a previous dataset from this Actor to calculate changes between two observations. Every video has its canonical URL, embed evidence URLs and an observation timestamp.

This is an unofficial tool, not affiliated with, endorsed by or sponsored by TikTok.

Scope

The Actor reads a public profile embed to discover video IDs, then reads each public video embed. It returns only the visible embed selection, typically about 10–13 recent videos per profile. maxResults is an upper bound, never a promise of that many rows. Full history and pagination are not provided. It does not retrieve hashtags or search results, employees, comments, private accounts, or videos requiring login. No cookies, login, captcha handling or blocked-IP rotation. No video, audio, subtitle or image files are downloaded, stored or redistributed. Profile biographies, nicknames, avatars and location are excluded. No personal-identity enrichment.

Input

All input values are arrays, including single-value settings:

{"profiles":["scout2015"],"maxResults":[10],"previousDatasetIds":[],"proxyModes":["DIRECT"],"disclaimerAck":["accepted"]}

• profiles: 1–30 handles, @handles, or https://www.tiktok.com/@handle URLs.
• maxResults: one integer, 1–300. Default [10] for the whole run.
• previousDatasetIds: zero or one dataset ID from this Actor, readable with the run's permissions.
• proxyModes: ["DIRECT"] or ["DATACENTER"]. The mode does not change after a block.
• disclaimerAck: ["accepted"]. You confirm authorization to query the profiles, compliance with relevant terms and lawful use of results, including your duties as personal-data controller where applicable. This does not waive our statutory obligations.

Output and comparisons

One row per unique video. Required fields: id, webVideoUrl, authorMeta.name, createTime, playCount, diggCount, commentCount, observedAt, and evidenceUrls. Other observed metrics include shareCount; unavailable collectCount stays null. Metadata includes video dimensions/duration, caption, hashtag labels and music attribution when exposed; media files and download links are excluded.

The output preserves the field names and nested structures of the three compared profile/video products. Field compatibility does not imply feature parity. Unobserved or excluded fields are null; fieldAvailability explains not_observed, excluded_by_policy, observed or derived.

Without a baseline, comparisonStatus is no_baseline and viewDelta, likeDelta, commentDelta, shareDelta, collectDelta, elapsedSeconds and viewsPerHour are null. A readable prior observation of the same video and author with an earlier timestamp gives matched. Deltas are current minus previous; decreases are retained. A missing video gives no_match. Inaccessible or invalid baselines produce baseline_unavailable or invalid_baseline. A current view count alone is never described as growth or a trend. viewsPerHour is an observed two-point rate, not a forecast.

coverageScope=profile_embed_visible describes the source selection, not the account's complete history. The OUTPUT record reports requests, retries, failures, discovered counts, block rate, and complete or incomplete. Notices are not dataset result rows.

Pricing

Plan	Startup per unit	Per video-result
FREE	$0.001	$0.0100
BRONZE	$0.001	$0.0080
SILVER	$0.001	$0.0065
GOLD	$0.001	$0.0052
PLATINUM	$0.001	$0.0052
DIAMOND	$0.001	$0.0052

The platform manages apify-actor-start: one unit per GB of run memory, minimum one. The default 256MB run has one startup unit. Startup can apply even to an invalid input or a run with zero results. We never issue a second custom startup charge. video-result is charged once for each unique validated video snapshot delivered. Comparisons are included. No result charges for duplicates, retries, notices, blocked requests, missing required metrics or internal failures. At FREE list pricing, 10 delivered videos with default memory cost $0.101 in Actor events. Your platform plan governs any other applicable usage charges; owner tests do not demonstrate customer revenue.

Charging stops at the run's spending limit. A durable ledger and final dataset IDs prevent duplicate event calls across restarts. If a charge response cannot be reconciled, processing stops without charging it again. Already paid rows are retained. There is no separate feature cap for FREE users.

Requests and limits

Requests start at most once per 1.05 seconds. Timeouts, 429, 5xx and empty bodies receive at most three retries with 1/2/4-second exponential backoff. Only DATACENTER transient retries obtain a fresh proxy session. Login, captcha, 401/403 and redirects are not retried. Five consecutive blocked requests end with incomplete. Parsing or source failures remain visible; they are not reported as zero-valued metrics or billable successful results.

Embed availability and selection may change. This product has no full-history pagination. Review coverage and field availability before interpreting a snapshot.

Leave a review

A short review on the Reviews tab, good or bad, helps other people find this Actor and tells us what to build next. Found a bug or need a feature? Open an issue on the Issues tab and we'll look into it.