TikTok Shop Videos Scraper — Affiliate Video Transcripts

Paste a TikTok Shop product link and get the affiliate videos TikTok shows on that product's page — the creator, the caption and hashtags, views, likes, when it was posted, how long it runs and the "Commission paid" label — and what each video actually says, as a timestamped transcript. No login, no cookies, no proxy to configure. $0.01 per product read with something new, $0.004 per video row, and $0.015 per transcript on the Apify free plan, $0.014 on Bronze ($0.0075 on Silver and Gold); a repeat read of a product that finds nothing new is one $0.0065 check. Platform usage included, nothing charged for starting a run. Press Start with nothing set and the run returns three built-in sample rows so you can see the exact output shape first — nothing is read from TikTok and there is no result fee.

Using an AI agent? Pin this actor in Apify's MCP server with one link: https://mcp.apify.com?tools=steadyfetch/tiktok-shop-video-scripts-scraper

• Actor id: steadyfetch/tiktok-shop-video-scripts-scraper
• Input: { "products": ["https://www.tiktok.com/shop/pdp/<product-slug>/<product-id>"], "transcripts": true, "maxVideosPerProduct": 20 } — "products" is the one field you have to set; "transcripts" adds the spoken words.
• Cap the bill: set maxTotalChargeUsd on the run (a run option, not Actor input), e.g. 0.50 — the run stops when it reaches it, and the last row says what it did not deliver.
• How often it changes: the videos TikTok lists on a product page change slowly (a repeat read saw the same 19 videos in the same order, the newest of them 294 days old on one product), so a weekly schedule is the useful cadence — repeat memory skips the videos you already have, and a product with nothing new costs one $0.0065 check.

Unofficial. Not affiliated with, endorsed by, or sponsored by TikTok or ByteDance Ltd.

What one row looks like

Each product you send gives one video row per affiliate video, then one product row that closes it. The examples below use made-up placeholder values, never real data.

A video row:

{
  "rowType": "video",
  "product": {
    "id": "1234567890123456789",
    "title": "Example Product Name",
    "price": 19.99,
    "currency": "USD",
    "soldCount": 12000,
    "shopName": "Example Shop",
    "link": "https://www.tiktok.com/shop/pdp/example-product/1234567890123456789"
  },
  "video": {
    "id": "7000000000000000000",
    "link": "https://www.tiktok.com/@creator.handle/video/7000000000000000000",
    "creatorName": "Creator Name",
    "views": 125000,
    "likes": 4200,
    "comments": null,
    "shares": null,
    "postedAt": "2026-01-15T18:30:00.000Z",
    "durationSeconds": 42,
    "caption": "This is the one thing I use every morning #tiktokshop #example",
    "hashtags": ["tiktokshop", "example"],
    "affiliateLabel": "Commission paid"
  },
  "transcript": {
    "text": "Okay so this is the one thing I use every morning…",
    "language": null,
    "source": "captions",
    "durationSeconds": 41.8,
    "segments": [{ "start": 0, "end": 3.4, "text": "Okay so this is the one thing" }]
  },
  "transcriptStatus": "transcribed",
  "transcriptReason": null,
  "transcriptRetryable": false,
  "transcriptCharged": true,
  "repeat": false,
  "status": "delivered",
  "charged": true,
  "chargedEvents": { "affiliate-video": 1, "transcript": 1 },
  "retryable": false,
  "statusReason": null,
  "input": "https://www.tiktok.com/shop/pdp/example-product/1234567890123456789",
  "sourceIndex": 0,
  "scrapedAt": "2026-10-22T09:00:00.000Z"
}

A product row:

{
  "rowType": "product",
  "product": {
    "id": "1234567890123456789",
    "title": "Example Product Name",
    "price": 19.99,
    "currency": "USD",
    "soldCount": 12000,
    "rating": 4.7,
    "reviewCount": 850,
    "shopName": "Example Shop",
    "link": "https://www.tiktok.com/shop/pdp/example-product/1234567890123456789"
  },
  "videosListed": 19,
  "videosNew": 19,
  "status": "delivered",
  "charged": true,
  "chargedEvents": { "product": 1 },
  "retryable": false,
  "statusReason": "TikTok lists 19 affiliate videos on this product's page; 19 new ones delivered above. You asked for up to 20; the page lists no more. The product fee was charged once for this product."
}

The transcript is trimmed for this page; a real row carries it in full, with every timestamped segment. The last row of every run is the _summary receipt (status: "run_summary"): products read, videos delivered, transcripts produced and how many came from TikTok's own captions, what was skipped, what stopped the run, and the totals for all four charged events.

What you get

Column	What it holds
rowType	product or video
product	the product's id, title, price and currency, how many have sold, and its shop and link (the product row adds its rating and review count)
videosListed · videosNew	on the product row: how many videos the product page lists, and how many of them your account did not have yet
video	the video's id and link, the creator's display name, views, likes, when it was posted, its length, the caption, the hashtags pulled out of it, and the affiliate label TikTok shows on it
video.comments · video.shares	always null in this build: the product page does not carry those two counts
transcript	the spoken words, where they came from (captions or speech), their language when the words were read from the audio (null on captions), the length and the timestamped segments; null when no words were produced
transcriptStatus · transcriptReason · transcriptRetryable · transcriptCharged	whether a transcript was produced, why not when it was not, whether a re-run could change that, and whether it billed
repeat	whether your account already had this video
status · charged · chargedEvents · retryable · statusReason	whether this row was billed and for which events, so your invoice reconciles from the dataset itself
input · sourceIndex · scrapedAt	the exact value you sent, its place in your list, and when it was read

TikTok lists the videos it picks for the product page — its top videos for that product, not the newest — up to about 20 per product.

What to set

• products — TikTok Shop product links, one per line: https://www.tiktok.com/shop/pdp/<product-slug>/<product-id>, https://www.tiktok.com/shop/pdp/<product-id>, https://shop.tiktok.com/view/product/<product-id>, or the bare product id. A pasted block of lines is read as a list.
• transcripts (default true) — adds the spoken words to every delivered video.
• maxVideosPerProduct (default 20, 1 to 20) — videos per product.
• newVideosOnly (default true) — the repeat memory described below.
• maxRunSeconds (default 1800, 60 to 3,600) — the run stops cleanly before this and reports what is left. A time limit ends the collecting, never the delivering: videos already in hand are always written out.

Every optional field accepts null, which means "use the default".

Price

Four charged events, and each lands only on work that was actually delivered.

Event	When it is charged	Price
product	once per product read that delivered at least one video new to your account	$0.01
affiliate-video	once per video row delivered	$0.004
transcript	once per video whose spoken words were actually produced	$0.015 on the Apify free plan · $0.014 on Bronze · $0.0075 on Silver, Gold and above
check	once per product a repeat read found nothing new on	$0.0065

The arithmetic on a typical job: one product, 19 videos, transcripts on, on Gold = $0.01 + 19 × $0.004 + 19 × $0.0075 = $0.2285. On the Apify free plan the same job is $0.371, and on Bronze $0.352. Next week, the same product with nothing new is one $0.0065 check. Platform usage is included in every price — there is no separate compute bill on top, and no start fee. Set maxTotalChargeUsd on the run and it is a hard ceiling: the run stops cleanly under your maximum.

Most of these videos already carry TikTok's own captions, and the words are taken straight from them; transcript.source says captions when that is where they came from and speech when the audio had to be transcribed. You pay the same either way.

Honest rows: what "not charged" actually means

• A video whose words could not be produced is still delivered as a normal row with transcript: null and a reason on transcriptReason, and the transcript event is not charged for it — only the video row is.
• A long video with no captions: speech-to-text runs on videos up to 2 minutes and 12 MB. A longer or bigger video without TikTok's own captions ships its row with transcriptStatus: "transcript_too_long", no transcript and no transcript charge. Captions have no length limit.
• A product link from a shop outside the US is refused on its own uncharged row with a one-line reason; the rest of your list runs as usual. So is a product TikTok reports as sold outside the US.
• Anything that is not a TikTok Shop product — a creator's profile, a single video link, another site's address — is refused on an uncharged row naming the mistake.
• A read that did not go through says so on an uncharged row and the run finishes successfully — a failed read is never billed and never dressed up as a product with no videos. retryable says whether a re-run could change the answer.

How often the data changes

The set of videos TikTok lists on a product page moves slowly: on one product a repeat read returned the same 19 videos in the same order, the newest of them 294 days old. View and like counts keep moving, so re-run for fresh counts.

Re-running the same products does not re-buy what you already have. With newVideosOnly on (the default), every run remembers the videos it delivered to your account for each product, in a key-value store in your own Apify account, and a later run skips them: not delivered again, not charged, never re-transcribed. A product with nothing new charges one $0.0065 check instead of the $0.01 product event. That makes a weekly schedule on the same products cost only what is new. Switch newVideosOnly off to get every listed video back on each run; a video your account already had is marked repeat: true. A run started with a scoped API token in restricted-access mode needs key-value store Read, Write and Create permission (or Actor runs set to Full access) to use that memory; without it the run says so on its status line and cannot skip what you already have.

Run it on a schedule: save your input as a Task (Actor page → Save as a new task), then add that task to an Apify Schedule (Console → Schedules → Create) with a cron such as 0 7 * * 1 for every Monday.

What this build does not do yet

• US TikTok Shop only. A product link from another country's shop is refused per item, uncharged.
• No comment or share counts. The product page does not carry them, so video.comments and video.shares are always null.
• Up to 20 videos per product — what the product page lists. This actor does not search shops by keyword or walk a creator's whole profile.
• Spoken words come from TikTok's own captions when the video has them (most do), otherwise from speech-to-text for videos up to 2 minutes and 12 MB. A longer uncaptioned video ships its row without a transcript, and the transcript is not charged.

If this actor is not the door you want

• A creator's own videos, with transcripts: TikTok Transcript Scraper — TikTok Subtitles to Text, No Login
• Any single video link, from TikTok or elsewhere: Podcast Transcription Scraper — Speech to Text, Audio & Video

Use it as an API endpoint (Standby)

You can also call this actor like an API: send a product and get its videos back in the same response, with no run to start and no dataset to read. The address is https://steadyfetch--tiktok-shop-video-scripts-scraper.apify.actor, and the actor's Endpoints tab lets you try it from the browser. It takes one product per request — product (a product link or id), plus maxVideosPerProduct and transcripts — by GET or POST.

curl -H "Authorization: Bearer <YOUR_APIFY_TOKEN>" \
  "https://steadyfetch--tiktok-shop-video-scripts-scraper.apify.actor/?product=1234567890123456789&maxVideosPerProduct=10&transcripts=true"

The answer is JSON: the product row as product, the video rows as videos, the receipt as summary and every uncharged fact as notes — the same rows a run writes to its dataset. A request is charged with the same events and the same prices as a run with the same settings. Good to know: the endpoint has no account memory, so newVideosOnly does not apply there and every request returns and charges the videos it reads. To skip what you already collected, run the actor as usual.

Limits and good manners

Only public product pages and public videos are read. This actor may fail when the platform changes — failed items are never charged.

If it earned its keep, a rating helps other buyers find it, and saving the actor keeps it one click away.

Support

Something off, or a column you need that is not here? Open an issue on the Issues tab.