Invented specifics
Addresses, opening hours and prices arrive fully formed and completely unsourced. They read exactly like the ones that are true.
Verifiable venue intelligence
Your AI just invented that restaurant's address.
Ask an assistant where to eat and it will answer with total confidence: an address that moved two years ago, a price nobody quoted, a rating nobody gave. Power Scraping is the opposite bargain. Run collection on declared profiles; we enrich the posts and hand back records in which every single field cites the evidence it came from. What we cannot evidence, we return as unknown.
Run hosted collection on managed workers after a one-time Instagram login connection, or use the Power Scraping Collector in your own desktop browser. Hosted jobs and schedules continue while your browser is closed.
- Every derived field carries a citation to the exact caption, post detail or venue lookup it came from.
- Unknowns stay unknown. A null is never quietly filled in.
- Ungrounded model output is discarded atomically, not softened.
- Every job records what was collected, when, and under which declared basis.
The problem
Confident invention is the default failure mode
A language model asked to summarise a venue will produce a fluent answer whether or not the facts exist. There is no marker on the output that separates "the creator said this" from "this seemed plausible". If you are publishing a guide, briefing a client or feeding a recommendation engine, that difference is the whole job.
Laundered hearsay
A throwaway comment becomes "widely praised". A single visit becomes a rating. The chain back to who actually said it is gone by the time you see the summary.
No way to check
When a client challenges one field, you have to re-do the research to answer. There is nothing to point at, which means there is nothing to defend.
The answer
Verifiable extraction, in three steps
The engine is the boring part. The product is the audit trail it leaves behind.
Declare what you may collect
You name each profile as an authorized target and state the basis: your own account, a signed creator contract, a documented client mandate, a public profile with a recorded purpose, or public-interest research — plus a reference we can audit. Nothing runs against an undeclared handle.
Collect and enrich automatically
Hosted workers collect captions, visible comments and media using your encrypted connected session. Image and video OCR adds text evidence; remote AI and Places can add enrichment when their providers are configured. For local collection, the optional browser extension sends its fixed post fields. Every derived claim must cite evidence.
You get records you can defend
Any field whose citations do not resolve is dropped, not downgraded. Every job records what was collected, when, and under which declared authorization basis.
Hosted automation or your own browser
| Capability | Hosted workers | Browser Collector |
|---|---|---|
| Setup | Choose Connect Instagram in the dashboard; no installation | Install and pair the extension |
| While your browser is closed | Jobs and unattended schedules continue | Collection waits for your browser |
| History | Resumable campaigns for owned or contract-authorized profiles | Bounded recent posts |
| Evidence and files | Captions, visible comments, media, OCR, review files and archives where produced | Fixed post fields and evidence-backed records |
Remote AI and Places enrichment depend on configured providers. Complete-history campaigns report checkpoints and observed coverage; they cannot guarantee access to private, deleted or blocked posts.
A real record
One post in, one citable record out
This is the shape of what the API and the MCP connector return, abridged to the fields that matter here. Note the last row: the creator never gave a rating, so the field is null and says so.
- Record
rec_8d31f0a4· from a single Instagram post- Venue
- Trattoria da Oscar
- City
- Milan
- Prices named
- €14, €22 — price signal
mid_range - Sentiment
- positive
- Explicit rating
- unknown — the creator never gave one
| Field | Value | Evidence | Quoted from the source |
|---|---|---|---|
venue.name |
Trattoria da Oscar | Caption ev_01 |
"finally back at @trattoriadaoscar for the cotoletta" |
location.city |
Milan | Geotag ev_02 |
"Milano, Lombardia" |
price.mentions_eur |
14, 22 | Caption ev_03 |
"cotoletta €22, calice di rosso €14, worth every cent" |
assessment.sentiment |
positive | Caption ev_03 |
"worth every cent" |
venue.place_id |
null | none | No places provider is asked to match the venue, so no third-party place id is attached; the venue is named from the post's own location tag and caption. |
assessment.explicit_rating |
null | none | No rating appears anywhere in the post, so the field is returned as unknown rather than inferred from tone. |
An enrichment pass that cites an evidence id which does not resolve has its whole field discarded before the record is written. There is no "low confidence" tier that lets an ungrounded guess through.
Compliance posture
Your session, your targets, the controls written down
Hosted jobs and schedules use a session you connect explicitly; browser jobs use your paired extension. We encrypt hosted sessions, enforce declared targets and plan allowance, and stop when Instagram requires attention.
Only targets you are entitled to
A job runs only for a profile declared as an authorized target, with an authorization basis and an auditable reference; a job for an undeclared or revoked handle is refused. Posts of any other account are dropped on your device and again on our server.
It stops at the first warning
An Instagram warning, a challenge or a logout stops collection. Reconnect a hosted session through the dashboard when needed. In the hosted viewer you interact with the actual Instagram page running on our server, including personal login and two-factor checks. Your device encrypts input for that isolated browser, which handles it transiently; we retain the validated encrypted session for later jobs, not your password as an account credential. The optional browser extension never reads cookies. Never send passwords or verification codes through chat or support.
Everything is on the record
Every collection job writes an audit event naming the actor, the target and the declared basis. Browser jobs need your click; hosted jobs and schedules run after setup. Acceptance of the current acceptable-use version is required before any job will run at all.
Your records stay yours
We delete a target's data on request, keep records only for your plan's retention window, and never pool, sell or re-sell them.
Pricing
Three tiers, metered on posts processed
One metered unit: a post the service actually processed. No seat games, no per-field charges, and jobs stop at the allowance. Hosted and browser collection share the same allowance.
Free
Free
Prove the evidence trail on a handful of posts before you commit.
- 1 authorized target
- Evidence-grounded enrichment
- CSV and JSONL download of results
- 7-day retention
Most popular
Pro
£20 /month
No VAT charged
For one operator running continuous venue intelligence.
- 10 authorized targets
- MCP connector for Claude and ChatGPT
- Evidence-grounded enrichment
- 90-day retention
Agency
£800 /month
No VAT charged
For teams collecting on behalf of named clients.
- 200 authorized targets and 10 seats
- 365-day retention, priority support
Questions
The things people ask before signing up
What exactly is a citation?
A pointer from one field of a record to one numbered evidence item collected from the source, together with the modality (caption, location label, venue lookup), a short verbatim excerpt, the method that produced the field and its confidence. You can quote it straight back to whoever is challenging the field.
What happens when the model is not sure?
Nothing gets written. A field whose citations do not resolve against the collected evidence is discarded atomically before the record is stored. The record still arrives, with that field null rather than filled in.
Which profiles can I collect?
Ones you are entitled to have collected: your own accounts, creators you hold a signed contract with, clients who have given you a documented mandate, or subjects of public-interest research. You declare the basis and a reference per target, and you revoke it the moment it stops being true.
Do I need to install anything?
Hosted mode requires no software installation. Choose Connect Instagram in the dashboard to open a secure hosted Instagram browser, then personally complete login and any two-factor verification. After that, jobs and schedules run on our workers while your browser is closed. Browser mode uses the optional Collector extension on desktop Chrome, Edge or Brave. You can manage jobs and download results from any browser. Read the privacy notice for hosted input handling and encrypted session retention.
What does the Collector send you?
A fixed list of fields from the declared profile's posts: post id, handle, link, dates, caption, location label, media type, links to the media, and like and comment counts. Never comments, other page text, your feed, your messages or who you follow. Posts of any other account are dropped on your device and again on our server.
Can automatic mode get my Instagram account restricted?
Yes, it can. Hosted automation and browser automatic mode can trigger restrictions. In browser manual mode, you open posts yourself and press Send this post. Automatic mode is off until you turn it on in a browser and acknowledge the risk: it then opens the profile and its posts in a visible tab at a slow, fixed pace with small daily caps. Instagram may still treat automated activity as a breach of its terms and respond with warnings, temporary action blocks or restrictions on your account — that risk is yours, not ours. The Collector stops at the first warning, challenge or logout; a challenge pauses that browser for 24 hours and switches automatic mode off until you turn it back on. You must follow Instagram's terms and have a lawful basis for every profile you collect.
What happens if Instagram shows a warning or a challenge?
The job stops and says why. We do not evade access controls of any kind, and we will not add that capability. A refusal you can see beats data you cannot explain.
How does the MCP connector fit in?
Connect Power Scraping to Claude, ChatGPT or an IDE agent and the same evidence-grounded records become callable in a conversation: list targets, start a collection, poll it, then read records and search them. The assistant quotes the citations instead of inventing the gaps. Hosted collection from a chat uses an already connected session; browser-mode jobs need the Collector paired on a desktop browser. See the quickstart.
What is a processed post?
One collected post the service processed in the current billing period. Posts skipped by a filter are not metered, and downloading results you already have never counts again.
Can I get my data out?
Yes. Every finished job has Download CSV and
Download JSONL buttons in the dashboard, and the same files come from
GET /v1/jobs/{job_id}/export in the REST API — every record, with its
evidence citations intact.
Start with 50 posts and check the citations yourself
No card. Install the Collector, declare one target, run one collection, and read the evidence behind every field it returns.