Social data, first-hand by design

Why this exists now

The official routes closed, one after another. Instagram’s API requires a login and only serves accounts that have connected themselves to your app. X removed its free tier in 2026 and now charges per resource returned — US$ 0.005 a post, read from their own pricing documentation. Reddit shut off unauthenticated endpoints in May 2026; its commercial access is a negotiated agreement rather than a published price, and we do not quote a figure for it because we could not confirm one at the source.

So a whole category of small builders got locked out at once — and a market of resellers grew in the gap, each one selling the same scraped pages at a markup, most of them describing it in language that implies an access they do not have.

How the collection works, in plain words

We read public pages — the same ones you see logged out, in a browser, without an account. We do not log in. We hold no account on any platform for this purpose. We have no partnership, no authorisation and no official access from any platform, and any vendor in this market who implies otherwise is worth reading twice.

The collection will be ours: our own client, our own residential proxy — designed and measured, and not yet running. No proxy is contracted and no collector is deployed. The reason for the choice is not that it is cheaper to say so: a reseller in the middle can change what comes back without telling you, and can be a direct competitor to the thing you are building on top of it.

What this does not solve

Public data is still personal data under GDPR and LGPD. Collecting it from a public page does not make it exempt, and if you store what comes back, you become the controller of it. Every vendor in this category pushes that responsibility to the customer; the difference is whether they say so on the pricing page.

Scraping also conflicts with these platforms’ terms of service. That is civil exposure rather than criminal, and Meta v. Bright Data (2024) found that logged-off collection of public data is not bound by terms you never accepted. It is a real position with a real trade-off, and it is ours on purpose rather than something we discovered by accident.

What is built, honestly

No social endpoint is live yet. There is nothing to call, nothing to buy, and no waiting list dressed up as a launch. When an endpoint answers, it appears here with a real response pasted into the page and the cost per thousand calls next to it — and not one day earlier.

What is live is the other product: the trend archive, self-serve today.

Reading while you wait

• Which Instagram scrapers still work in 2026What broke, what survived, and how to tell whether the fault is yours or theirs.
• How to read an Instagram public profile without logging inThe method, the failure modes, and the part everyone gets wrong about rate limits.
• Public data is still personal dataGDPR and LGPD for people who scrape — written for builders, not lawyers.
• The Reddit API closed in May 2026 — what still answersEvery remaining public surface, measured: the one that returns 200 with the wrong content, and the one that survives without scores.
• X killed its free tier — the real cost of the alternativesPriced per resource returned. What that works out to at real volumes, and the 24-hour rule that halves some bills.
• Every vendor in this market, measuredReal pricing tiers read from primary sources. No affiliate links.