For product teams building on mention data

Build social listeninginto your product

Mention data, AI filtering, and monitoring infrastructure over one API — with webhooks and MCP where they fit. You own the customer experience; Octolens runs the listening layer underneath.

API, webhooks & MCP on every plan • 15+ sources, one schema • Full API access during trial
Your UIyourproduct.com/dashboard
Anyone compared these two for a mid-size team?
Their new release fixed the export problem for us
Show HN thread mentions the category leaders

Your brand, your components — live mention data underneath

Octolens API200 OK
POST /api/v2/mentions
Authorization: Bearer WORKSPACE_KEY

{
  "items": [{
    "title": "Anyone compared these two…",
    "source": "reddit",
    "relevanceScore": "high",
    "sentimentLabel": "Neutral",
    "keyword": "your customer's brand",
    "tags": ["buy_intent"]
  }],
  "nextCursor": "…"
}

Trusted by brands people talk about online

Cursor logo
Sentry logo
Vercel logo
PostHog logo
Railway logo
Render logo
Supabase logo
PlanetScale logo
Prisma logo
Juicebox logo
Modal logo
Metabase logo
Tally logo
Lovable logo
Medusa logo
CodeRabbit logo
What teams build

Three Things People Are Building Right Now

Not integrations of Octolens — products of their own, with Octolens as the listening layer inside them.

White-label client dashboards

A GEO/AEO platform ships a Social Listening tab to its clients: its own UI and brand, with a workspace, quotas, and mention data per client from Octolens underneath — including trials before paid access.

client signs up → workspace provisioned → your tab fills

Intelligence features inside your product

An AI-visibility product streams live Reddit and X conversations into its own Communities section — keyword and competitor feeds per customer, created over the API instead of by hand.

manual setup works for 7 clients — the API works for 500

Internal GTM tools

A security company pipes scored mentions into the seller command center it already built — social signal next to its own data, in its own visualization layer.

webhook → your warehouse → your views

Multi-tenant by design

One Account Up Top. A Workspace Per Customer Underneath.

The same workspace architecture agencies run on, exposed over the API — so your product can provision, cap, and meter customers without a human in the loop.

Your product on Octolens

Your product

Your auth
Your UI
Your billing
Your brand

Child workspaces per customer

Customer Aisolated data · own API keys
Customer Bisolated data · own API keys
Customer Cisolated data · own API keys
Customer Disolated data · own API keys
provisioned via API

Octolens infrastructure

Collection across 15+ sources
AI relevance · sentiment · tags
Quotas & metering

Programmatic provisioning

POST /api/v2/agency/workspaces creates a child workspace per customer — settings applied, onboarding skipped, ready for keywords. Onboarding a customer is an API call, not a checklist.

Isolation by default

Each workspace has its own data and its own API keys, scoped read/write/admin. Customers — and their integrations — never see across the wall.

Caps and pools you control

Set per-workspace mention and keyword limits and flex behavior over the API. Commercially, embedded plans price on your aggregate keywords and mentions — not per seat.

Metering for your billing

GET /api/v2/org/usage returns per-cycle mention and keyword consumption with plan limits and flex state — the numbers you need to run your own tiers and trials.

Multi-workspace provisioning runs on agency and embedded plans — we set the flow up with you.

Developer essentials

Everything You'll Ask in the Evaluation Call

Read/write REST, webhooks, and MCP — one schema across every source. The full reference lives in the API docs.

curl -X POST "https://app.octolens.com/api/v2/mentions" \
  -H "Authorization: Bearer WORKSPACE_KEY" \
  -d '{ "limit": 25 }'

# → { "items": [ { "title", "body", "url", "source",
#      "relevanceScore", "sentimentLabel",
#      "keyword", "tags" } ], "nextCursor": "…" }

Read/write REST with structured JSON: every mention arrives pre-scored for relevance and sentiment, tagged, and cursor-paginated.

Key scopes

read, write, and admin API keys per workspace — write implies read, admin implies write. Scoped to exactly one workspace.

Rate & usage limits

500 requests/hour per organization with X-RateLimit-* headers on every response and Retry-After on 429s. Exports: 5/hour, 2 concurrent.

Backfill

Every new keyword gets a 7-day backfill, then collects continuously. Octolens isn't a historical archive — the archive builds from the day a keyword is added.

On-demand search

POST /api/v2/search runs a one-off, AI-scored search over the last 1, 7, or 30 days — with quota headers so you always know what a call cost.

Sources

15+ platforms — Reddit, X, LinkedIn, Hacker News, GitHub, YouTube, and more — through one endpoint with one schema. News, podcasts & newsletters on Scale.

Spec & docs

Full OpenAPI spec at octolens.com/openapi.json. Docs: API overview, webhooks quickstart, and the MCP server reference.

Commercial path

Start on a Trial. Scale on Terms That Fit.

Builders range from a weekend prototype to a platform provisioning hundreds of customer workspaces — one fixed package would obscure the decision, so there isn't one.

Prototype this week

Self-serve

The full API — read/write REST, webhooks, MCP — is on every plan from day one, including the trial. Most builders validate the integration on Pro before any call with us.

A roster of client workspaces

Agency & multi-workspace

Running listening for clients rather than inside a product? The agency setup — isolated workspaces, shared pools, central billing — has its own page and its own terms.

High-volume, programmatic

Embedded / OEM

Provisioning workspaces from your signup flow, pooled quotas, per-workspace caps, volume pricing on aggregate keywords and mentions. There's no fixed package — because the right shape depends on what you're building.

Builder Questions Answered

Yes — that's the intended use. The REST API is read/write: pull scored mentions into your own components, create keywords and feeds programmatically, record relevance feedback, and drive it all from your product's backend. Webhooks push mentions to you in real time, and the MCP server exposes the same data to agents. Your customers never have to see Octolens.

No — and that's a deliberate line. Octolens supplies the data and infrastructure; you build the experience in your own components, your own brand, your own product. If you want a finished interface instead of building one, that's the Octolens app itself.

Yes. On agency and embedded plans, POST /api/v2/agency/workspaces creates an isolated child workspace per customer — own data, own API keys — and PATCH lets you set per-workspace mention and keyword caps and flex behavior. Wire it to your signup flow and onboarding stops being manual. We set the provisioning flow up with you.

Every new keyword gets a 7-day backfill, then collects continuously — the archive builds from the day monitoring starts. Octolens is not a deep historical archive; if you need a one-off look further back than your collection window, on-demand search covers the last 1, 7, or 30 days. Plan your customer onboarding around that: provision the workspace the day they sign.

500 API requests per hour per organization, shared across that org's keys, with X-RateLimit-Limit/-Remaining/-Reset headers on every response and Retry-After on 429s. Exports run at 5 per hour with 2 concurrent. Mention and keyword quotas come from the plan, and on multi-workspace setups you control each workspace's share.

GET /api/v2/org/usage returns per-cycle mention and keyword consumption, plan limits, and flex state; per-workspace caps let you enforce tiers. Builders typically map their own pricing tiers to workspace limits and read usage into their billing system — including free trials for their customers before paid access.

15+ platforms through one endpoint and one schema: Reddit, Twitter/X, LinkedIn, Hacker News, GitHub, YouTube, TikTok, Bluesky, DEV, Product Hunt, Stack Overflow, Medium, and reviews — with news, podcasts, and newsletters on the Scale plan. Every mention arrives with AI relevance, sentiment, and tags already applied.

We want to. If you're exposing Octolens inside an automation platform, assistant, or your own product's integration directory, talk to us — completed integrations can be reviewed and featured alongside the ones at octolens.com/integrations, and we're happy to co-announce.

API, webhooks & MCP on every plan

Build social listening into your product

Ship a prototype on the trial this week. When it's time for customer workspaces and volume terms, book the embedded-pricing call.

No credit card required · API access on all plans · Cancel anytime