Bluesky's growth story is hard to ignore: from 10M users in late 2024 to 43M+ registered users by early 2026, with more than 50,000 community-built custom feeds. X API pricing keeps climbing, and brands are increasingly curious — but "lots of users" and "your buyers are here" are different questions.
Here's a data-driven playbook: how to test whether Bluesky fits your ICP, and how to build reach with Starter Packs, feeds, and search — without turning it into another spam channel.
The state of Bluesky: growth and biases
The platform's real strengths for B2B research:
- Open AT Protocol — posts and identities live on a portable, searchable protocol, so third-party tools can index conversations without exorbitant API paywalls
- Custom feeds — 50,000+ community-built algorithmic feeds let you find exactly the topic signal you want
- Starter Packs — bundle up to 150 accounts/feeds into one shareable link; the primary viral loop on the platform
The honest biases: Bluesky skews early-adopter, dev-advocate, journalist, open-web. If your ICP is conservative enterprise IT, this may not be their watering hole — yet. For dev tools, AI infrastructure, and forward-leaning SaaS, it's increasingly relevant.
Step 1 — Phrase-test before you invest
Run the same test you'd run anywhere: does your category get discussed here?
- Search problem phrases (the pains you solve), not just your brand
- Search competitor names and "alternatives to X" phrasings
- Check volume and recency — a busy feed from 2024 is not momentum
If your phrase test returns crickets, skip Bluesky for now. Re-test quarterly, as platform coverage changes fast.
Step 2 — Discovery mechanics that actually work
Custom feeds: Follow feeds built around your niche — AI, devtools, indie SaaS, specific languages. They replace the algorithmic "For You" wall with topic signal.
Starter Packs: Find and join Starter Packs in your niche (100–150 accounts each). Two moves:
- Join — instantly connected to a curated audience
- Create one — curate 100–150 relevant accounts/feeds in your space; it becomes a shareable link others discover you through
AT Protocol search: Because the protocol is open, search surfaces (both native and third-party) index posts broadly. Search intent language like "what do you use for X" or "switching from Y."
Step 3 — The engagement loop
Bluesky rewards authenticity and punishes pitch-drops:
- Add value publicly — answer questions, share teardowns, post genuinely useful thread-ables
- Build reach via Starter Packs — join niche packs and create your own as you grow
- Soft outreach only — after a real exchange, not from a cold DM blast
- Disclose affiliation — same rule as Reddit and every other community
Step 4 — When X still wins vs when Bluesky wins
| Job | X still strong | Bluesky wins |
|---|---|---|
| Vendor announcements | Yes | Partial |
| Large-audience launches | Yes | No (yet) |
| Dev advocates / open-web early adopters | Mixed | Yes |
| Journalists & experimental feeds | Mixed | Yes |
| B2B problem-phrase research | Expensive (API) | Open, cheap |
The Bluesky vs X snapshot has the full reasoning — and reminds you to re-verify quarterly, because both platforms shift.
Step 5 — Measure with a timebox
- Weekly: 2–3 short sessions scanning your feeds (15 min each)
- Monthly: review which posts drove replies, follows, or DMs
- Use the free Bluesky Analytics tool to audit any profile's engagement patterns before you commit weeks there
If nothing converts after 4–6 weeks of honest effort, re-run the phrase test and reallocate. The Bluesky customer discovery guide covers the audit workflow in depth.
Related reading
- Bluesky vs X (Twitter) for B2B signal: 2026 snapshot
- Bluesky for founders: read an audience with free analytics
- Bluesky customer discovery guide
- Mastodon and the Fediverse: market research cautions