How to do customer discovery on Twitter / X
Do customer discovery on X by collecting real questions and jobs from public posts, then talking to those authors.
Short answer. Discovery on X is reading the job in public. DROP gathers posts around your site or a rival’s so you interview from evidence, not from a blank calendar.
Skip the survey link
People already wrote the problem. Asking them to fill a form is extra work. Start from their sentence.
Choose a window without assuming complete coverage
The initial logged-in preview samples recent needs on X and Reddit, then older posts or alternative wording when useful. Paid monitoring adds LinkedIn and Threads and searches up to 30 days within the allowance. Neither mode promises a complete archive or a matching result.
Write notes from the post, not the persona
“SaaS founder in NYC” is a guess. “I cannot get my first ten paying users without an audience” is a brief.
Write a hypothesis and a disconfirming example
Use four fields: assumed job, source evidence, alternative explanation and next question. Illustrative hypothesis: small teams need feedback deduplication. A request for duplicate-vote merging supports investigation; a request for enterprise approval controls may contradict your initial audience assumption. Preserve both. Ask about current workarounds before treating either post as demand for your product.
Find a research question
Illustrative exercise — not a real customer result.
Worth checking: “I reconcile bookings manually every night” suggests asking what cannot be automated today.
Do not count: A profile describing someone as a founder does not tell you their current problem or buying authority.
Next step: Record the workflow and one unknown to ask about. A conversation can invalidate your idea as well as support it.
Try it step by step
- Collect the author’s stated task and exact obstacle before assigning a persona.
- Check which capability of your product addresses it and identify anything you still need to ask.
- Use a relevant, permission-respecting conversation to test that interpretation. Track learned constraints as well as positive responses.
Using DROP for this task
DROP is a social listening and customer discovery tool for product makers and content creators. Paid monitoring supports 4 platforms: X, Reddit, LinkedIn and Threads. Focus includes up to 500 post reads per paid period; Growth includes up to 3,000. Reads are not guaranteed leads. The initial logged-in preview samples X and Reddit.
Inspect a product example · Inspect a creator example · Pricing and limits. Both examples are fictional and need no login. Read a real DROP snapshot with original-source excerpts and limitations.
Can a profile replace an interview?
No. A public post provides context; it cannot establish unstated needs, private attributes or a future purchasing decision.
How should I measure discovery quality?
Count distinct relevant conversations and what you learned from them. Keep raw posts, reviewed opportunities and paid conversions as separate measures.