Back to blog
EN7 min read·May 20, 2026

Demand Discovery for SaaS: How to Find Real Buying Signals

A practical system for monitoring public sources, scoring customer intent, and turning scattered demand into qualified follow-up.

Most early-stage teams do not have a lead problem first. They have a signal problem.

The market is already talking. Founders complain about broken workflows on Hacker News. Developers open issues because a tool does not fit their stack. Buyers ask for alternatives in niche communities. Customers leave clues in support tickets, sales notes, and review threads.

The problem is that these signals are scattered. By the time someone copies a link into Slack, the moment is often gone.

What Counts as a Demand Signal?

A demand signal is not just a mention of your category. It is a moment where someone shows pain, urgency, or active search behavior.

Strong examples:

  • "We are looking for an alternative to..."
  • "This is getting too expensive for our team."
  • "How do you handle this workflow at scale?"
  • "Does anyone know a tool that works with..."
  • "We tried X, but it breaks when..."

Weak examples:

  • Generic market commentary
  • Promotional posts
  • Broad opinions with no clear problem
  • Hiring or vendor spam

The goal is not to monitor everything. The goal is to find the few signals that are worth a human response.

Build a Source Map

Start with five source types:

  1. Public communities where your buyers ask for help
  2. Developer issue trackers where tools break or workflows fail
  3. RSS feeds from niche blogs, forums, changelogs, and newsletters
  4. Manual imports from sales calls, support tickets, and saved links
  5. Approved social or partner sources where commercial monitoring is allowed

Do not start with every channel. Pick the sources where pain is specific and public enough to act on.

Score Before You Respond

Most monitoring tools stop at keyword matching. That creates noise.

A better system scores each item for intent:

  • 9-10: active search for a solution, vendor, migration path, or urgent fix
  • 7-8: clear pain with enough context for a helpful response
  • 4-6: relevant market signal, but weak urgency
  • 1-3: noise, promotion, generic discussion, or no buyer fit

This is where AI helps. Claude can read the full context and separate "interesting topic" from "someone needs help now."

Keep the Follow-Up Human

The best demand discovery workflow does not auto-spam people. It creates a short queue for review.

For every qualified signal, prepare:

  • Source and community
  • Original link
  • Intent score
  • Reason for the score
  • Suggested reply draft
  • CRM stage
  • Owner and next action

The human still decides whether to respond, how to respond, and whether the source allows that action.

The Weekly Operating Rhythm

For a small team, a practical rhythm looks like this:

  • Monday: review new high-score signals and tag themes
  • Tuesday to Thursday: respond to the best opportunities
  • Friday: review which sources produced real conversations
  • Monthly: add or remove sources based on signal quality

The metric is not "number of mentions." The metric is qualified conversations created from real customer pain.

Where Glean Fits

Glean monitors approved sources, RSS feeds, Hacker News, and manual imports, then scores each item with Claude. The best signals flow into a dashboard with reply drafts and CRM follow-up.

That means the team spends less time refreshing tabs and more time helping people who already have the problem.