Four kinds of signal, with examples
The examples below are illustrative. They show how to reason about public statements; they are not quotes from real prospects. Context, date, and product fit matter more than a particular keyword.
| Public statement | What it supports | What still needs checking |
|---|---|---|
| “What do people use to schedule follow-ups for a small team?” | An active search for a solution | Workflow, budget, and whether the question is still open |
| “We are replacing our CRM next month.” | A current switching event | Who decides, required integrations, and migration needs |
| “I spend Friday rebuilding this report.” | Recurring pain and a manual workaround | Whether they want a new tool or a better process |
| “Interesting product!” | Attention or curiosity | A relevant problem and willingness to act |
The five checks behind a useful shortlist
DemandBolt’s qualification model considers pain, fit, timing, reachability, and evidence quality. Treat each dimension as a question you must answer from the source or a conversation. A score is a prioritization aid, not a probability of conversion.
- Pain: is a concrete problem described, and does it recur or carry a cost?
- Fit: can your current product solve the stated job for this buyer?
- Timing: is there a recent request, deadline, switch, or active decision?
- Reachability: is there an appropriate public or professional route for a relevant response?
- Evidence: can you inspect the original source, its date, and enough surrounding context?
Why plausible signals fail
A job posting can suggest a team’s priorities without meaning it wants software. Funding can change spending capacity without creating the problem you solve. A complaint can be real while the writer has no authority to change the process.
Old threads, copied posts, promotional replies, and unresolved identity assumptions lower confidence. Do not fill those gaps with generated details. Preserve the unknowns and choose a next action that tests them.
Disqualifiers should override enthusiasm. If the person needs an unsupported platform, asks for an outcome you cannot deliver, or cannot be contacted appropriately, do not raise the score just to make the list longer.
Capture enough evidence to review your decision
Keep a source link and capture date, a concise factual summary, the buyer’s stated job, your fit hypothesis, and the most important unanswered question. You should be able to explain the selection without asking someone to trust an opaque model.
Focus on intentionally public professional and business context. Do not prospect using private or sensitive personal traits. Public availability alone does not make every use appropriate.
Match the action to the strength of the evidence
For a clear, active recommendation request, a short relevant answer or invitation to a worked example can be appropriate. For a vague complaint, first understand the job. For an old resolved issue, a new sales message is unlikely to help.
After the conversation, update the decision using what happened. Record a correction if the person was a poor match. Your research system becomes more useful when it remembers negative evidence as well as positive replies.
The buying-signal worksheet compares evidence across the five checks. It gives a transparent review recommendation and a list of gaps; it does not manufacture prospects or purchase intent.