ConversionJune 6, 2026·7 min read
Conversation-based lead scoring: the signals that actually predict
Classic scoring on firmographics and page views plateaus. The strongest signals sit in the support and sales conversations you're not mining.
JG
Jérôme Gambiez
Founder, ResponZ
A lead asking a precise question about a paid use case converts 4 to 6× better than an identical-looking lead who only asked about pricing. Conversation-based scoring captures exactly that delta.
Conversational signals to weight
- Mentions of specific business use cases (integration X, workflow Y) — strong intent predictor.
- Questions about security, compliance, contracts — typical of an active buying cycle.
- Personalised demo request vs generic demo.
- Multiple people from the same company chiming in — internal evaluation signal.
- Reply volume over 14 days — sustained engagement vs ping-and-go.
Anti-signals people forget
- Pricing-only questions with no use case.
- Free email domain + unknown company.
- Short conversation pattern (< 3 messages then ghost).
Building the model
With 200 to 300 labelled deals (won / lost), a simple logistic regression on these signals already extracts 80% of the value of a fancier model. Don't wait for a data team to start.Wire scoring to routing
A score is only useful if it triggers action. Useful thresholds: score > 80 → direct assignment to a senior AE; 50–80 → personalised engagement sequence; < 50 → self-serve resources.
Recompute continuously, not in nightly batches
B2B intent windows are hours, not days. Daily batch scoring misses half the opportunities. Re-scoring on every inbound message is very doable and changes the game.