SignalScore — Lead Scoring Engine

Every inbound lead scored on behaviour and fit — so the team calls the right people first, within minutes.

Marketing · sales ops 35% more closed deals · 52% fewer wasted calls

SignalScore — Lead Scoring Engine — main screenshot

Why we built it

The sales team worked leads top-of-inbox instead of best-first. Hot prospects waited days; tyre-kickers got instant calls. SignalScore reads behaviour (pages viewed, questions asked in chat, downloads) plus fit, and reorders the call queue in real time.

Project scope

Real-time scoring from web + chat-agent signals
Fit rules (industry, size, budget hints) layered on behaviour
Auto-prioritised call queue with 'call within 5 min' alerts
Nurture routing for low scores
Score explanations — reps see WHY it's hot
Feedback loop: closed-won recalibrates the model

Technology used

Next.jsTypeScriptClaude AI (signal analysis)MongoDBWebhooksSlack/email alerts

Result

35% more closed deals · 52% fewer wasted calls

Representative project — client details anonymised for confidentiality. Ask us for a live walkthrough.

Build something like this →

Inside the product

screenshot — Mobile hot-list: the three leads to call right now.