1 min lesson
Scoring and routing - make it explainable
Rebuild the parts of "Scoring and routing - make it explainable", then say why each one matters.
Step 1 of 2
Scoring and routing - make it explainable
A score nobody can explain gets ignored by reps and overruled by leadership. Keep it a transparent weighted sum to start, with thresholds that map cleanly to routing actions.
A startable ICP score → routing map
- Signals
- Company size band, funding stage, tech-stack fit, title seniority, engagement recency.
- Form
- Transparent weighted sum to 0–100; resist a black-box model until you have labeled outcomes.
- 75+
- Route to a named AE, round-robin within the matching segment/territory, SLA timer starts.
- 40–74
- Route to a shared queue or SDR for qualification.
- <40
- Drop to automated nurture; never silently discard.
- Fallback
- No rep matches the territory → escalate to a default owner, alert and log; never 404 a lead.
Reliability is part of the design, not an afterthought
This is production plumbing, so a senior answer covers failure modes before being asked: idempotent capture so retried webhooks don't double-create, retries with backoff on provider timeouts, dead-letter handling for leads that fail enrichment and rate-limit awareness so a vendor's 429 doesn't drop leads on the floor. A candidate who designs only the happy path is designing a demo, not a system.