An electrical contractor sends 200 quotes per month. 70 close. The other 130 go into a CRM graveyard — "follow up next week" that nobody follows up on. The owner assumes those were tire-kickers. Some were. But buried in that pile are 15–20 quotes from serious buyers who went with a competitor because nobody called them back on day 3.
The problem isn't lead quality. It's that the business has no model for when a quote dies. They treat every open quote the same — either it converts or it doesn't. But quotes have a survival curve, and it's not flat.
The Survival Curve of a Quote
In medical statistics, survival analysis models time-to-event: how long a patient survives after diagnosis. The same math applies to quotes. The "event" is conversion. The survival function is the probability a quote is still alive (not yet won or lost) at time days after being sent:
From historical data across three trades — HVAC, electrical, plumbing — the empirical survival curve shows the same pattern:
| Days since quote | Still alive | Conversion if reached now |
|---|---|---|
| 0–2 | 92% | 45% |
| 3–5 | 64% | 28% |
| 6–10 | 35% | 12% |
| 11–21 | 18% | 5% |
| 22+ | 8% | < 2% |
The drop-off is brutal. By day 6, two-thirds of your quotes are effectively dead. But most trade businesses don't follow up until day 7 or later — after the window has already closed.
The Hazard Function
The hazard function measures the instantaneous risk of losing a quote at time , given it's survived until then:
For trade quotes, the hazard spikes between day 2 and day 5. That's the window where the homeowner is comparing 2–3 quotes they requested over the weekend. After day 5, the hazard drops — not because the quote is safer, but because the customer has already chosen someone else and just hasn't told you.
Cox Regression: What Accelerates Death
A Cox proportional hazards model identifies which quote features accelerate or delay the "death" of a quote:
| Covariate | Hazard ratio | Interpretation |
|---|---|---|
| Quote > $5K | 1.4× | Bigger jobs die faster — more comparison shopping |
| Sent on Friday | 1.6× | Weekend quotes compete with Saturday-morning Googling |
| Follow-up within 48h | 0.5× | Cuts death rate in half — the single biggest lever |
| Repeat customer | 0.3× | Existing relationships survive 3× longer |
| Photo/video included | 0.7× | Visual quotes build trust, reduce need to shop around |
The 48-hour follow-up result is the headline. It's not "following up is good" — everyone knows that. It's that following up within 48 hours specifically halves the hazard rate. Day 3 is too late for high-value quotes. Day 1 is optimal.
The Prioritized Follow-Up Queue
Instead of a flat list of open quotes, rank by expected remaining value — the probability the quote is still alive times the ticket value:
A $12,000 panel upgrade sent 2 days ago ranks above a $400 outlet install sent 8 days ago — even though the outlet job is "older" and feels more urgent. The model tells the salesperson: stop chasing corpses, call the one that's still breathing.
Revenue Impact
200 quotes/month × $2,800 avg ticket Before (random follow-up, day 5+ avg): Close rate: 35% → 70 jobs → $196K/month After (survival-prioritized, day 1–2 follow-up on top quartile): Close rate: 42% → 84 jobs → $235K/month Delta: +$39K/month → +$468K/year Cost: One dashboard. Zero new leads.
The insight that changes how you build trade SaaS: the CRM isn't a database — it's a triage system. Every quote has a half-life, and if your software can't tell the owner which quotes are dying right now, you're just a prettier spreadsheet.