Restoration Company KPIs: The 7 Numbers That Predict Whether You Scale or Stall
Revenue is the vanity metric. These seven KPIs — recon margin, AR aging, supplement hit rate, first-pass acceptance, cycle time, revenue per seat, and owner-dependence — tell you whether your restoration company scales or stalls.
Ask most restoration owners how the business is doing and you’ll get a revenue number. Revenue is the vanity metric. It tells you how busy you are, not whether you’re building something that scales, or something that will collapse the second you step away.
Seven numbers actually predict that. Track these and you can see a ceiling coming a quarter out. Ignore them and you’ll keep mistaking motion for progress, right up until the year you do $8M and take home less than you did at $3M.
1. Recon gross margin
The single most honest number in the shop. Not revenue, not net, recon gross margin: what’s left on the reconstruction work after materials, labor, and subs, before overhead. It’s where the money is actually made or quietly given away.
Our cohort runs a 52% recon margin. If you don’t know yours this week, you don’t have a margin problem yet, you have a visibility problem, and that’s worse, because you can’t fix what you can’t see. Margin leaks hide in mis-scoped line items, uncaptured supplements, and subs nobody re-bid in two years. The fix starts with a live, per-job margin view pulled straight from your books, not a month-end spreadsheet you build after the damage is done.
2. AR aging (days sales outstanding)
How long your money sits in someone else’s account after the job is done. Every day a file sits in 60-, 90-, 120-day AR is a day you’re acting as the carrier’s interest-free bank.
This is the leak owners feel but rarely measure. They “know” AR is bad; they can’t tell you the DSO trend. One shop we work with had written off $214K in aged invoices, then an AR follow-up agent collected it in 60 days. Nobody on the team had even tried. Watch the trend, not the snapshot, and chase aging on a cadence instead of when cash gets tight.
3. Supplement hit rate
The percentage of files where you actually capture the supplement the job earned. The scope grows on almost every loss, mid-build discovery, hidden damage, code upgrades, but if nobody owns the supplement, that money walks. Quietly. Every week.
Most shops have no idea what their hit rate is, which means it’s low. Our recent cohort runs 78%. The difference between a 50% and a 78% hit rate on a $5M shop is real six-figure money you already earned and didn’t bill. This is the core of scope defense and supplement workflow: a system that flags the supplement before the file closes, with the justification already written.
4. Carrier first-pass acceptance
What share of your scopes the desk adjuster accepts without kicking back for rework. Every rejected scope is a second touch, a delay, and a hit to your TPA scorecard. First-pass acceptance is the cleanest proxy for whether your documentation is carrier-grade.
Tuned right, this runs around 84% in our cohort. Below 60% and you’re one program review away from losing volume. The fix isn’t arguing harder, it’s writing the scope the carrier can’t cut: IICRC-referenced, photo-cited, mapped to the right line items the first time. (New to how carriers read these files? Start with the Xactimate operator’s guide.)
5. Estimate cycle time
Hours from inspection to a submitted estimate. This is the number that quietly caps your growth, because in most $1M–$15M shops the estimator is the owner, and the owner is finite. When cycle time depends on one person’s nights and weekends, the whole business is throttled to that person’s calendar.
Cut estimate cycle time and two things happen: you get paid faster, and you stop being the bottleneck. Shops that put AI on the first draft run 3–5× the scope output per seat, with the estimator moving from typing to QA.
6. Revenue per production seat
Total production revenue divided by the people who actually move it, estimators, PMs, crew leads. This is your leverage number. Two shops at $5M look identical until you see one runs it with nine people and the other with sixteen. The first one is a business; the second is a job that pays badly.
You grow revenue per seat by removing the low-judgment work, the typing, the lookups, the follow-up calls, so each seat spends its hours on the decisions only a human should make. That’s the whole point of putting AI on the work: leverage, not headcount.
7. Owner-dependence ratio
The percentage of revenue that still routes through you, the estimates you write, the carrier calls you take, the decisions your team won’t make without you. It’s the hardest number to face and the one that decides whether you have an asset or a very stressful job.
No buyer pays a premium for a business that walks out the door when the owner does. Drive this number down and everything else compounds: margin you can see, AR that gets chased, supplements that get captured, scopes that ship without you. That’s the difference between a shop that hits the $1M–$15M ceiling and one that breaks through it.
How these compound
None of these live alone. A clean scope (4) lifts your supplement hit rate (3) and your margin (1). Faster cycle time (5) and more revenue per seat (6) pull the owner out of production (7), which finally frees you to work the AR (2). Move one and you nudge the others. Ignore them and they decay together, that’s the quiet slide from a $3M shop that pays well to an $8M shop that doesn’t.
Where to start Monday
Pick the one number you can’t answer right now. For most owners that’s recon margin or supplement hit rate. Spend an hour getting an honest baseline, not a guess. Whatever you find is your starting line. You can’t manage your way to scale on a revenue number; you manage your way there on these seven.
If you want help getting them visible and moving, that’s exactly what we do, see restoration business consulting for owner-operators, or apply for a 30-minute Operating Diagnostic and we’ll tell you which number is costing you the most.
Keep reading
Read by an R360 operator-founder. Want one at your table? Apply for the diagnostic