Roofing Sales Manager KPIs and Pipeline Metrics to Track

Most roofing sales managers can tell you last month's revenue and how many jobs got installed. Far fewer can tell you, on a Monday morning, which rep is about to miss quota, which lead source is quietly burning money, or how much approved claim money is sitting unbilled in their own system. Those three answers are the difference between managing a team and just watching one.

The gap is almost never effort. It is that the numbers a roofing team naturally collects — jobs sold, revenue, maybe a close rate someone calculated once — are lagging indicators. They tell you what already happened, after it is too late to change it. The metrics that let you actually steer a sales team are the ones one or two steps upstream: how fast a new lead gets contacted, how many inspections turn into a real estimate, how long a supplement sits before someone touches it again. Those move first, and revenue follows them by weeks.

What follows is the scorecard a roofing sales manager can actually run a team on. Every stage of the pipeline, the specific KPIs that matter at each stage, honest benchmarks and the ranges good teams land in, the math worked out with real numbers, and the operational habits that keep the data clean enough to trust. It is built for retail and storm/restoration roofing alike, and it flags where the two diverge. Where the topic touches insurance claims and supplements, it stays strictly on the side a contractor is allowed to operate on — documenting your own scope and writing an accurate estimate — and it spells out the lines you legally cannot cross.

Why most roofing KPI dashboards are useless

Before the list, a warning. A dashboard with twenty-five metrics on it is a dashboard nobody reads. The failure mode is collecting everything and acting on nothing. A second failure mode is worse: tracking metrics that are real but lagging, then being surprised when they go bad.

Three rules separate a scorecard that runs a team from a wallpaper dashboard.

Track the input, not only the output. "Revenue" is an output. "Inspections set per rep per week" is an input. You can coach an input on Tuesday. You cannot coach revenue — it already happened. Every output metric on your board should have at least one input metric feeding it that a human can change this week.

Every number needs a denominator and a stage. "We got 40 leads" means nothing without "and set 22 inspections from them." A close rate of 35% is meaningless until you say close rate of what — of leads, of inspections, of estimates presented? Half the arguments in roofing sales meetings are two people using the same word for different denominators. Define the stage gates once, in writing, and make everyone use them.

Speed is a metric, not a vibe. The single most predictive number in most roofing pipelines is how many minutes pass between a lead arriving and the first real contact attempt. It is measurable, it is coachable, and it is usually the cheapest fix available. Teams that ignore it leave the most money on the table.

With those rules set, here is the pipeline.

The roofing sales pipeline, stage by stage

You cannot measure conversion without agreed-upon stages. The exact names matter less than the discipline of moving a deal forward only when a real event happens — not when a rep "feels good" about it. Here is a clean six-stage pipeline that fits both retail and storm work, with the entry event that earns each stage.

Two discipline points make or break this.

An immutable first-touch source. The lead source recorded at stage 1 must never be overwritten later. When a rep marks a door-knock lead as "referral" three weeks in because the homeowner mentioned a neighbor, your entire source-attribution and cost-per-lead math turns to noise. Lock the original source at creation and let later context live in a separate field. This is the most common silent corruption in roofing CRMs, and it makes every marketing decision downstream a guess.

A loss reason, every time. A deal cannot move to Lost without a reason from a fixed list: price, went with competitor, claim denied by insurer, no decision / ghosted, not a fit, timing. Free-text loss reasons are useless in aggregate. A structured list turns your loss column into the most valuable coaching data you own.

Retail vs. storm: where the pipeline forks

A retail re-roof pipeline is mostly linear: lead, inspect, estimate, present, close. Storm and restoration work adds a parallel track that runs after the sale on the production and supplement side, and that track holds real money. A storm pipeline effectively has two pipelines stapled together — the sales pipeline above, and a claim/production pipeline that runs claim documented, estimate written, scope agreed by carrier, built, supplemented, depreciation released, paid. Most roofing teams measure the first pipeline obsessively and the second one barely at all, which is exactly backwards from where the leaks are. We will come back to the second pipeline, because it is where the largest unmeasured dollars usually hide.

The conversion KPIs: where deals die and money leaks

These are the rate metrics. Each is a ratio between two adjacent stages, and each one tells you something a rep can change.

1. Speed-to-lead (time to first contact)

Definition: Minutes (not hours) between a lead being created and the first genuine contact attempt — a call placed, a text sent, an answered conversation.

Why it leads everything: Decades of cross-industry response-time research point the same direction — the odds of reaching and qualifying a lead drop sharply when first contact slips from minutes into hours. In roofing, where a storm-chasing competitor may be three doors down the same afternoon, the decay is brutal. A lead contacted in five minutes and one contacted the next morning are not the same lead anymore.

How to track it: Timestamp lead creation and timestamp first outbound attempt. Report the median, not the average — one rep who forgot a lead for four days will wreck an average and hide that the team is mostly fine, or mostly not.

Honest benchmark: Aim for a median first contact under 5 minutes for inbound web leads during business hours, under 30 minutes for after-hours handled by an answering setup. If your median is measured in hours, this is almost certainly the single highest-ROI fix on your board, ahead of any new lead source.

Worked example: A team buys 100 web leads a month at $80 each — $8,000. At a 4-hour median response they reach 55 and set 18 inspections. They cut median response to 8 minutes with no other change and reach 78, setting 27. Same spend, same closers, nine extra inspections a month. At a 35% close rate and a $12,000 average job, that is roughly three extra jobs and about $36,000 in revenue from a scheduling change, not a marketing change.

2. Contact rate

Definition: Of leads created, the percentage you actually reach with a two-way conversation.

Why it matters: A low contact rate masquerades as a lead-quality problem. Reps will swear the leads are junk when the truth is half were never reached because nobody called back fast enough or enough times. Separate "reached and not interested" from "never reached" — those demand opposite fixes. The first is a list or messaging problem; the second is an effort and cadence problem.

How to track it: Contact rate = (leads with a logged two-way conversation) / (total leads). Pair it with attempts-per-lead. A healthy retail inbound lead deserves at least 6–8 contact attempts across call, text, and email over the first two weeks before you call it dead. Most teams quit after two.

Benchmark range: 60–80% contact rate on inbound web/form leads is achievable with a real cadence. Aged or list-based outbound runs much lower and that is fine — just do not compare them on the same axis.

3. Lead-to-set rate

Definition: Of leads, the percentage that become a booked inspection with a date and time.

Why it matters: This is the first true qualification gate and the cleanest measure of your intake/SDR function. It blends contact rate and persuasion. A rep with great close numbers but a weak set rate is leaving deals upstream they never get to pitch.

Benchmark range: 35–55% of contacted inbound leads turning into a set inspection is a reasonable target band; against all leads it lands lower because of the contact-rate drag. Always state which denominator you mean.

4. Set-to-inspect (show) rate

Definition: Of inspections set, the percentage where the inspection actually happens.

Why it matters: No-shows and cancellations are pure wasted windshield time and a hidden capacity killer. A 70% show rate means three of every ten scheduled trips are dead miles. Confirmation texts the day before and morning-of, plus tightening the set-to-inspection gap to within 48 hours, are the standard levers.

Benchmark range: 80–90% show rate is attainable with confirmation discipline. Below 70% and you have a scheduling or qualification problem, not a closing problem.

5. Inspect-to-proposal and proposal-to-close

Definition: Of inspections, the percentage that produce a presented proposal; and of presented proposals, the percentage that sign.

Why split them: Folding these together hides the most common rep failure mode — inspecting roofs and never presenting a price. A rep at a 90% close rate on proposals presented but who only presents to half the roofs they inspect is not a great closer; they are an avoidant one. The split exposes it instantly.

Benchmark range: For retail, an inspect-to-proposal rate of 70%+ and a proposal-to-close rate of 30–50% are healthy bands. Storm/restoration close rates run higher when there is an active, carrier-acknowledged claim, because the homeowner already has a reason to act. Track them separately or the blended number lies to you.

6. Overall close rate

Definition: Won deals divided by some agreed denominator — usually inspections, sometimes qualified leads.

The trap: "Close rate" is the most abused metric in roofing because nobody agrees on the denominator. Close rate of leads, of inspections, and of proposals can be 15%, 35%, and 45% for the same rep in the same month. Pick one denominator for the headline number (inspections is the most honest for a sales manager), publish it, and never quietly switch it.

Here is the full funnel worked end to end so the compounding is visible.

Eleven jobs from a hundred leads. Now notice where the biggest absolute drop is: lead to contacted (28 lost) and contacted to set (32 lost). The close rate at the bottom (41% of proposals) is healthy. If this manager spends the month coaching closing technique, they are polishing the strongest part of the funnel while 60 leads die upstream from slow contact and weak setting. The funnel table is how you avoid that mistake. Always look for the largest absolute loss, not the lowest percentage.

A word on benchmarks and where to get honest ones

Every range in here is a working band, not a law. Roofing conversion varies with market, lead source, ticket size, season, and whether you are retail or storm. Two things keep benchmarks honest. First, your own trailing 12-month numbers are a better benchmark than any industry figure, because they control for your market and your team — the goal is to beat last quarter, not a number from a webinar. Second, be suspicious of any close rate or cost figure quoted without a denominator and a source; if a vendor claims a 60% close rate, ask close rate of what, measured how. The only benchmarks worth pinning to a wall are the ones you can reproduce from your own pipeline data. Build the band from your own history first, then compare outward.

The speed and efficiency KPIs

Conversion tells you whether deals advance. These tell you how fast and how efficiently, which is where cash flow and rep capacity live.

Sales cycle length

Definition: Median days from lead created to contract signed, for won deals.

Measure the median and watch the trend, not the absolute number — retail and storm cycles differ wildly. A lengthening cycle is an early warning: deals are stalling somewhere, often at proposal. Segment cycle length by lead source and by rep. A source with a great close rate but a 60-day cycle ties up working capital and rep attention that a 14-day source would free.

Pipeline velocity

This is the most underused number in roofing sales, and it bundles four KPIs into one dollar figure that answers "how much revenue is this pipeline generating per day."

Pipeline velocity = (open qualified deals x win rate x average deal value) / sales cycle length in days

Worked example: 60 open qualified deals, 38% win rate, $13,000 average job, 30-day cycle.

(60 x 0.38 x 13,000) / 30 = 296,400 / 30 = $9,880 per day

The power is in the sensitivity test. Improve win rate from 38% to 43% and velocity rises to about $11,180/day. Cut the cycle from 30 to 24 days at the original win rate and it jumps to about $12,350/day. Now you can argue, with numbers, whether to invest in closing training or in speeding up the proposal step — instead of guessing. Run this monthly per rep and per source.

Average deal value and gross margin per job

Revenue per job is obvious; margin per job is the one that actually matters and the one roofing teams skip. A rep who sells high volume at thin margin because they discount to close can outshine, on a revenue board, a rep who protects price and sells profitable jobs. Put estimated gross margin next to revenue on every rep's line. If your system can pull job costing back in after production, track actual margin against the estimate — the gap between sold margin and delivered margin is where a surprising amount of profit quietly disappears.

Activity metrics (use with care)

Doors knocked, calls made, inspections run, follow-ups logged. These are leading and coachable, but they are inputs, not goals. The danger is rewarding activity for its own sake — a rep can knock 300 doors and set nothing. Use activity metrics diagnostically: when a rep's results dip, activity tells you whether the problem is effort (low activity) or skill (high activity, low conversion). Those need opposite conversations. Never rank reps on raw activity alone; rank on outcomes and use activity to explain them.

The money KPIs: cost per lead, CAC, and cost per win

This is where marketing spend meets reality. Roofing teams routinely track cost per lead and stop there, which is like judging a fishing trip by bait price.

Cost per lead (CPL)

Total spend on a source divided by leads generated. Useful, but dangerous alone — the cheapest leads are frequently the worst, and a low CPL on leads nobody can reach or close is a trap.

Cost per acquisition / cost per win

This is the number that should drive your budget.

Cost per win = total source spend / jobs won from that source

Worked comparison: Two sources, $10,000 each this month.

Source A has a CPL four times cheaper and looks like the winner on the metric most teams watch. On cost per win and on revenue, Source B wins outright — better close rate and bigger jobs. A manager optimizing on CPL would cut exactly the wrong source. This single table reframes most roofing marketing budgets.

Return on ad spend and the actual-vs-estimate discipline

ROAS = revenue from a source / spend on it. The harder, more honest practice is tracking projected results against actual results per campaign. When you launch a mail drop or a canvass push, you estimate response, leads, and wins. Recording the estimate and then the actual closes the loop and kills the recurring fantasy that "that campaign worked great" with no evidence. Over a few cycles the gap between estimate and actual becomes your most reliable planning input. Most roofing teams never write the estimate down, so they can never be wrong, which means they can never get better.

LTV and repeat/referral rate

Roofing is not pure one-and-done. Repairs, maintenance, gutter work, and especially referrals extend the value of a won customer. Track referral rate (jobs sourced from a prior customer) as its own line; it is usually the highest-margin, fastest-closing source you have and the easiest to grow with a deliberate ask. A customer who refers two neighbors is worth several times the cost-per-win you paid to land them, and that math justifies real investment in post-job follow-up.

Where RoofPredict fits: turning this scorecard into something you run, not rebuild

Everything above assumes you can actually see these numbers. In practice most roofing managers reconstruct them by hand — exporting the CRM, pasting into a spreadsheet, arguing about denominators, and producing a dashboard that is two weeks stale by the time it is done. The metrics are right; the plumbing is the problem.

RoofPredict is the operations platform a roofing contractor runs the outreach and revenue cycle on, and it is built around exactly this scorecard rather than bolted on after. A few specific places it does the work described above:

The lead pipeline with an immutable first-touch source. The discipline points that make or break attribution — a locked original source and a structured loss reason — are enforced in the pipeline itself: new, contacting, appointment, inspected, won/lost, with the first-touch source recorded at creation and not overwritable later. That alone fixes the silent corruption that makes most cost-per-source math untrustworthy. Because it syncs two-way with 13 CRMs — HubSpot, ServiceTitan, JobNimbus, AccuLynx, Jobber, Housecall Pro, Salesforce, Pipedrive, Leap, Roofr, SalesRabbit, CompanyCam, plus Zapier and CSV — you do not have to rip out the system your reps already live in to get clean stage data; it reads and writes to what you have.

The results funnel with actual-vs-estimate and cost-per-win. The delivered → views → form → calls → leads → wins funnel is tracked per campaign with cost-per-lead and cost-per-win computed for you, and — this is the part most tools skip — it holds your up-front estimate next to the actual result and against an industry benchmark. That is the actual-vs-estimate discipline above, automated, so the loop gets closed whether or not anyone remembers to write the estimate down. A/B campaign variants let you test mail copy or a list segment and read the cost-per-win difference instead of arguing about which "felt" better.

None of this is magic. The scoring underneath the targeting side is roof-age banding plus storm-exposure heuristics, not a crystal ball, and the platform is honest about that. What it does is make the scorecard a live instrument instead of a monthly spreadsheet excavation.

The pipeline most roofers never measure: production, supplements, and recoverable depreciation

For storm and restoration teams, the sales pipeline is only half the revenue picture. The other half runs after the contract is signed, and it is where the largest pile of unmeasured money usually sits. These are not sales KPIs in the traditional sense, but if a sales manager owns revenue, they own these too — because a job that is sold but stuck unbilled in supplement limbo is not revenue, it is risk.

First, the legal frame, because this is where roofers get themselves in trouble. As a contractor, you may inspect the roof, document damage thoroughly with photographs, and prepare an accurate, Xactimate-aligned estimate to repair your own scope of work. You may state facts about your scope to the carrier. You may not, for a fee, negotiate or "handle" or adjust the claim, interpret the homeowner's policy or what it covers, promise a specific payout or that approval is coming, promise the deductible will be waived or absorbed, advertise a "free roof," or represent the homeowner against their insurer. Those last items are unlicensed public adjusting and, in the case of the deductible and free-roof promises, are illegal in most states under insurance fraud and consumer-protection statutes. The safe operating frame is simple: you document thoroughly, you write an accurate repair estimate, and you hand it to the homeowner. The homeowner files the claim and the insurer decides coverage. Every metric below lives entirely on the documentation-and-estimate side of that line.

Supplement aging

Definition: For each open supplement (additional scope you have documented and submitted on your own repair estimate), the number of days since it was submitted or last touched.

Why it is the biggest leak: A supplement is money for work you have documented and are entitled to bill, but it requires follow-up to move through the carrier's process. Supplements do not chase themselves. Without an aging report, they silently rot — a documented item submitted in March that nobody followed up on by June is real money you earned and never collected. An aging bucket (0–15 days, 16–30, 31–60, 60+) instantly shows your manager where collectible dollars are stalling and who needs to follow up today.

How to track it: Every open supplement gets a submitted date and a last-action date. Sort descending by age. The 60+ bucket is your triage list every single week. Pair it with a follow-up cadence so nothing in 16–30 ages into 60+ untouched.

Recoverable depreciation released vs. outstanding

On replacement-cost claims, the carrier typically holds back depreciation and releases it after the work is completed and documented. That recoverable depreciation is money you have earned but cannot collect until you submit completion evidence and a final invoice. Teams lose real dollars here purely through dropped paperwork — the job is done, the homeowner is happy, and a five-figure depreciation release is sitting uncollected because nobody assembled the completion photos and final invoice. Track recoverable depreciation as released vs. outstanding, with an outstanding-aging view just like supplements. This is one of the highest-dollar, lowest-effort recovery opportunities in a restoration business.

Deductible tracking

The homeowner's deductible is the homeowner's legal responsibility, full stop. You may not absorb it, waive it, or rebate it — doing so is fraud in most jurisdictions and it is a fast way to lose your license and your business. The legitimate metric is simply tracking that each deductible is invoiced to and collected from the homeowner. A clean deductible-collected line protects you on audit and keeps your books honest; a pile of uncollected deductibles is both a cash problem and a compliance flag.

Packet completeness

Before anything goes to a carrier, a documentation packet is either complete or it is not — photos, measurements, the line-item estimate, code references where applicable, completion evidence for depreciation. Incomplete packets are the number-one cause of delay and pushback. Scoring packet completeness before submission, as a percentage, turns "why is this stuck" into "this packet is at 70%, it is missing slope photos and the final invoice." That is a metric you can act on in an afternoon.

How RoofClaim measures the second pipeline for you

This is the second specific place RoofPredict does the work. RoofClaim is the integrated claim revenue-cycle side of the platform, and it is built to keep that after-the-sale pipeline from leaking — on locked, compliance-gated, contractor-documentation-only templates, so it never drifts into the claim-handling territory the law forbids.

What a supplement or production manager actually does with it:

• Claim intake linked to the home, so the documentation, photos, and estimate all live against the specific roof and job rather than scattered across inboxes and trucks.
• Upload, auto-classify, and OCR of claim documents — carrier and contractor estimates, photos, denial letters, invoices — so the paperwork is searchable and the data is extracted instead of retyped.
• Opportunity detection that maps the estimate's line items against a roofing knowledge base and flags missing scope, code-required items, and missed supplements, each with an evidence anchor and pricing. This is the documentation help, not claim negotiation: it shows you, with evidence, what your own repair estimate may be under-scoping so you can document and price it accurately before it goes to the homeowner.
• Recoverable-depreciation autopilot that runs the completion-evidence and final-invoice checklist so the depreciation release stops falling through the cracks.
• Supplement aging plus a follow-up cadence and packet-completeness scoring — the exact aging report and completeness metric described above, generated and maintained for you, with the cadence built in so nothing rots in the 60+ bucket.
• Deductible tracking and a claim-inbox email triage that turns the flood of carrier correspondence into a worklist.

The outputs are documents you are entitled to produce as a contractor: supplement packets, depreciation-release letters, deductible invoices, missing-docs letters, and audit reports — all on UPPA-gated, locked templates that keep you on the document-and-estimate side of the line. RoofPredict's role is which roofs likely qualify by age and storm exposure and the photo-and-scope documentation workflow. It never handles the claim. The homeowner files; the insurer decides.

Building the actual scorecard: weekly, monthly, and per-rep

Knowing the metrics is not the same as running them. Here is the operating rhythm that turns the list into management.

The weekly sales-manager scorecard

Keep it to one screen. These are the numbers you review every Monday, per rep and team-total:

1. Median speed-to-lead (last 7 days) — the early-warning metric.
2. New leads and contact rate — is intake working.
3. Inspections set and run — the activity that produces revenue.
4. Proposals presented — the avoidance check.
5. Deals won and dollars sold — the output.
6. Pipeline value and velocity — forward-looking revenue.
7. Aged deals (anything stuck in a stage past its expected time) — the triage list.

For storm teams, add two lines: supplement 60+ aging total and recoverable depreciation outstanding. Those two numbers are often larger than the week's new sales, and almost nobody puts them on the sales board.

The monthly review

Monthly is for the slower-moving, higher-altitude numbers: cost per win by source, ROAS, actual-vs-estimate on campaigns, sales cycle trend, average margin per job, referral rate, and full-funnel conversion by source. This is where budget decisions get made — which sources to scale, which to cut, where the funnel is leaking in aggregate.

The per-rep scorecard and the coaching conversation

Each rep gets the same funnel, so you can see where their number diverges from the team. The diagnostic logic:

Notice the last row. The fix is not coaching the rep — it is fixing the system. Half of what looks like a rep problem in roofing sales is actually a process problem wearing a rep's name, and the scorecard is what lets you tell the difference.

One more diagnostic that managers miss: read the loss-reason mix per rep, not only the close rate. Two reps can both close 35% of proposals while losing for completely different reasons. A rep losing mostly to "price" needs value framing and objection handling. A rep losing mostly to "no decision / ghosted" has a follow-up and urgency problem, not a price problem. A rep losing to "went with competitor" may be slow to the proposal, letting a faster contractor get there first. The close rate alone hides all three; the structured loss reason names them. This is exactly why the loss-reason list from a fixed set, captured every time, earns its keep — it is the cheapest coaching data you will ever collect, and it costs one dropdown click per dead deal.

Setting quota and pipeline-coverage targets

A scorecard without a target is just a thermometer. Work backwards from the revenue goal to set rep targets that are grounded in your own conversion math rather than wishful thinking. If a rep needs $80,000 in sold revenue this month and the average job is $12,000, that is about 7 jobs. At a 40% proposal-to-close rate they need roughly 17 proposals; at a 75% inspect-to-proposal rate, about 23 inspections; at a 50% set-to-inspect-with-shows allowance, enough sets to absorb no-shows. Chain your own conversion rates backward and the activity target falls out as arithmetic, not a guess. Then apply a pipeline-coverage ratio — most teams want roughly 3x the revenue goal in open qualified pipeline to hit quota, because not every open deal closes. If a rep carrying a $80,000 quota has only $90,000 of open pipeline, they are under-covered and the time to fix it is now, not at month-end when nothing can be done about it.

A 30-day implementation plan

If you are starting from a messy CRM and gut-feel management, do not try to track everything at once. Sequence it.

1. Week 1 — define the stages and gates. Write down the six stages, the entry event for each, the loss-reason list, and lock first-touch source. Get every rep to agree. Nothing else works until this is clean.
2. Week 2 — instrument speed-to-lead and contact rate. These are the highest-ROI metrics and the easiest to fix. Set up the timestamps and alerts. Start reporting the median daily.
3. Week 3 — stand up the weekly scorecard. The seven lines above, per rep. Run your first real Monday review off it. Expect the data to be ugly the first few weeks; that is the point.
4. Week 4 — add the money and (for storm) the supplement/depreciation lines. Cost per win by source, and the supplement aging and depreciation-outstanding totals. Now you can see the leaks the sales board was hiding.

After 30 days you have a scorecard you run the team on, not a dashboard you rebuild every month.

Keeping the data clean enough to trust

No scorecard survives dirty data, and roofing CRMs get dirty fast — reps in trucks, fields left blank, deals that sit in "contacting" for two months because nobody marked them lost. A few habits keep the numbers honest enough to act on.

Make the required fields actually required. Source at creation, stage entry events, and loss reason on close should be enforced, not optional. A field that is optional is a field that is blank half the time, and a metric built on a half-blank field is a coin flip. If your system can block stage advancement until the gating field is filled, use it.

Run a weekly stale-deal sweep. Any deal that has not moved or been touched in a defined window — 14 days is a reasonable default for retail — gets surfaced and either advanced, followed up, or marked lost with a reason. Stale deals inflate your pipeline value and rot your velocity number. A pipeline full of zombie deals reports a healthy forward number that is pure fiction, and you will plan against it.

Reconcile sold against installed monthly. Sold revenue and installed revenue are different numbers, and the gap between them — cancellations, claim falloff, jobs that never got built — is itself a metric. A high sold number that does not convert to installed is a warning that reps are booking deals that do not hold, often by over-promising on the claim side, which is both a revenue problem and a compliance one.

Audit attribution quarterly. Pull a sample of won deals and trace each back to its recorded first-touch source. If the recorded source does not match reality, your cost-per-win math is wrong and every budget decision built on it is wrong. This audit is tedious and it is the single best way to catch source corruption before it compounds across a year of marketing spend.

Common mistakes that quietly break the numbers

• Vanity revenue. Celebrating total revenue while margin erodes. Always put margin next to revenue.
• Overwriting lead source. The single most common silent data-corruption in roofing CRMs. Lock first-touch at creation.
• One denominator for "close rate" on Monday, another on Friday. Pick one, publish it, never switch quietly.
• Coaching the strongest part of the funnel. Always attack the largest absolute drop, not the lowest percentage or the part that is easiest to talk about.
• Averaging speed-to-lead instead of taking the median. One forgotten lead hides a whole team's problem behind a bad average.
• Ignoring the post-sale pipeline. Supplements and recoverable depreciation aging out is real, earned, collectible money walking out the door because no metric was watching it.
• Confusing activity with results. Knocking 300 doors is not selling. Use activity to diagnose, never to rank.
• Promising on the claim side. Telling a homeowner the deductible is covered, or that approval is guaranteed, is not a sales tactic — it is fraud or unlicensed adjusting. Stay on documentation and estimating, every time.

What good looks like

A roofing sales manager running this scorecard well can, on a Monday, point to the rep whose set rate slipped and know to role-play the set conversation, not the close. They can tell which lead source earns its budget on cost-per-win even though its cost-per-lead looks high. They can name the supplement that has been aging 47 days and assign the follow-up before it ages another two weeks. They know their pipeline velocity in dollars per day and which lever — win rate or cycle time — moves it most. And they do all of it from a live instrument, not a spreadsheet they rebuild from scratch every month.

That is the whole point of measuring the pipeline: not to have numbers, but to know, every week, exactly where to push. The metrics in here are the ones that tell you. RoofPredict is built to keep them live — the ranked targeting that fills the top of the funnel, the lead pipeline and two-way CRM sync that keep the stage data honest, the results funnel that computes cost-per-win and holds actual against estimate, and RoofClaim's supplement aging, recoverable-depreciation autopilot, and packet-completeness scoring that finally put a number on the second pipeline. If you want to stop reconstructing your scorecard by hand and start running your team off a live one — honestly, with roof age treated as a range and a forecast treated as odds — that is the platform built for it. Book a demo and bring your messiest funnel; that is the one worth fixing first.

FAQ

What are the most important KPIs for a roofing sales manager to track?

Start with speed-to-lead (median minutes to first contact), contact rate, lead-to-set rate, set-to-inspect (show) rate, inspect-to-proposal and proposal-to-close rates, and overall close rate against one fixed denominator. Add the money metrics — cost per win by source and ROAS — and pipeline velocity in dollars per day. For storm and restoration teams, add supplement aging and recoverable-depreciation outstanding, which are often larger than the week's new sales and almost never appear on a sales board.

What is a good close rate for a roofing sales team?

It depends entirely on the denominator, which is why the number gets abused. Of proposals presented, 30–50% is a healthy band for retail. Of inspections, expect lower because not every inspection becomes a proposal. Of all leads, lower still after contact-rate drag. Storm/restoration close rates run higher when there is an active, carrier-acknowledged claim because the homeowner already has a reason to act. Pick one denominator for your headline number, publish it, and never quietly switch.

How fast should a roofing company respond to a new lead?

Aim for a median first contact under 5 minutes for inbound web leads during business hours, and under 30 minutes after hours with an answering setup. Response-time research across industries shows reach and qualification odds drop sharply once first contact slips from minutes into hours, and roofing is worse because a competitor may be on the same street the same day. Report the median, not the average, so one forgotten lead does not hide an otherwise healthy team.

What is the difference between cost per lead and cost per win, and which matters more?

Cost per lead is spend divided by leads; cost per win is spend divided by jobs actually won from that source. Cost per win matters far more because the cheapest leads are frequently the worst. A source with a high cost-per-lead but a strong close rate and bigger jobs can beat a cheap source on both cost-per-win and revenue. Optimizing your budget on cost-per-lead alone routinely leads managers to cut exactly the source they should scale.

How do I calculate roofing sales pipeline velocity?

Pipeline velocity = (open qualified deals x win rate x average deal value) / sales cycle length in days. It collapses four KPIs into one dollars-per-day figure. The value is the sensitivity test: change win rate or cycle length and watch velocity move, so you can argue with numbers whether to invest in closing training or in speeding up the proposal step. Run it monthly per rep and per lead source.

What is a supplement aging report and why does it matter?

A supplement aging report buckets every open supplement — additional scope you have documented on your own repair estimate — by days since it was submitted or last touched (0–15, 16–30, 31–60, 60+). It matters because supplements do not chase themselves; documented, collectible money silently rots without follow-up. The 60+ bucket is a weekly triage list. This is one of the largest unmeasured revenue leaks in restoration roofing, and it lives entirely on the documentation side of the claim, not on negotiation.

Can a roofing contractor handle or negotiate a homeowner's insurance claim?

No. A contractor may inspect, document damage thoroughly, prepare an accurate Xactimate-aligned estimate for their own repair scope, and state facts about that scope to the carrier. A contractor may not, for a fee, negotiate or adjust the claim, interpret the policy or coverage, promise a specific payout or approval, waive or absorb the deductible, advertise a free roof, or represent the homeowner against the insurer — those are unlicensed public adjusting or outright fraud in most states. The safe frame: you document and estimate, the homeowner files, and the insurer decides coverage.

Why shouldn't I let reps change a lead's source after it's created?

Because it silently corrupts every downstream number. When a rep relabels a door-knock lead as a referral weeks later, your cost-per-source, cost-per-win, and ROAS math turns to noise, and you can no longer tell which marketing actually works. Lock the first-touch source at creation and let later context live in a separate field. This is the most common silent data problem in roofing CRMs and it quietly invalidates marketing-budget decisions.

How many times should we follow up with a roofing lead before giving up?

At least 6–8 contact attempts across call, text, and email over the first two weeks for an inbound web or form lead. Most teams quit after two, which is why their contact rate looks like a lead-quality problem when it is really an effort-and-cadence problem. Separate leads you reached and that said no from leads you never reached — the first needs better messaging or targeting, the second just needs persistence and a real cadence.

What's the most common mistake roofing sales managers make with their metrics?

Coaching the strongest part of the funnel. Managers gravitate to closing technique because it is the most visible skill, while the biggest absolute losses are usually upstream — leads that were never contacted fast enough or inspections that were never set. Always find the largest absolute drop between two stages, not the lowest percentage or the topic easiest to discuss, and push there. A close runner-up: tracking lagging output metrics like revenue while ignoring the leading inputs you can still change this week.