Revenue does not disappear uniformly. It concentrates at a single stage, for specific reasons. Most operators are looking at the wrong one.
Every home improvement operator running meaningful marketing spend has the same collection of numbers on a dashboard somewhere. Leads generated. Appointments set. Demos run. Closes signed. Monthly revenue. Cost per lead. Every one of those numbers is accurate. None of them, on their own, identifies the stage where the business is losing the most money.
This is not a reporting problem in the technical sense. The CRM works. The spreadsheets add up. The metrics update on schedule. The issue is that reporting was built to describe what happened, not to isolate what is suppressing everything that happens after it. Those are different jobs. Most operators are asking their reporting to do a job it was never designed to do.
Start with a simple observation. In a well-run funnel, conversion rates at each stage are largely independent of one another. Set rate reflects activation — how many inbound leads reach a real conversation. Show rate reflects scheduling discipline and follow-through. Close rate reflects sales execution in the home. Cancel rate reflects contract-stage friction. Each stage is governed by different variables and solvable with different interventions.
But the stages are not equal in consequence. A failure at stage four — a cancel rate that is five points too high — costs the business the margin on those specific jobs. A failure at stage one — a set rate that is thirty points too low — costs the business everything that would have happened downstream of those appointments. The set appointments that never got scheduled. The issued appointments that never ran. The closes that never happened. The revenue that never entered the system at all.
A first-stage failure is not one leak. It is a multiplier on every subsequent leak the business does not even get the opportunity to experience.
This is the pattern that repeats across audited operations, regardless of market, vertical, or price point. A business with a set rate below forty percent and close rate at benchmark is not a business with a sales problem. It is a business whose sales team is performing correctly on a fraction of the demand the business paid for. The reporting does not flag this because the close rate looks fine. The close rate is fine. The problem is upstream of where the reporting is looking.
Across 31 audited home improvement operations, 9 markets, $23M in tracked marketing spend. Most operators are running at a fraction of what the same spend and the same team could produce, if the first-stage failure were resolved.
Thirty-eight percent is not a sales problem. It is a mathematical description of what happens when the first-stage gate fails quietly while every stage after it performs as expected. The downstream performance is real. The revenue it produces is real. And it is a fraction of what the same operation would produce if the gate opened.
Here is what the four funnel stages actually look like, and what each one is actually measuring:
The percentage of inbound leads that reach a qualified sales conversation. Governed by speed-to-contact, lead source fit, and setter execution. A failure here suppresses every stage that follows — because the operation never gets the chance to run those demos, close those jobs, or recognize that revenue.
This is where most operators are losing the most money, and where the reporting tells them everything is fine.
The percentage of set appointments that result in the sales rep sitting with a qualified buyer. Governed by confirmation cadence, appointment-setting discipline, and lead temperature at the moment of booking. Rarely the primary constraint — but reliably a secondary drag when activation is already weak.
A show rate below 70% usually reflects the quality of the activation upstream, not the scheduling itself.
The percentage of qualified demos that produce signed contracts. Governed by presentation quality, product fit, financing options, and rep skill. This is the stage most operators watch most closely — and the stage that is least likely to be the actual constraint in a business whose reports look healthy but whose revenue is suppressed.
When close rate is within benchmark and revenue is below target, the problem is not here. It is upstream.
The percentage of signed contracts that survive to installation and collection. Governed by deposit structure, contract clarity, install scheduling, and post-sale friction. A cancel rate above ten percent erodes margin on jobs already won — but the erosion is bounded by the revenue that actually made it through the previous three stages.
Cancel rate compounds the losses of the earlier stages. It rarely causes them.
The argument is not that close rate does not matter, or that cancel rate does not matter. Both matter. The argument is that they matter conditionally — they matter in the context of whatever volume survived the earlier stages. And if the earlier stages are suppressing volume, improvements downstream produce diminishing returns on an already-diminished base.
Fixing the wrong stage is worse than not fixing anything. It consumes time, attention, budget, and management focus on a problem that, resolved, does not change the outcome. The business doubles its close rate on forty demos a month. The same business, unfixed at stage one, would have been running one hundred demos with the same team and the same spend.
The constraint is not where the number looks worst. It is where the number suppresses every number after it.
This is the reason standard reporting — CRM dashboards, agency summaries, sales pipeline reviews — does not identify the constraint. Reporting is stage-local. It tells you what happened at each stage. It does not tell you which stage is suppressing the rest. That requires a different kind of measurement: one that compares each stage to a structural benchmark, quantifies the downstream cost of the gap, and sequences the stages by consequence rather than by severity.
That measurement is not complicated. It is just not built into the systems most operators already use. The CRM records events. The agency dashboard describes the marketing layer. Neither one is structured to produce a constraint analysis — and neither was designed to.
The practical implication of all of this is simple. If you are an operator running meaningful marketing spend, and the reports say everything looks fine but the revenue does not feel like what the spend and the team should be producing — the reporting is probably correct and the feeling is probably correct. They are measuring different things.
The stage that suppresses every stage after it is not visible in the reporting. The reporting describes the output of the system. Finding the constraint requires a different kind of question — one that asks not what happened, but what did not happen, and what it cost.
The 38% median referenced above was produced by a diagnostic system built from the same data this thesis draws on: 31 audited home improvement operations, 9 markets, $23M in tracked marketing spend. The diagnostic accepts nine numbers from the last thirty days of an operation's performance and returns a constraint analysis — the stage most suppressing revenue, the monthly dollar cost of the gap, and three directives in execution order.
It is open to any operator who wants to run their own numbers through it. $97.
Run the diagnostic →