The system was built to produce score-management. Not the operator. The data architecture, the reporting cadence, the comp structure, the inherited tools : all of it was designed around the score because the score was what was available. The diagnosis was never available. So nobody managed to it.
There is a version of this conversation that blames the operator. The metrics exist. The framework is not complicated. The shift from managing net revenue to managing retained revenue is, in principle, a decision. So why hasn't it happened.
The answer is not motivational. It is not a failure of curiosity or sophistication or ambition. The operators running home improvement companies at meaningful scale are not incurious people. They are people operating inside systems that were built, layer by layer, to produce exactly the behavior those systems produce. The score gets managed because the score is what the system was designed to surface. The diagnosis was never surfaced. So it was never managed.
Four structural walls explain this. Each one is independent. Each one is sufficient on its own to prevent the shift. Together, they produce a system so coherent in its score-orientation that managing the diagnosis feels : from inside the system : like managing something that does not exist.
The Four Walls
The signed contract record lives in the CRM. The cancellation and refund record lives in operations or accounting. The financing approval and fallout record lives in the financing platform. The install completion record lives in the field operations system. None of these were designed to talk to each other. None were designed to produce a cohort view of what a signed contract actually became.
The retained revenue calculation requires all four sources connected against a single cohort timeline. That connection has almost never been built in a home improvement operation, not because operators don't want it, but because no single vendor ever had an incentive to build it. The CRM vendor measures activity. The financing platform measures approvals. The accounting system measures settled periods. The diagnosis lives in the gap between all of them, and the gap has no owner.
The dashboards, the end-of-month rollups, the variance reports, the sales pipeline reviews : every standard reporting artifact in a home improvement operation was designed to answer one question: did we hit the number. Not: what is the number actually measuring. Not: what will this cohort produce in 90 days. Did we hit the number this period.
This is not a design failure. The reporting was built for the question being asked at the time it was built. Accounting systems settle periods. Sales dashboards track pipeline. Agency reports summarize spend. Each one is accurate for its purpose. None of them were ever asked to produce a diagnostic view of what the period's signed contracts will ultimately become.
The monthly close is the most score-oriented artifact in the operation. It arrives after the period ends, after the decisions have already been made, after the next period's spend has already been committed. It confirms what happened. It does not predict what is happening. An operation managed to the monthly close is always making decisions on information that is at least 30 days stale, and often 60 to 90 days stale on the cohort outcomes that would actually change those decisions.
This is the most consequential wall. The data wall and the reporting wall are architectural problems. They can be solved with infrastructure. The comp wall is a behavioral problem. It shapes what people optimize for, every day, in every decision they make.
Reps are paid on signed contracts. Not retained jobs. The commission triggers at signature, which means the rep's financial incentive ends at the moment the cancellation risk begins. A rep who closes hard and cancels hard earns more than a rep who closes at a lower rate and cancels almost nothing. The comp plan does not know the difference. It was not designed to know the difference.
Sales managers are measured against month-end revenue. Not cohort outcomes. The manager whose team hit 108 percent of goal in March is rewarded in March, regardless of what the March cohort retains in May and June. The incentive is to produce the score. The diagnostic outcome is not in the measurement.
Marketing directors hit CPL targets while the operation misses retention targets. The marketing director's performance review does not include the cancel rate by source. It includes cost-per-lead, lead volume, and sometimes issued appointment rate. The sources driving the highest cancel rates are often the sources producing the best CPL numbers. The marketing director is being rewarded for producing the inputs that the retention data would reveal as the most expensive inputs in the operation.
Each of these comp structures made sense when it was designed. Reps need to be paid promptly. Managers need a clear target. Marketing directors need measurable KPIs. The structures were not designed to be misaligned with retention. They were designed before retention was the thing being measured. The misalignment is a consequence of building incentive architecture around the score before the diagnosis existed as a concept.
Every operator in home improvement was trained by someone who was trained by someone who managed the score. The vocabulary of the industry is score vocabulary. Net revenue. Close rate. Cost per lead. Monthly volume. These are the numbers that get reported in the trade press, discussed at conferences, cited in benchmarks, and used to evaluate acquisitions. The industry's shared language is the language of the score.
The diagnostic frame : retained revenue rate, cohort tracking, post-sale retention by source : is genuinely new vocabulary in the industry. Not new as a concept in finance or operations broadly, but new as applied language in home improvement specifically. Most operators have never heard a peer use the phrase "retained revenue rate by cohort." It is not the language of any platform they use, any conference they attend, or any benchmark report they read.
This matters because mental models are not changed by information alone. An operator who receives a retained revenue analysis does not immediately begin managing to it. They receive it through the filter of the mental model they have been operating with for ten or twenty years. The analysis has to be translated into score language before it can be acted on, which often strips out the diagnostic content that made it valuable in the first place.
The mental model wall is the slowest wall to move. Infrastructure can be built. Reporting can be redesigned. Comp structures can be adjusted. The inherited frame through which an operator reads their business is not a switch. It is a gradual replacement of one vocabulary with another, and it requires repeated exposure to the diagnostic frame before the frame becomes the operating reality rather than an interesting external perspective.
Why This Matters for the Diagnosis
The four walls are not sequential. They do not need to be dismantled in order. But they interact. An operator who builds the data architecture still has reporting that presents the score. An operator who redesigns the reporting still has comp structures that reward score-management. An operator who adjusts comp still operates inside an industry whose shared vocabulary is the vocabulary of the score.
The difficulty is not motivational. The system was built to produce score-management. Understanding that is the first diagnostic act.
This is worth stating plainly because the alternative framing : that operators who manage the score do so out of laziness or incuriosity or resistance to change : is both inaccurate and unhelpful. It locates the problem in the operator rather than the architecture. It suggests that the fix is persuasion rather than infrastructure. And it misses the actual reason the shift is hard, which means it produces interventions that address the symptom rather than the cause.
The operators who have made the shift did not do so because they were persuaded. They did so because something forced a confrontation with the architecture : a diligence process that exposed the data gap, a quarter where the cohort outcome diverged sharply from the monthly close, a financing partner who began reporting cancellation rates by source. Something made the diagnostic data visible when the system had always kept it invisible. The visibility changed the management behavior. Not the persuasion.
When the Forcing Function Is the Market Itself
The operators who shifted because diligence forced them through it are not unusual cases. They are early cases. The diligence lens is no longer a transaction-specific framework. It is the lens the institutional capital in this industry has adopted as the standard for evaluating an operation's value.
A diligence team opening a home improvement operation does not ask what the net revenue closed at. They ask what the cohort that produced it retained. They do not ask what the blended CPL is. They ask what the cost per deposited dollar is by source. They do not ask what the close rate looks like. They ask what install-to-deposit conversion looks like at 90 days. The questions are not different from the questions in this article. They are the same questions, asked by the entity with structural authority over the operation's next chapter.
This is precisely the lens that PE firms evaluate opportunity through.
The diligence questions are not specific to a transaction event. Some operators are being measured by this lens because they took capital. Others will be measured by it when they raise it. The rest are being measured by it whether they know it or not, because the market has already adopted it as the standard.
Four walls. Each one structural. Each one the product of rational decisions made before the diagnosis was the thing being measured. None of them the operator's fault. All of them the operator's problem.
See what managing past the four walls looks like in operation.
View the Sample Brief →