Each brand can report correctly. The numbers still don’t compare.
Each brand reports in good faith, from its own systems and by its own definitions, and each number is right on its own terms. Financial reporting, BI platforms and portfolio-monitoring systems bring those numbers together, and capital, compensation and acquisition decisions are made on what they show.
But bringing numbers together doesn’t make them comparable.This is why it happens, and what has to be established before the comparison can be trusted.
Every acquisition arrives with its own definition of what counts.
Brands may operate on different CRMs, estimating and production platforms, accounting systems, marketing tools and spreadsheets. The same word can describe different events, and the same line can hold different costs.
| The word | At one brand | At another | At a third |
|---|---|---|---|
| A lead | Any contact, including purchased lists | A homeowner who asked for an estimate | Only a scheduled appointment |
| An appointment | A booked time | Confirmed and issued to a rep | Issued, net of cancellations and resets |
| A demonstration | Any completed in-home presentation | Only a presentation with all decision-makers present | Only a presentation that ended with a price |
| A sale | A signed contract | A contract net of rescissions and financing declines | A job that has reached production |
| Gross margin | After sales commissions, booked in cost of goods | Before commissions, booked in sales and marketing | Before installation labor, booked in operating expenses |
| Brand EBITDA | After its share of shared-service costs | Before shared-service costs | Including proposed management add-backs |
The definitions can change even when the metric names don’t.
How to standardize KPI definitions across portfolio companies
Different trades. Different operating models. One requirement before comparison.
HVAC, plumbing and electrical
- Booked calls
- Dispatches
- Completed service calls
- Estimates
- Replacement opportunities
- Memberships
- Sold work
- Installed revenue
Each creates a distinct operating population, with its own timing questions.
Roofing and exteriors
- Inspection
- Claim
- Contingency
- Supplement
- Contract
- Production milestones
- Completion
Retail and insurance work follow different paths. Windows, doors and siding bring their own sales and installation cycles.
Restoration
- Loss
- Emergency response
- Mitigation
- Authorization
- Insurance approval
- Remediation
- Reconstruction
- Collection
Water, fire and mold work can each move through these stages on its own path.
Bath and home remodeling
- Lead
- Appointment set
- Confirmed
- Issued to a rep
- Demonstration
- Contract
- Rescission
- Financing
- Retention
- Production
- Completion
Any of these stages can be where a brand decides the sale happened.
Goverisyn doesn’t assume these businesses should share one operating model. The measurement basis has to reflect how each business actually operates before portfolio comparison can be established.
Which measures compare across trades, and which belong to one
One brand. One month. Four correct close rates.
The problem doesn’t need two different businesses to appear. Take one brand in one month. Nothing is wrong with the records, yet the month produces four close rates, depending on which stage counts as the sale and which population is the denominator.
- Leads
- Scheduled appointments
- Appointments issued
- Demonstrations
- Gross sales
- Net sales, after cancellations
- Jobs in production
Every calculation is mathematically correct. They are not measuring the same thing.
Put brands side by side, and a difference in measurement is easily mistaken for a difference in performance. See what a definition costs
One brand’s records for one month support four correct gross margins.
The P&L has its own vocabulary, and every acquired brand learned it in its own accounting system. Where financing fees, installation labor and sales commissions are booked decides the margin a brand reports, before anyone has done anything differently on a job.
- Installed revenue
- Financing fees
- Materials
- Installation labor
- Sales commissions
Every margin is calculated correctly. Thirty-two points separate them, and none of it is performance.
Where the P&L vocabulary splits
| The item | At one brand | At another | At a third |
|---|---|---|---|
| Revenue | When the contract is signed | When the installation is complete | When final payment is collected |
| Financing fees | A reduction of revenue | A cost of goods | A selling expense |
| Sales commissions | Paid at signing | Paid at installation | Clawed back on cancellation, or not |
| Warranty | Accrued at installation | Expensed when the work is done | Carried by the shared-service company |
| Manufacturer rebates | A reduction of materials | Other income | Recognized when the cash arrives |
| A cancelled deposit | Revenue reversed | Kept as income | Left in backlog until written off |
Goverisyn records how each brand books these items, and reports gross margin both as submitted and on the basis your platform approves, side by side.
A margin is a classification decision before it is a performance measure.
Find the measurement problem while it’s still an internal correction. Not an external credibility event.
Each of these moments turns an internal difference into an outside question. The answer has to be on record before it starts.
Each board meeting
The board asks why two brands differ.
On recordWhat each brand counts, and the evidence behind it.
At each deal
An acquisition adds a brand.
On recordHow it defines close rate, margin and EBITDA, recorded before its figures enter the portfolio total.
Each test date
A lender tests a covenant.
On recordThe basis of each covenant figure, not only its value.
Once, with the most at stake
A refinancing or exit begins.
On recordEach adjustment, with its support and the decision that accepted it.
With a record kept as each change happens
Without one rebuilt for each question
Without a record, the basis is rebuilt each time it is asked for: in spreadsheets, email threads and the memory of whoever built the last version. How much of your reporting is reconstructed each month?
Goverisyn addresses what happens before comparison.
Goverisyn is portfolio measurement governance software. It establishes when performance can defensibly be compared across acquired residential home-services businesses, and reports each figure on that basis. It works alongside the systems a firm already uses, from each brand’s connections, exports, extracts or monthly submissions; the engagements that set it up are separate.
For every measure, Goverisyn records and applies the definitions the client approves, and makes the calculation, source authority, period treatment, reconciliation and supporting evidence explicit. Accounting, CRM, estimating and production systems remain systems of record; fund accounting, valuation and investor reporting remain with the systems built for them.
What a number needs before you can decide on it.
Each level depends on the one beneath it. A number can’t be compared until it is defined, scoped and substantiated.
Compared with each otherSame definition, record-level evidence
Stops hereExclusion open: 2 demos unresolved
Stops hereDefinition differs and is recorded; Excel totals only
Illustrative · Brands A–D from the close-rate comparison on the homepage.
Where a number stops short of comparable, Goverisyn doesn’t round it up. It records where the number stopped and why.
Goverisyn does not manufacture comparability.
Every comparison is identified by what the records support, with the reason preserved.
Restated under the portfolio’s definitions
Where the underlying records allow, Goverisyn restates a brand’s results under the portfolio’s approved definitions.
The limitation travels with the measure
Where the records support comparison only with a stated caveat, that limitation remains attached to the measure.
The number is reported, not compared
Where the records cannot support the portfolio definition, Goverisyn says so, and states which comparison it cannot support.
Missing information is not estimated simply to create the appearance of comparability.
Standardization does not settle comparability.
Giving two fields the same name does not establish that they measure the same population.
Mapping two systems to the same KPI does not resolve differences in definitions, exclusions, timing or status treatment.
Goverisyn makes those differences explicit before they become portfolio conclusions.
From reported numbers to defensible portfolio measures.
From“Here are the numbers from our brands.”
To“Here are the measures we can substantiate, how they were constructed, and where comparison is defensible.”
The first question shouldn’t automatically be:
Why is the second brand underperforming?
It should be:
Are these two percentages measuring the same thing?
Only after that question is answered does the performance comparison become meaningful. The same question, asked of an AI tool
Before you compare performance, establish that you’re measuring the same thing.
Derwin Lucas founded Goverisyn after a career in management and information systems in home services. About Goverisyn