Every brokerage has had the meeting. The managing broker, the finance lead and a team lead each have a different count of last quarter’s deals. Nobody is lying. Each person counted something slightly different, and each was right by their own rules.
The real cost comes afterwards. People stop trusting the reports, so they build their own. Each new spreadsheet adds another private definition.
The fix is not a better dashboard. It is a short, written set of shared definitions that every report must follow.
Where the disagreements come from
Disagreements usually have one of five causes.
1. The event being counted is not the same
“Closed” can mean the day the agreement of purchase and sale became firm, the day conditions were waived, the closing date on title, or the day the commission was received. Each is a real event. Each lands in a different month. A deal that went firm in March, closed in May and paid in June can be counted in three different quarters by three honest people.
2. The date field is not the same
Even when people agree on the event, they may filter on different dates. One report uses the date the record was created in the CRM. Another uses the date an agent last edited it. A third uses the expected close date, which agents tend to update optimistically. Choosing the wrong date field quietly moves numbers between periods.
3. The population is not the same
Does the count include deals where the firm acted for both sides? Referrals sent out to another brokerage? Leases as well as sales? Deals by agents who have since left? Each inclusion or exclusion is a defensible choice. The problem is making it silently and differently in each report.
4. Double counting and splits
A deal with two co-listing agents is one deal for the firm and a share of a deal for each agent. If the team report adds up the agent reports, the firm appears to have done more business than it did. Volume and commission splits make this worse, because a share of a dollar figure is easy to sum twice.
5. The source is not the same
The CRM says one thing, the transaction management system says another, and the accounting system says a third. Each holds part of the truth. Without a rule about which system is authoritative for which fact, people pick the one that suits the question they were asked.
What a good definition looks like
A shared definition is not a sentence in a slide deck. It is a short, structured entry that leaves no room for interpretation. Each one should answer the same questions.
- Name. One name, used everywhere. If the firm says “firm deal” in one place and “accepted offer” in another, pick one.
- Plain-language meaning. One or two sentences that a new agent could understand.
- What counts. The exact event, record type and status that qualifies.
- What does not count. The exclusions, written out, because this is where most arguments live.
- Date that assigns it to a period. One field, named.
- Source of truth. The system and field the number is taken from.
- Owner. The person who decides when the definition changes.
Here is an example of the difference.
Weak definition: “Closed deals: deals that closed in the period.”
Strong definition:
| Element | Entry |
|---|---|
| Name | Closed transaction |
| Meaning | A sale or lease where title transferred or the lease term began, and the firm earned a commission. |
| Counts | Opportunity records with status Closed, transaction type Sale or Lease, and a recorded closing date. |
| Does not count | Referrals paid out to other brokerages; deals that firmed but later collapsed; internal test records. |
| Period date | Closing date on the transaction record, not firm date or payment date. |
| Source of truth | Transaction management system; CRM mirrors it nightly. |
| Owner | Managing broker |
| Splits | Counted once for the firm. Agent reports show credit share separately and are never summed to produce a firm total. |
How to build the set
Most firms run on a small number of measures. Start with those.
- List the numbers people actually argue about. Look back at the last few leadership meetings. Which figures caused a debate, a follow-up email or a “let me check”? Those are your first definitions.
- Collect every current version. For each number, find every report or spreadsheet that shows it. Write down how each one calculates it. This is uncomfortable but essential. It shows you the real spread of definitions in the firm.
- Pick one, deliberately. Choose the definition that best matches how the firm is managed. A brokerage that pays agents on closing should probably count on closing date. A leasing team measured on activity may care more about the date terms were agreed. Write down why you chose it.
- Name the alternatives you rejected. If people still need the firm-date view, give it its own name, such as “Firmed transaction”, with its own definition. Two clearly named measures are far better than one ambiguous one.
- Assign a source of truth for each fact. Decide which system wins when they disagree. Then fix the other systems or stop reporting from them.
- Have the owner sign off. A definition without an owner will drift. The owner is the only person who can change it, and changes are dated.
Making the definitions stick
Build them once, in one place
If each report calculates “closed transaction” itself, the definitions will diverge again within months. The calculation should live in one shared layer, a data model that every report reads from. When the definition changes, it changes once and every report updates together. This is the core reason firms move from spreadsheets to a proper business intelligence model.
Show the definition next to the number
Put a short description, or a link to the definition, on every report tile. The conversation then moves from “your number is wrong” to “this report uses closing date and you were thinking of firm date”.
Control how data is collected
Many arguments start upstream, at data entry. If agents can type any value into a “deal status” field, the definitions downstream cannot be applied reliably. Use controlled pick lists for statuses, transaction types and sources. Make the date fields that define periods mandatory at the stage where they become known.
Keep a change log and reconcile
When a definition changes, record the date, the old rule, the new rule and the reason, or year-on-year comparisons will mislead. Once a month, trace a handful of records from the shared number back to the source systems.
A worked example
The following is a hypothetical example, for illustration only.
A brokerage holds its quarterly review. The sales report shows more closed transactions than the finance report. When the two lists are compared record by record, the gap has three causes:
- Several deals firmed late in the quarter but close next quarter. Sales counted them on firm date; finance did not.
- A few deals were co-listed. The sales report summed agent reports, so each of those deals appeared twice.
- One referral paid out to another brokerage was recorded as a closed deal in the CRM.
Each is an unwritten definition. Once the firm writes down “Closed transaction” and “Firmed transaction” as separate measures, counts splits once at firm level, and excludes outbound referrals, both reports show the same figure. The quarterly review spends its time on why the number moved, not on which number is right.
Shared definitions also make tools more useful. Analytical models and automated summaries are only as good as the definitions underneath them. Good definitions do not replace judgment; they give judgment something solid to work with.
Where to start
- Pick the three numbers that caused the most debate in your last leadership meeting.
- For each, gather every report that shows it and write down how each one counts.
- Choose one definition per number, name the alternatives separately, and assign an owner.
- Publish the definitions somewhere every agent can read them, and put a link on each report.
- Set a monthly half-hour to reconcile one shared number against its source system.
Multiunit builds business intelligence for real estate firms, starting with exactly this work: agreeing the definitions, then building them once so every report uses them. If you would like help with your first set, get in touch through the Contact page.
This article is general information, not legal, financial or brokerage advice. Examples and figures are illustrative unless a source is given.