THE KEY TAKEAWAY
A mortgage peer comparison is useful when its population, metric definitions, and reporting period match the question. Treat a gap as a starting point for investigation, not proof of a cause.
Choose the question before choosing the peer group
“How do we compare?” is too broad to guide a useful analysis. A lender reviewing production expense needs a different comparison from a team studying geographic lending activity. Define the decision first: capacity planning, expense review, production mix, or another specific management question.
Then describe the comparison population. Relevant dimensions may include channel mix, business size, product mix, geography, and reporting period. No peer group perfectly matches a business. Record important differences so the comparison remains understandable after it leaves the analyst’s screen.
Match the metric and the denominator
A metric can have the same name across reports while using different calculations. Confirm which expense categories are included in cost per loan and which loan event defines the count. Check whether revenue is shown in dollars per loan or basis points and whether balances and periods are comparable.
For an internal review, retain both your own calculation and the benchmark definition. If a reconciliation is possible, document it. If it is not, label the limitation instead of presenting a precise-looking gap as a directly comparable operating result.
Understand what each data source can answer
The MBA’s performance reports describe mortgage origination and servicing revenues, expenses, and production mix. Those measures can provide operating context when the report population and definitions fit the question.
Public HMDA data serves a different purpose. The CFPB describes it as information about mortgage activity and lending patterns, with public records modified to protect privacy. It should not be treated as a substitute for a peer’s complete internal expense or profitability data.
Before building a chart, list which fields the selected source contains and which conclusions those fields cannot support. Data availability does not make two datasets interchangeable.
Read the distribution, not just the average
An average compresses differences between organizations. Where available, examine the range or distribution, sample coverage, and changes across periods. A comparison can become misleading when a small or shifting group is treated as a stable industry standard.
- What organizations are included?
- Which reporting period and definitions are used?
- How different is our channel or product mix?
- Does the gap persist across comparable periods?
- What internal evidence would help explain it?
Translate a gap into a focused investigation
If a relevant comparison suggests higher expense, investigate the underlying categories and operating context before setting a reduction target. Differences may involve staffing capacity, volume, vendor scope, or accounting treatment. The benchmark identifies a question; source records and business context help answer it.
GlassStack includes peer comparisons in its reporting and analysis offering. A useful conversation starts with the available comparison data, its definitions, and the decisions the lender hopes to inform. No benchmark should imply guaranteed performance or a universal target.
Common questions
Can HMDA show a competitor’s cost per loan?
Public HMDA data describes mortgage activity and lending characteristics. It is not a replacement for internal operating expense reporting.
Should a lender automatically target the peer average?
No. First check comparability and understand why a gap exists, then choose a target that fits the business.
Sources & further reading
External references provide industry context. The practical review framework above is GlassStack’s editorial guidance.