Insurance Jobs vs Retail Jobs: How to Do the Cost-Per-Acquisition Math Honestly

Why a Blended Cost Per Acquisition Misleads

Cost per acquisition, or CPA, sounds like one number, but it hides a choice about what counts as the acquisition. A booked appointment, a signed contract, an approved claim, and a collected payment are four different finish lines, and insurance work and retail work reach them in a different order. Divide total marketing spend by total jobs and you get a figure that describes neither business. It flatters whichever job type is quietly subsidizing the other.

The cleaner approach is two ledgers. Every appointment gets tagged at creation with its source and its job type, and every cost follows that tag. When the ledgers disagree, the disagreement is the finding. The CPA glossary entry defines the base metric, and this post is about keeping it honest once insurance work enters the mix.

The Insurance Funnel Has Costs the Retail Funnel Does Not

An insurance job passes through an inspection, a claim, an adjuster meeting, a scope decision, often a supplement, and sometimes a re-inspection before a contract is safe to schedule. Each stage consumes labor, and some stages end with a denial or a partial approval and no job at all. Those losses belong in the insurance ledger. If they are written off as general overhead, insurance work looks cheaper than it is.

Add the outcome split. An inspection that ends in a repair scope and one that ends in a replacement scope produce very different revenue, and the Verisk 2025 averages show how far apart the outcomes can sit: $17,631 for a replacement against $4,699 for a repair. Count the cost of the inspections that ended in a small scope, or the average ticket in your model will be better than the one you actually sell.

The Retail Funnel Hides Its Costs in Time

Retail work can look cheaper per appointment because nobody negotiates with an insurer, but its costs sit in time. Homeowners compare bids, think it over, and ask for a second visit, and every one of those touches belongs to the job. Follow-up labor, repeated quoting, financing paperwork, and the occasional price concession all accrue to the retail ledger, and they stretch the gap between a booked appointment and a signed contract.

Timing also distorts comparisons. Retail jobs often resolve within the sales cycle, while insurance jobs can sit in claim limbo for a long stretch. Compare them by appointment-month cohorts, so you judge January’s insurance appointments only after they have had the time insurance work needs, and never against retail jobs that resolved faster. Otherwise the slower business always looks like it is failing.

Falling Claim Volume Makes Last Year’s Storm CPA Obsolete

Claim volume has moved as well: hail claim activity fell 23.6% year over year in Q1 2026, and roofing claims are down 11.9% through August, per Loveland Innovations’ summary of Verisk data. When fewer roofs generate claims, the cost of finding each qualifying homeowner rises even if your marketing does not change, because the pool you are fishing in has shrunk. A storm-lead CPA from last season is a historical figure, not a forecast.

Rebase it every season, and rebase it by market. A territory that kept its claim flow and one that lost it will carry very different insurance CPAs, and averaging them hides the market where the money is being lost. Retail CPA should be steadier, since it does not depend on weather, which is one more reason to keep the ledgers apart: the steadier one gives you a baseline to judge the volatile one against.

Decide on Contribution Margin, Then Source the Steady Side

The final comparison belongs at the level of contribution margin per completed job: revenue collected, minus materials and labor, minus the fully loaded acquisition cost from that job type’s own ledger. Include no-shows and reschedules in the cost, because they inflate CPA whether or not anyone is billed for them. Review monthly, and write down which assumptions changed so the next review compares like with like.

Once the numbers are honest, the retail line is often the one you can scale on purpose, and it is the line VA Horizon’s roofing appointment setting feeds with exclusive, double-confirmed retail appointments. It runs pay per booked appointment, with no-shows replaced, never billed, which keeps that acquisition cost easy to read in your ledger.

Sources

The external data in this article draws on the sources below. Figures described in the text as estimates or industry triangulations are directional and are not attributed to a single dataset.

• Insurance Business Magazine, Verisk 2025 roof report
• Loveland Innovations, "What Q1 2026 Data Tells Us About Roof Damage and Insurance Claims" (Verisk data)

FAQ