Blog · Board and management reporting · Insurance
Insurance lead conversion rate tracks inquiry cohorts through quotes to bound business. Define the buying decision being counted, calculate each stage by source, and reconcile agency exports without counting multiple carrier quotes as multiple prospects.
Insurance lead conversion rate measures how many inquiries reach a defined outcome, usually bound business, within the same observation window. Divide distinct inquiries with a qualifying bound policy by the eligible starting inquiries. Then calculate the quote stages separately so agency managers can distinguish a placement problem from an inquiry-quality or follow-up problem.
Inquiry-to-bound conversion rate = Inquiries with qualifying bound business / Eligible starting inquiries × 100
Inquiry-to-quote rate = Inquiries receiving a qualifying quote / Eligible starting inquiries × 100
Quote-to-bound rate = Quoted inquiries reaching qualifying bound business / Quoted inquiries × 100
The unit here is a commercial insurance buying decision, identified by a stable inquiry identifier. A business requesting several carrier options for the same coverage need remains one inquiry. A separate line of coverage or later buying decision needs an explicit rule so staff count it consistently.
Define bound using a documented agency outcome supported by placement records. A sent quote or a verbal expression of interest is not sufficient for this report. The Texas Department of Insurance glossary distinguishes a binder as a temporary insurance contract providing proof of coverage before the permanent policy is received. Align your reporting status with the agency's documented binding process and evidence.
For this example, every counted bound inquiry first received a quote. If your export shows bound business without a quote-stage record, investigate the missing history or report that route separately before multiplying stage rates.
The general stage conversion definition explains the starting-population principle. Insurance reporting adds the need to distinguish inquiry records, carrier submissions, quotes, placements, and policies.
Start with the agency's inquiry register, not just its bound-policy list. A production file contains outcomes but cannot supply the unsuccessful inquiries needed for the denominator.
| Field | What it establishes |
|---|---|
| Inquiry identifier | One commercial buying decision across all related records |
| Inquiry-created date | The cohort the prospect entered |
| Original referral or acquisition source | The source receiving credit under the written rule |
| New-business or renewal flag | Populations to compare separately |
| Coverage line and intended effective date | Comparable buying situations |
| First qualifying quote date | Whether and when the inquiry reached the quote stage |
| First qualifying bound date | Whether and when it reached the bound stage |
| Outcome and reason | Open, declined, lost, or bound at the cutoff |
Join related quote and policy records to the inquiry identifier before counting outcomes. Keep original source stable in this cohort report; a later producer assignment is a different field. Preserve rejection and loss reasons rather than excluding difficult inquiries after seeing their outcomes.
If a quote date or binding date is missing, use a documented resolution process. A blank date does not prove that the stage never occurred. The inquiry file, placement log, and policy register should explain the difference.
A US commercial property and casualty agency reviews new-business inquiries created in January 2026. Each inquiry receives a full 60-day observation window. The figures are illustrative and do not represent an industry benchmark.
Each row counts distinct inquiry identifiers. The quote and bound columns are nested outcomes from the same starting cohort; rates are rounded to two decimal places.
| Source | Eligible inquiries | Quoted inquiries | Bound inquiries | Inquiry to quote | Quote to bound | Inquiry to bound |
|---|---|---|---|---|---|---|
| Professional referrals | 100 | 80 | 48 | 80.00% | 60.00% | 48.00% |
| Agency website | 160 | 96 | 32 | 60.00% | 33.33% | 20.00% |
| Purchased leads | 240 | 96 | 24 | 40.00% | 25.00% | 10.00% |
| Total | 500 | 272 | 104 | 54.40% | 38.24% | 20.80% |
For professional referrals, the calculations are 80 / 100 for inquiry-to-quote, 48 / 80 for quote-to-bound, and 48 / 100 for inquiry-to-bound. The total uses summed counts: 104 / 500 = 20.80%. It does not average the source percentages.
With source in A, inquiries in B, quoted inquiries in C, and bound inquiries in D, the formulas for E2, F2, and G2 are:
=IF(B2=0,"",C2/B2)
=IF(C2=0,"",D2/C2)
=IF(B2=0,"",D2/B2)
Format the cells as percentages and fill down through row 4. The overall inquiry-to-bound rate is:
=IF(SUM(B2:B4)=0,"",SUM(D2:D4)/SUM(B2:B4))
The arithmetic identity is:
Inquiry-to-bound rate = Inquiry-to-quote rate × Quote-to-bound rate
Using the original fractions, (272 / 500) × (104 / 272) = 104 / 500 = 20.80%. Use unrounded rates for the check. The quoted-inquiry count cancels only because both stages refer to the same cohort and every bound inquiry in this example passed through the quote stage.
The population also reconciles to mutually exclusive positions at the cutoff:
| Position after the observation window | Calculation | Inquiries |
|---|---|---|
| No qualifying quote recorded | 500 - 272 | 228 |
| Quoted, without qualifying bound business | 272 - 104 | 168 |
| Qualifying bound business | From bound register | 104 |
| Total starting cohort | 228 + 168 + 104 | 500 |
The unconverted groups include both open and unsuccessful inquiries. Keep those statuses underneath the summary rather than calling every unconverted inquiry a loss. Reconcile the 104 bound inquiry identifiers to the placement register; the number of individual policies may be different.
This differs from lead-to-opportunity conversion, which measures an earlier CRM handoff rather than insurance placement.
Professional referrals lead on both stages here. That supports reviewing their business fit and placement process, but it does not prove that every additional referral would convert at the same rate. Coverage line, intended effective date, carrier appetite, and the completeness of the initial submission can differ by source.
Purchased leads show lower progression before and after quoting. Review reasons separately: inability to reach the prospect calls for a different response from an inquiry outside the agency's target coverage or a quote lost on terms. The agency manager owns the counting policy; producers and marketing owners supply the explanations and proposed actions.
Binding counts also do not establish agency income. The Washington insurance glossary defines commission as the portion of premium the producer keeps as compensation. Use insurance commission-rate variance when the question is earnings, and referral-source income attribution when the question is durable income by source.
Bring a file or Google Sheet with inquiry identifiers, source, creation dates, and quote and binding outcomes. Covirage maps the columns, its tools compute and check the source-cohort rates, and the external AI model explains the results with every figure traced to its rows. Create a free Covirage account, set up your business, and bring a sample of your own export.
See Covirage for insurance brokers for the supporting industry approach.
For inquiry conversion, count the distinct buying decisions that reach binding, rather than every resulting policy. A business may buy several policies from one inquiry. Policy production and customer acquisition are useful separate measures, but they need their own labels and denominators.
Count an inquiry as quoted once when it receives a qualifying quote under your written stage rule. Several carrier options for the same buying decision do not create several quoted inquiries. Retain carrier quote records separately for placement analysis.
Report renewals and new business separately before presenting any combined total. Their starting relationships and buying decisions differ. If you combine them, preserve the separate counts so the reader can distinguish a changing mix from a change in conversion.
They remain unconverted at the stated observation cutoff, but they are not necessarily lost. Retain open, declined, lost, and bound statuses separately. Compare equally mature cohorts so a source with slower placement is not penalized simply for having less follow-up time.