Lead scoring for insurance: rank carriers, MGAs, and agencies by fit and timing
Carriers, MGAs, agencies, and TPAs buy in different ways. Here is how to score each one with public regulatory data, filing and acquisition signals, and buying group coverage.
Lead scoring for insurance is the practice of ranking carriers, managing general agents, agencies, brokerages, and third-party administrators by how well they match the customers you win and how ready they are to buy. Insurance breaks the usual pattern in three ways: the buyer types have little in common, most fit data sits in public regulatory records, and timing follows filings, renewals, and system replacements.
Everything here concerns B2B vendors scoring the carriers, MGAs, agencies, and TPAs they sell to. It is not about scoring consumer insurance leads or about underwriting risk scores. The model never uses policyholder, claimant, or consumer data, and nothing here is insurance or regulatory advice.
Why generic lead scoring fails in insurance
An off-the-shelf model pays out for headcount, a senior title, and a spike in web visits. Across carriers, MGAs, and agencies, it goes wrong in five ways.
| What is different | What to do instead |
|---|---|
| “Insurance” is four or more industries, so a carrier and a ten-person agency land in one model | Score segment first, and run separate models |
| Headcount is a poor size measure: a lean specialty carrier scores below a big call center | Use direct written premium for carriers and MGAs, producers for agencies |
| Licensing limits where a product can be used | Score lines of business and states licensed |
| Core systems decide integration | Make the policy, claims, or agency management system a fit factor |
| Formal review and long cycles make early clicks overstate readiness | Score buying group coverage, and use slow decay |
Start with the first row, because segment changes everything else.
| Buyer type | What it does | Best size measure |
|---|---|---|
| Carrier (insurer) | Holds the risk, files rates and forms, pays claims | Direct written premium by line |
| Managing general agent (MGA) | Underwrites and often handles claims on a carrier’s behalf | Premium placed, programs, carrier partners |
| Agency or brokerage | Sells and services policies for several carriers | Producers, locations, premium placed |
| Third-party administrator (TPA) | Handles claims or administration for carriers and self-insured employers | Claims volume, clients, lines handled |
The NAIC’s Managing General Agents Act defines an MGA narrowly, as a person who manages all or part of an insurer’s business and underwrites premium equal to 5% or more of the insurer’s policyholder surplus. Many firms that use the name sit outside that definition, so score on what the firm does.
Size also runs to extremes. The NAIC reported $974.97 billion in property and casualty direct premiums for 2024, with the top 10 groups holding 51.40% of it. At the other end, the Big “I” estimated 39,000 independent agencies in 2024. The two cannot share one set of weights.
Fit factors for insurance accounts, and where the data comes from
Fit data is usually patchy in a CRM. Insurance gives you more to work with, because carriers, producers, and most TPAs are licensed, and regulators publish much of what they collect.
| Fit factor | Why it matters | Where the data comes from |
|---|---|---|
| Segment (carrier, MGA, agency or brokerage, TPA) | Each buys different things through a different process | Your own classification, checked against license type |
| Lines of business (P&C, life, health; personal or commercial) | A commercial underwriting product needs a commercial book | State department company profiles; NAIC market share reports |
| Premium volume | The best size measure for a carrier or MGA | NAIC market share reports; annual statements through NAIC InsData |
| States licensed | Sets where your product can be used | State department lookups; producer licensing records |
| Number of producers and locations | The best size measure for an agency or brokerage | The agency’s site, discovery calls, licensing data |
| Core system (policy admin, claims, agency management) | Decides whether you can integrate | No regulator collects it. Ask on discovery calls, read job postings, or buy technographic data |
The public sources, and what each one gives a vendor:
- State insurance departments. Insurance is regulated by the states, and the NAIC keeps a directory of 56 insurance jurisdictions. California’s insurance company profiles, for example, show license status, company type, state of domicile, and the lines of insurance a company is authorized to transact.
- NAIC market share reports. The annual reports rank the top 25 groups and companies by countrywide direct written premium for lines including personal and commercial auto, workers’ compensation, homeowners, and life.
- NAIC financial statements. Through InsData, the NAIC sells the annual and quarterly statements of more than 4,800 insurance companies, with up to 10 years of filings, and offers bulk data by quote.
- Producer licensing. NIPR’s Producer Database holds licensing information for producers, agencies, and companies. It is a subscription service.
- TPA licensing. An NAIC chart shows that most jurisdictions license or register TPAs.
Two cautions. Public data is rich for carriers and thin for agencies, which file no premium statements. And no filing names the policy, claims, or agency management system an organization runs. On the carrier side, core suites come from vendors such as Guidewire and Duck Creek, which both sell policy, billing, and claims applications to P&C insurers. On the agency side, examples include Applied Epic and Vertafore’s AMS360. Score against the list of systems you integrate with.
Segment sets the factor list, and your ideal customer profile sets how heavily each one counts for carriers, MGAs, or agencies. For the steps from profile to weights, see our five-step framework for building an ICP scoring model.
Timing and buying signals in insurance
Segment, premium, and licensing tell you which organizations to pursue. Filings, deals, and seasons tell you when.
| Signal | What it usually suggests | Where to find it |
|---|---|---|
| New product, rate, or form filing, or a new state | A launch or expansion needs tooling | SERFF Filing Access, state department filing searches |
| Core system replacement announced | Connected systems are re-evaluated before go-live | Press releases, job postings, discovery calls |
| Agency acquired, or acquiring others | Systems and vendors get consolidated | Press releases, trade press, deal reports |
| New chief underwriting, claims, or information officer | A new officer tends to reopen vendor choices in the first year | Press releases, professional networks |
| Regulatory change with an effective date | A dated requirement gets its own budget line | NAIC and state department announcements |
| Catastrophe season or a major event | Claims teams are at capacity and not buying | Weather agencies, the carrier’s footprint |
| Policy and reinsurance renewal periods | Underwriting and executive time is committed | Ask on a discovery call |
The NAIC describes SERFF Filing Access as online access to publicly available rate and form filings. Participating states are listed on the SERFF site, and what each state marks public varies. A carrier filing a new commercial product in six states is showing you its next year of work.
Agency consolidation is the other large one. OPTIS Partners counted 520 announced agency deals in the first three quarters of 2025, 7% fewer than a year earlier. For an agency seller, the acquirer is often the real account.
Seasonality matters too. The National Hurricane Center states that the Atlantic hurricane season runs from June 1 to November 30, with a peak on September 10. A property claims team in an exposed state has little time for a pilot in those months. Renewal dates differ by account, so ask on the first call and store the answer as a property. Carrier decisions take many months, so give filings and leadership changes a slow decay. You can look up every signal in this section in our buying signals glossary.
Scoring the buying group
How many people decide depends on the segment. Forrester’s benchmark is that 95% of B2B purchases involve three or more people. A carrier’s group is larger and more formal, while a small agency’s may be one owner.
| Role | What their engagement tells you |
|---|---|
| Line-of-business owner | A book of business has a problem, and you may have a champion |
| Underwriting | The product is being tested against how risk is priced |
| Claims | The claims team is working out how it would run day to day |
| IT | Someone is checking the fit with the policy or claims system |
| Compliance | Regulatory fit is under review |
| Information security | Formal vendor review is under way |
| Procurement or vendor management | The carrier is processing a purchase |
| Agency principal or owner | At an agency, the decision and the budget in one person |
Give coverage its own line in the model: the count of separate functions at the carrier or MGA where at least one contact is engaged. Four underwriters is one function. An underwriter, an IT lead, and a compliance officer make three.
After that, weight the engaging role by segment. At a carrier or large MGA, a first engagement from compliance, information security, or procurement means review time is being spent on you. Weight it far above a repeat visit from the champion you already have. At an agency, weight the principal, and do not penalize the account for having no security team.
Disqualifiers specific to insurance
- The wrong segment, such as a captive agent who cannot choose their own systems.
- Lines outside your product’s scope, such as a life-only carrier for a P&C claims product.
- States or a core system you cannot support.
- Below your minimum premium or producer count.
- A change freeze during a core system replacement. The announcement is a signal. The implementation often blocks anything that touches the core.
- An agency being acquired. The buyer of the agency chooses its systems and vendors.
- Consumers and policyholders looking for a quote or a claim, consultants, and students.
A worked example: scoring a demo carrier
A vendor of claims software for commercial P&C insurers scores a demo regional carrier on fit. The carrier is midway through a core system replacement, and that is the one factor that subtracts.
- Segment and lines of business (commercial P&C carrier) 30% 8.7
- Direct written premium ($410 million) 25% 7.8
- Core system you integrate with 20% 6.4
- States licensed (14 states) 20% 8.3
- Change freeze during core replacement 5% -4.4
The math: 8.7 at 30% is 2.61, 7.8 at 25% is 1.95, 6.4 at 20% is 1.28, and 8.3 at 20% is 1.66, which adds to 7.50. The disqualifying factor subtracts 4.4 at 5%, or 0.22, for a fit score of 7.3. Three of the four fit factors can be filled from public records.
The carrier’s engagement, broken out by function:
| Signal (last 180 days) | Function | Counts toward engagement |
|---|---|---|
| VP of claims sat through a claims workflow demo | Claims | Strongly |
| Compliance manager asked for data handling documentation | Compliance | Very strongly |
| New commercial auto filing in three states, four months ago | Account event | Strongly |
That leaves the carrier with a 7.3 fit, claims and compliance both engaged, and a fresh filing. The freeze is a reason to ask when go-live is planned.
Plays by score band
Start from these cutoffs and adjust them by segment, moving each one to where conversion clearly rises.
| Engagement 6.0 and up | Engagement 3.0 to 5.9 | Engagement under 3.0 | |
|---|---|---|---|
| Fit 7.0 and up | The account owner moves this week. List who speaks for underwriting, claims, IT, and compliance, and have data handling and security documents on hand | Reach out with a reason tied to a filing, a deal, or a leader | Find a contact in each function you have not reached |
| Fit 4.0 to 6.9 | Qualify the core system, lines, and states on a short call | Nurture carriers, MGAs, and agencies separately, and rescore quarterly | Not worth a rep’s time yet |
| Fit under 4.0 | Verify the segment and license data. If it is right, reply and move on | Send content, nothing more | Do nothing |
Two timing rules sit on top of the grid. For property claims buyers, schedule pilots outside the months their catastrophe exposure peaks. For underwriting buyers, avoid their heaviest renewal period. A carrier’s RFP or a direct request is answered in every band.
How to measure and tune the model
Judge the model on one question: do your top-scored carriers, MGAs, and agencies produce pipeline and revenue at a higher rate than the others? Since a carrier deal can take many months, look at early stages as well, such as meetings held, and compare conversion by band within each segment.
| What you see | What it usually means | What to do |
|---|---|---|
| Carriers convert alike in every band | Premium or headcount is carrying the model | Bring in lines written and the policy or claims system, then reweight |
| Carriers you would want score low | A weighted factor has no data | Fill it from state and NAIC records. Missing data contributes 0 |
Per segment model, change a weight or two in a pass and no more. For the common ways a model goes wrong, read our guide to where teams get ICP scoring wrong. Our account scoring worksheet is a blank template you can fill in once per segment.
Where BoostIdeal fits
For HubSpot teams selling to insurance, BoostIdeal is a scoring engine that fits a segment-first, account-first model.
- Each segment’s weights are set by your team. You decide how premium, lines, states, and core system count. AI assists with analyzing your carrier and agency data and drafting a starting point.
- Scores attach to any record: accounts, contacts, leads, and custom objects. A record can also be scored with data from another, such as a carrier on how many buying functions its contacts cover.
- The reasons sit next to the score. Every score is out of 10, and its factors write back to HubSpot properties that your workflows, lists, and agents can use.
- Each segment can have its own model, such as one for carriers and one for agencies, running side by side.
- A new filing date or producer count on a record flows into its score: scores recalculate hourly by default, and in near real time with webhook sync.
When it is not the right fit. Look elsewhere if your carrier and agency records are not in HubSpot and you would not want to bring them in as CSV uploads, or if you want a vendor to hand over a number you never question.
Frequently asked questions
What is lead scoring for insurance?
For a company that sells to the insurance industry, it is how you rank carriers, MGAs, agencies, and TPAs and their contacts by fit and buying readiness. Fit comes from segment, lines of business, size, states licensed, and core system. Readiness comes from events and buying group coverage.
Is this the same as scoring insurance leads or underwriting risk?
No. Consumer lead scoring ranks people who might buy a policy, and underwriting scores rate the risk of insuring them. This is B2B scoring of insurance organizations as potential customers for a vendor, with no policyholder data.
How is selling to insurance carriers different from selling to insurance agencies?
Carriers are few, large, and formal, with underwriting, claims, IT, compliance, and security all involved. Agencies are many and small, and an owner often decides alone. Size a carrier by premium and an agency by producers, and run a separate model for each.
Where can I find public data on insurance companies for account scoring?
State insurance departments publish company lookups with license status and authorized lines. The NAIC publishes market share reports by line and sells annual and quarterly financial statements through InsData. SERFF Filing Access shows public rate and form filings in participating states.
What are the strongest buying signals in insurtech lead scoring?
New product or state filings, an announced core system replacement, an agency acquisition, a new underwriting, claims, or technology leader, and an RFP. Once a deal is open, compliance or information security showing up for the first time is among the clearest signs.
Does premium volume work as a fit score on its own?
It is the best single size measure for a carrier or MGA, but two carriers with the same premium can write different lines in different states on different systems. Agencies report no premium, so use producer count there.
The takeaway
Insurance is several industries sharing one label, and a single scoring model cannot serve them all. Decide the segment first, size carriers by premium and agencies by producers, and fill lines and states from the records regulators publish. Add the core system, and let filings, acquisitions, and seasons set the timing. Leave each factor and weight in plain view, so a rep can tell why a carrier or agency sits at the top.