Facio
All resources

Buyer evaluation toolkit

MGA Software Evaluation: A Practical RFP Scorecard

Ask every vendor to run the same insurance journey and retain the evidence. This suggested scorecard turns an attractive demo into a comparable operating decision. Adjust the weights to your business before the first evaluation.

Updated 18 November 2026 · 7 minute read

Set the decision rules before scoring vendors

Use a 0–5 evidence scale: 0 means not demonstrated; 1 means a roadmap or slide claim; 2 means a scripted demonstration; 3 means your scenario works in a test environment; 4 means an operator can complete it and reopen the saved result; 5 means the full journey is repeatable with documented exception handling and agreed acceptance evidence.

Calculate each weighted result as weight × score / 5. The weights below total 100. A high total does not override a mandatory requirement: record unmet security, contractual, market or financial controls separately and resolve them before selection.

  • Policy lifecycle and transaction integrity: 25 points
  • Product configuration and underwriting authority: 20 points
  • Bordereaux, finance and reconciliation: 20 points
  • Migration, integrations and data ownership: 15 points
  • Security, permissions and auditability: 10 points
  • Implementation, support and three-year cost: 10 points

Run one realistic policy journey

Provide the same synthetic risk, product wording, limits, rates and authority boundaries to each vendor. Quote and bind the risk, issue the documents, process a mid-term change, then cancel it. Include a referral that an ordinary operator cannot approve.

Record the policy versions, approval decision, premium movements and document outputs. Reopen the saved policy as another authorized user. A corrected screen without a persistent record is not sufficient evidence.

  • Can the operator trace every premium and commission movement?
  • Does a historical document preserve the terms that applied when it was issued?
  • Does an unauthorized bind fail with a clear next step?
  • Can a correction be explained without overwriting history?

Change a product during the evaluation

Add a question, change a rate factor and introduce a referral threshold in a draft product version. Review and publish it through the normal operational controls. Create a new quote, then amend a policy issued under the previous version.

Separate configuration an authorized business user can manage from custom development, one-time import work and future roadmap items. Ask for the owner, lead time and cost of each category.

  • Draft, review, publication and rollback evidence
  • Existing policy behavior after a new version is published
  • Questions, coverages, clauses and pricing linked to the same product definition

Reconcile the outputs, not just the dashboard

Generate the carrier's required risk and premium bordereaux from the test journey. Reconcile totals to the policy transactions, taxes, commissions and settlement records. Include one rejected row and one correction in a subsequent reporting period.

Require a trace from a reported value back to the source transaction and approval. Record any spreadsheet steps still required, who owns them and how errors are detected.

  • A traceable sample bordereau with an exception and correction
  • Gross, net, tax and commission reconciliation
  • A repeatable period-close and resubmission procedure

Test migration and exit before committing

Use a small, representative synthetic migration batch containing an active policy, a historical endorsement, an unpaid balance and a document. Validate counts, financial totals, relationship links and effective dates after import.

Request an export of those same records in documented formats. Evaluate data ownership, access to documents, API coverage, rate limits, implementation responsibilities and the cost of leaving the platform. Price migration and exit effort alongside subscriptions.

  • Import reconciliation and exception report
  • Documented integration and authentication boundaries
  • An export with policy history, documents and financial movements
  • A three-year cost model including implementation, integrations, usage and support

Keep a decision log with evidence

For each requirement, record the owner, weight, score, evidence link, limitation and next action. Have operations, underwriting, finance and technology score the portions they will own. Resolve conflicting scores by repeating the scenario with the relevant operator.

Facio is one vendor you can evaluate with this framework. The scorecard describes suggested buyer tests; it is not a claim that any vendor, including Facio, has passed every requirement for your product or market.

Evaluation checklist

  1. 1Freeze your scope and mandatory requirements
  2. 2Agree weights and use the same synthetic scenario for every vendor
  3. 3Retain saved records, documents and reconciliation outputs
  4. 4Separate current capability from configuration work and roadmap promises
  5. 5Assign unresolved gaps an owner, cost and acceptance date
  6. 6Choose against operational evidence and total cost

Frequently asked questions

What should an MGA software RFP include?

Include your programs, authority model, policy lifecycle, product change process, carrier reporting formats, integrations, migration scope, security requirements and total cost assumptions. Attach an evaluation scenario and the evidence required to accept it.

How should we score a policy administration system demo?

Use a consistent evidence scale and business-specific weights. Give more credit to repeatable operator journeys with saved results than to slides or scripted demonstrations. Keep mandatory requirements outside the weighted total so they cannot be averaged away.

Are these weights a market benchmark?

No. They are a suggested starting point from Facio. Adjust them with your operations, underwriting, finance and technology teams before evaluating vendors.

Continue your evaluation