Skip to main content
Available to claimEvidence: SupportedConsumer insurance
A balance scale weighing a price-tagged insurance quote against a car key, with the beam level.Opportunity score: 51 out of 100 — Worth exploring

Rate Check

Tells you if your car insurance quote is fair, before you sign.

Published · Last assessed · Methodology v5.1 · How ideas are researched and assessed

Problem in brief

A driver with a car insurance quote in hand has no straightforward way to find out whether the number is reasonable for someone with their vehicle, coverage and ZIP code. Comparison sites exist but reviewers complain they must hand over personal data and then click through to carrier sites to see an actual quote, with one consumer reporting roughly 50 unwanted insurer calls afterwards. The same uncertainty hits after a claim, where people currently hire independent appraisers or public adjusters to dispute a total-loss valuation. Most drivers cope by ringing several agents or posting their numbers online and hoping someone tells them they are being overcharged.

Who has this problem
Individual personal auto insurance shoppers and policyholders in the US who are comparing quotes or facing a claim decision
How they cope today
Calling multiple agents to compare answers and posting on Reddit to crowdsource guidance and rates
What it costs them
Decisions on claims handling and major purchases stall amid uncertainty, with no documented or trustworthy answer
Opportunity explored
Rate Check, described below
A driver at a kitchen table shrugs while three hands offer differently sized stacks of blank insurance paperwork.

The problem in detail

Drivers can't tell whether a quote is reasonable or what to do after a claim, because every agent gives a different answer and there's no public price reference for someone with their car, ZIP code and record. They end up calling three agents and posting their numbers on Reddit hoping strangers will tell them if they're being overcharged.

Proposed product

A web tool where a driver uploads their declarations page or quote PDF and gets a plain-English verdict: how their premium compares to other drivers with similar age band, vehicle, coverage limits and ZIP code, which coverages are driving the price, and which line items are unusual. Submitted policies (stripped of personal details) build the benchmark pool, so every user both gets and gives a data point. A second layer answers the recurring 'what should I do' questions — whether to file a claim, accept a total-loss valuation, add a teen driver — using the user's actual coverage numbers and their state's rules, with the answer written down so they can show it to the adjuster.

Why now

Existing comparison sites now function mainly as lead funnels — The Zebra sits at 2.9/5 on one aggregator, with reviewers saying they must hand over data and then click through to carrier sites to see any real number, and a BBB complaint describes roughly 50 unwanted insurer calls after using it. That resentment is the opening for a tool that gives a verdict without entering the lead funnel. Otherwise this is a long-standing problem nobody has solved well, and one state regulator (Delaware) actually moved to repeal its own public rate comparison tool.

Smallest useful version

A single-state, single-line (personal auto) version: paste or upload a quote, answer five questions, and receive a benchmark percentile plus three flagged line items. Seed the pool by parsing publicly filed rate tables and a few hundred policies collected from an insurance subreddit audience, then charge nothing until the benchmark feels credible.

Potential ways to charge

  • Free benchmark — £0: percentile, three flagged line items, in exchange for an anonymised copy of the policy to seed the pool (free tier earns the data and the paid upsell)
  • Full policy report — £12 one-off: line-by-line breakdown, which coverages drive the price, state-specific notes, downloadable PDF (typical for comparable one-off consumer report tools; no price in the evidence)
  • Claim decision write-up — £35 one-off: a written, citable answer on a specific decision (file or not, total-loss valuation, adding a teen driver) using the user's actual coverage numbers, formatted to show an adjuster (anchored well below the paid independent appraisers and public adjusters the evidence shows consumers already hire for total-loss disputes)
  • Annual renewal watch — £20 per year: re-benchmark at each renewal and alert when the premium moves out of band (typical for comparable subscription consumer tools)

Routes to early customers

  • Insurance-focused subreddits where drivers already post their numbers asking strangers whether they are being overcharged — the behaviour the idea is built on, and the stated seeding route for the policy pool
  • Total-loss and claim-dispute content: people searching for how to challenge a CCC One valuation or invoke the appraisal clause are already paying appraisers and adjusters for this answer
  • Review and complaint threads about quote aggregators, where users specifically object to the calls that follow — position as the no-lead-form alternative
  • State insurance department consumer pages and rate comparison tables, which attract exactly the person looking for a price reference

Main risks

  • Cold start: the benchmark may never be dense enough at ZIP-and-vehicle level from user submissions, and SERFF-derived filings may not reproduce a real quote closely enough to give a trustworthy percentile
  • A percentile may be interesting but not actionable — if the user's next step is still to call three agents, they will not pay
  • The 'no public price reference' premise is partly wrong: several state regulators publish comparison tables, and Delaware moved to repeal its own on the reasoning that carriers' calculators now fit an exact profile
  • Commercial products already index approved rate filings, so the data moat may be shallow and an incumbent could add a consumer-facing verdict
  • Giving written answers on whether to file a claim or accept a valuation edges toward regulated advice; licensing and liability exposure in each state is unresolved

Questions to test first

  • Build the SERFF-based estimator for one state and one carrier set, then test it against twenty real declarations pages collected from a subreddit — measure how close the predicted premium is to the actual one before building any product
  • Post a free manual review offer in an insurance subreddit: ask for dec pages, return a written verdict by hand, and count how many people submit and how many say it changed what they did
  • Offer the claim-side write-up (total-loss valuation challenge) at a fixed price to people in dispute threads, since the evidence shows this job is already bought from humans — test price acceptance there first
  • Compare the output against the free state DOI comparison tables for the same profile, to check whether the tool says anything the free public source does not
  • Get a one-hour read from an insurance regulatory lawyer on whether written claim-decision guidance requires licensing in the target state

Evidence and sources

The sources this assessment is based on. Evidence level: Supported, counted from these sources as described in how ideas are assessed. Findings are what a source shows; anything the research only inferred is marked as an inference.

Research findings

  • SupportsThird-party review aggregator page for thezebra.com (263 customer reviews, rated 2.9/5) · 10 July 2025

    Shows: aggregated user reviews of incumbent

    Reviews on this aggregator rate The Zebra 2.9/5 and include complaints that users must hand over data and then click through to carrier sites (GEICO, Progressive) to see an actual quote, with one reviewer calling it "Pretty worthless". Observed evidence that existing comparison sites do not deliver a straight answer on whether a given premium is reasonable, which is exactly the gap Rate Check targets.

  • SupportsBBB complaints page for The Zebra (auto insurance comparison site)

    Shows: support-complaint / product review of incumbent comparison tools

    BBB complaint text on The Zebra's profile includes a consumer saying they received roughly 50 unwanted calls from insurers after using the site and accusing it of selling phone data. This is an observed, non-community-sourced complaint supporting the candidate's premise that today's path to a price reference (quote-aggregator forms) carries a lead-generation cost users resent — the opening for an upload-a-dec-page tool that gives a benchmark without entering the lead funnel.

  • SupportsCommercial total-loss dispute/appraisal service pages (certified auto appraisers, Total Loss Toolkit, snapclaim, public-adjuster referrals)

    Shows: paid-service demand for the claim-side job (total-loss disputes)

    A whole cottage industry of paid independent appraisers, public adjusters and 'total loss dispute' guides exists, instructing consumers to demand the insurer's CCC One valuation report, refuse the first offer, and invoke the policy appraisal clause. Observed proof that people pay humans to do, case by case, the 'what should I do after a claim, written down for the adjuster' layer the candidate proposes — demand is real and currently served manually and expensively.

  • Context · inferenceSERFF Filing Access portals (AZ DIFI, Oregon DFR) plus commercial rate-filing trackers (boringrate, insurancefiling.com, rateauthority)

    Shows: data-availability / competitive landscape for benchmarks

    Carrier rate, rule and form filings are public via SERFF in most states, and several commercial products already index and roll up approved auto rate filings by state and carrier. Inference: the underlying rating factors needed to judge whether a quote is reasonable are partly obtainable without a user-contributed pool, which both lowers the cold-start barrier for buildability and means Rate Check's crowd-sourced benchmark is not the only possible source of truth.

  • Weakens the caseDelaware Register of Regulations — proposed repeal of Regulation 610, Automobile Premium Consumer Comparison · 1 November 2018

    Shows: regulatory record on public price references

    Delaware's insurance regulator ran an online auto rate comparison tool from 2006 letting consumers compare rates by driver profile and ZIP, then moved to repeal it on the reasoning that insurers' own calculators now fit an exact consumer profile. This weakens the 'no public price reference exists' premise (several state DOIs publish comparison tables) and signals that a generic rate-comparison layer alone has been tried and abandoned by at least one state.

Where the problem was reported

Public posts in which people described this problem, grouped into one problem before research began.

Looked for, not found

  • Freelance marketplace listings surfaced for insurance advisory and coverage analysis, but the visible work is life/health policy second opinions and commercial/claims coverage analysis for businesses — no real result showed individual US drivers paying freelancers to benchmark a personal auto declarations page. Whether paid freelance demand exists specifically for personal-auto premium review is unknown from the results obtained; no search-volume data for such a service was retrieved either.

This is a researched opportunity, not a guarantee of commercial success. The evidence level and sources show how much is known; the risks and questions to test show what is not.

Chosen for a shared category, customer or problem.

  • Individuals and households

    Plate Back

    When a US driver switches auto insurers, the new carrier is meant to report the policy electronically to the state DMV.

  • Individuals and households

    Claim Clock

    When a car insurance repair or a boiler or home-emergency claim stalls, the insurer or its approved repairer controls the facts that decide it: whether the damage counts as wear, whether a part is really on back order, when an engineer will actually attend.

  • Individuals and households

    In Writing

    After a small crash, a driver has days to decide whether to file a claim, what to tell the other party's insurer, and whether to accept a waived deductible or pay out of pocket.

  • Individuals and households

    Job Proof

    Homeowners who have just paid an HVAC company for an installation, major repair or service visit often have no way to judge whether the work was actually done properly.

  • Individuals and households

    Mail Vault

    People who have used one email provider for twenty years or more have no easy way to search, archive or migrate their full mail history.

More individuals and households ideas →

The method behind this idea, written up as guides you can apply to your own.

  • How to find product ideas worth building

    Find product ideas by starting from problems people describe in their own work, not from brainstorms. A practical method with real examples, from the research behind Daily Product Idea.

  • How to evaluate a product idea before you build it

    Evaluate a product idea by scoring the evidence rather than your enthusiasm: six dimensions, a written reason for each score, the assumptions it rests on and the cheapest test to run first.

  • How to validate a product idea before building it

    Validate a product idea with evidence of what customers do, not what they say: find the assumption that would kill it, run the cheapest test that can fail, and set the pass mark first.

  • Problem Signals Report, autumn 2026

    Six weeks of community posts, counted: where real problems surface, how people work around them, and why most product ideas are easy to build but hard to differentiate.

All 12 guides →

£49One buyer only

Claim