Compare: insurance quote comparison
Compare is medical insurance quote comparison software for UAE brokers. It reads insurer and TPA quote PDFs, learns each layout once with AI and then extracts that layout with no model cost, and builds side-by-side comparisons in Excel, PDF and a shareable proposal, with a client CRM, RFQs and renewals around it.
Built for teams like yours
- Account managers and placement teamsA stack of insurer PDFs turned into one comparison the client can read, without retyping premiums and benefit limits into a spreadsheet.
- Brokerage ownersBenchmarks against the brokerage's own book, win and loss by insurer, and a view of what each client bought.
- Renewals teamsA renewal compared against last year's plan on premium, cost per member and every benefit that moved, with a reminder before it falls due.
What Compare does
Learns each layout once
Each new quote layout is fingerprinted and read by a model once to learn an extraction recipe. Every later quote in that layout is read by the recipe, with no model call. Re-uploaded files are recognised before any spend.
Opened for a client, every time
Every upload starts with a named client. The record carries through the batch, the review queue and the finished comparison, so no deliverable is ever unattributed.
Comparisons clients can read
Excel, HTML, PDF and a dashboard, with frozen headers, the best value per benefit highlighted and a weighted ranking with a Recommended badge. Benefits no insurer quoted are left out.
80 fields, 11 sections
Premiums, network, inpatient and outpatient, maternity, dental and optical, wellness, waiting periods, age bands, deductibles, direct-billing hospitals and claims service. The field list is edited as configuration.
Renewals and revisions
Renewals are compared with the plan the client bought. A revision diff shows who joined the panel, who dropped and what each insurer changed between drafts.
CRM, proposals and analytics
Client directory and documents, RFQ intake linked to the comparison it produces, shareable proposals with maker-checker approval, and premium benchmarks and win and loss from your own book.
Where the AI is, and where it is not
What the AI does
- Learns an extraction recipe for each new quote layout, once.
- In consensus mode, two independent model reads must agree before a value is used.
- Flags layouts whose saved recipe has started to disagree with live reads.
What stays with rules or your team
- Once a layout is learned, extraction runs on the saved recipe with no model call.
- Disagreements and doubtful values go to a person in the review queue.
- A per-batch spend ceiling stops a run of new layouts from running up cost.
- Rankings follow the weights you set, and the broker decides what to recommend.
Built for the UAE, and for your stack
UAE compliance
- Premiums and costs in AED
- Benchmarks scoped by emirate and policy type
- Maker-checker approval before a proposal is shared
- No internal identifiers or extraction cost on client-facing pages
Connects to
- Anthropic API (optional, for layout learning and consensus)
- Read-only REST API with bearer tokens
- Excel, HTML and PDF exports
In production
A product example, not a client story: how Compare handles quotes for UAE medical insurance brokers. It reads insurer quote PDFs, learns each layout once with AI, then extracts later quotes in that layout at no model cost, and builds the side-by-side comparison and a proposal.
What does Compare do?
Compare turns insurer and TPA quote PDFs into a side-by-side medical insurance comparison. It reads each quote, extracts premiums and benefits into 80 fields, and renders the comparison in Excel, HTML, PDF or a shareable proposal. Around it sit a client CRM, RFQ intake, renewals and analytics for UAE brokers.
How can extraction cost nothing after the first quote?
The first time Compare sees a new quote layout, a model reads it and learns a recipe for that layout. Every later quote in the same layout is read by the recipe, without calling a model. A re-uploaded file is recognised by its content before extraction, so it costs nothing either.
What happens when the AI is not sure?
Doubtful values go to a person in the review queue before they reach a comparison. In optional consensus mode, two independent model reads must agree before a value is used, and any disagreement goes to review. A spend ceiling per batch caps what a run of new layouts can cost.
Can it handle scanned quotes?
Yes. Compare reads the text and tables in a PDF and falls back to OCR for scanned quotes. Each layout gets its own fingerprint, and a nightly check flags layouts whose saved recipe has started to disagree with live reads, so a quiet change in an insurer's template gets noticed.
How does Compare help with renewals?
A renewal is compared against the plan the client bought, on premium, cost per member and every benefit that moved, not as a fresh comparison. A daily sweep surfaces renewals that are coming due, and a revision diff shows what each insurer changed between drafts.
Will clients see our internal data?
No. Client-facing proposals and the main dashboard never show layout fingerprints, job identifiers, internal status codes or extraction cost. Proposals go through maker-checker approval before they are shared. Premium benchmarks and win and loss figures are built from the brokerage's own book and stay inside the brokerage, and per-document extraction cost is visible only to an internal list of people.
Related
Solutions it runs
Industries
Other products
Start with one job
Opens an email to info@elitestack.ae with your sentence in it. Or WhatsApp us on +971 56 121 1011.