Compare Options Across Websites

Introduction

“Let me open a few more tabs” is how an hour disappears. Each site formats pricing differently — monthly vs annual, per seat vs usage, footnotes on page four — and by the time you have a spreadsheet, you are not sure you captured the same fields for each option.

Dume for Chrome can read the pages you already have open, normalize the data into one schema, and give you a side-by-side table plus a short recommendation. You make the call; Dume removes the copy-paste tax.

How It Works

Step 1 — Open candidate pages in separate tabs

One product or plan per tab works best. If a site uses heavy modals or login walls, expand the sections you care about (pricing, FAQ, spec sheet) before asking Dume.

Step 2 — Tell Dume what decision you are making

A clear goal changes the comparison: “cheapest for 10 seats” vs “best support for enterprise” vs “fastest to integrate with Salesforce”.

Step 3 — Request extraction + normalization

Ask Dume to pull the same column set from every tab — e.g. price, limits, contract length, cancellation, data residency.

Step 4 — Review footnotes and caveats

Dume will flag when a field is missing, unclear, or only in fine print. That is your signal to verify on the vendor’s site or in sales email.

What Dume Compares

• Pricing models — list, discounted, usage tiers, annual vs monthly
• Limits — seats, API calls, storage, rate limits, support response times
• Features — must-haves vs nice-to-haves; integration lists
• Terms — trial length, minimum commitment, refund policy
• Qualitative signals — reviews, ratings, or positioning copy (with source notes)

Example Prompt

I’m choosing a helpdesk tool for a 25-person team.

For each tab I have open (pricing + feature pages):
1) Extract: name, list price for ~25 agents, annual vs monthly, trial terms
2) Extract: SLA, channels supported (email/chat/phone), hours of support
3) Extract: native integrations I care about: Salesforce, Slack, Jira
4) Note any hidden fees (per seat add-ons, AI usage, onboarding)

Output:
- One markdown table with identical columns for every option
- “Unknown / not published” where data isn’t on the page
- Top 2 recommendations with tradeoffs and who each is best for
- List of 5 questions I should ask sales before signing

Example Comparison Table (Shape)

Customization

• "Weight price 40%, integrations 40%, support 20%"
• "Assume we are in the EU — call out data residency on each site"
• "I’m non-technical — explain jargon in one line per term"
• "Output as CSV I can paste into a deck"

Tips for Best Results

• Keep all candidates at the same stage — e.g. all on public pricing, or all post-demo quotes
• If a site has regional pricing, say which country you are buying from
• Screenshots are optional; visible page content is usually enough
• When two products look tied, ask for a decision matrix with weighted scores

Limitations

Web pages change, and some vendors hide terms behind sales calls. Treat Dume’s table as a working draft — confirm numbers, contract language, and compliance requirements before purchase.