How the Analytics1 Score works

The Analytics1 Score is a 0 to 100 score for a tool within one category, built only from what the vendor publishes about its own product: features, prices, plans, privacy documents, integrations and support. Every fact is recorded with the page it came from and the date we read it, and the points follow fixed rules, so anyone with the same evidence gets the same score. Version 1.0; 88 of 98 category entries are scored.

Partner status has zero effect

Some tools we rank pay us a commission through affiliate links, and some vendors buy a labelled featured placement. Neither is an input to the score. The scoring code does not read our partner list or our placement records, and a build check fails if it ever does. Affiliate partners and sponsors are scored by the same rules as tools we have no relationship with, and a sponsored vendor that also appears in a ranking appears at its own score. Paid placements sit in a separate band marked “Sponsored · not part of the ranking”.

The seven criteria and their weights

Each criterion earns 0 to 5 points; points are multiplied by the criterion's weight and summed to 100.

Why these weights. A tool that cannot do the job is the wrong tool at any price, so core jobs carry the most weight. Pricing and privacy come next because they are the two things small businesses and agencies most often find out too late: what it will cost once a trial ends, and whether it can be used lawfully with EU visitors. Free plans, set-up effort, integrations and support matter, but less.

Core jobs by category

Each category has five core jobs, the work a buyer in that category most needs done. A tool earns a core-job point only where its own pages say it does the job; a point is not given for something we inferred.

Evidence rules

• Only the vendor’s own website, documentation, help centre, trust or legal pages and pricing page count. No review sites, no forums, no memory.
• Every fact is stored with its URL and the date it was read. Each ranked entry has a “Score breakdown” listing the points and the pages behind them.
• A fact we could not find scores zero. That is deliberate: a buyer can only rely on what a vendor publishes. It also means a capable tool with thin public documentation scores lower than it might deserve.
• A tool is scored only with evidence for its core jobs, a pricing fact and at least three of the other five criteria. Otherwise it is listed as “Not yet scored” at the end of the ranking, alphabetically.
• Ties are broken by core-job points, then alphabetically.
• Scores compare tools within a category. A 90 for a call tracker and an 80 for a survey tool say nothing about each other.

Limitations

• We have not tested these products in live accounts. The score measures what a vendor documents, not how well the product works in practice, how good its data is, or how its support performs.
• Feature claims are taken at the vendor’s word when they appear on its own pages.
• Pages change. Evidence was read on 19 October 2026; a price or plan may have moved since. Each entry shows its date.
• Prices are banded on the amount printed, in the currency printed, without exchange-rate conversion, and on the vendor’s cheapest paid plan. That plan may not suit your volume.
• Integration counts are the vendor’s own stated figure or what its directory lists; vendors count differently.
• Some pages could not be read automatically (sites that block automated requests or load everything with JavaScript). Where that left gaps, the tool scores lower or is not yet scored.
• Five core jobs cannot capture every buyer’s needs. Read the “Best for” line and the guide on each page, not only the number.

Reproduce it

The complete evidence file, the rules and every computed score are published as JSON at /data/analytics1-score.json. The scoring code applies the rules above to that file and nothing else.

Corrections and updates

If a fact about your product is wrong or out of date, send the page that shows the correct fact through the contact form. We re-read the page and re-score; we do not accept a correction without a public source. Buying a placement does not speed this up or change the outcome. We plan a full re-read of the evidence every six months.

Related

Top analytics tools 2026 lists every category winner. Our editorial policy covers how we write, and the disclosures page covers how we make money, including featured placements.