How we rank travel eSIMs

Plans are imported directly from each provider — through their own product feeds and affiliate catalogues — and refreshed on a daily schedule. Nothing on a comparison table is typed in by hand, so a price change at the provider reaches the site within a day rather than whenever someone remembers to update a post.

A plan that stops appearing in a provider’s feed is marked stale and drops out of the comparisons rather than lingering at a price nobody can buy.

Providers price in different currencies, bundle sizes and validity windows, which makes plans hard to compare on purpose. Every plan is converted to one canonical currency at import, then reduced to a single number:

cost per GB = total price ÷ gigabytes included

That number is what the tables sort on. It is derived at render time from the current price, never stored, so it cannot drift out of step with the price shown beside it.

Where a guide quotes a “from” price per GB for a provider, it is the lowest cost per GB across that provider’s live plans for that destination — not a headline rate from a marketing page.

Plans sold as unlimited have no meaningful cost per GB, so they are compared separately, on total price and on the daily fair-use allowance behind the word.

Almost every unlimited travel eSIM slows down after a daily threshold. We record that threshold only where the provider publishes it themselves, and we link to the page it was read from so you can check it. Third-party blog claims are not used.

Where a provider does not publish a figure, the tables say Not published rather than filling the gap with an estimate. That is a real finding about the provider, and it is more useful than a number nobody can verify.

Which local carrier an eSIM actually rides decides how well it works away from the big cities, so guides name it where it is known. Each row is marked Confirmed only when it has been verified against a live eSIM or a provider statement; otherwise it stays Confirming, with the reason given. Providers who decline to name their host network are shown as exactly that.

The score on each pick is an editorial rating, not a test result. It weighs the things this method can actually establish: cost per GB against the rest of the market at the sizes people buy, which host network the plan rides and how much of the country that covers, how clearly the provider publishes its fair-use and top-up terms, and whether the plan can be extended mid-trip.

It is not a measure of how the eSIM performed in someone’s hand. Where you see a rating on this site, it comes from the criteria above applied to live data and published terms.

Price and published policy are measured well here. Two things are not, and you should weigh guides accordingly:

• Real-world speed varies by handset, cell site and time of day. Where a guide reports measured speeds, they are individual observations with a date and a location attached, not a benchmark.
• These guides are built from live pricing and providers’ own published terms, not from hands-on trials of each plan in each destination. Where a page carries measured speeds or field observations, they are labelled as such with a date and a location; if a page carries none, there are none to report.

Some outbound links earn commission. Rankings are produced from the price data before any commercial consideration, no provider has ever paid for a placement, and commission rates play no part in the order. Providers we earn nothing from are ranked alongside those we do, on the same numbers.