Methodology
This is the most important page on eSIMBench. Every grade we publish is only as trustworthy as the method behind it, so we put that method here in full — the dimensions, the weights, the sources, the timing, the corrections, and the firewall between our scoring and our revenue. If you only read one page before trusting a number, read this one.
Why a benchmark, not another comparison site
Almost every eSIM comparison site ranks providers on price, and a few rank on speed. Price and speed matter, but they are two facets of a product that has at least eight. An eSIM with a brilliant price can have an app that fails on arrival, a no-refund policy, a hidden data cap behind the word "unlimited", or hotspot tethering that is silently throttled. A traveler who chose on price alone discovers the rest at the worst possible moment — abroad, with no data, and no easy way to get help.
eSIMBench exists to score the whole product. We treat travel eSIMs the way Consumer Reports treats appliances or a benchmark suite treats hardware: a fixed, public rubric applied identically to every contender, refreshed on a schedule, and published with its sources. The goal is a number a journalist can cite and a traveler can act on, because the working is shown.
The eight dimensions and their weights
We score each provider from 0 to 100 on eight dimensions, then combine them into a single Bench Score using the weights below. The weights reflect how much each dimension affects a real traveler's experience, not how easy it is to measure. They sum to 100%.
| Dimension | Weight |
|---|---|
| Speed & Coverage | 25% |
| Pricing Value | 20% |
| Plan Flexibility | 10% |
| App Quality | 10% |
| Customer Support | 10% |
| Hotspot & Tethering | 5% |
| 5G Access | 10% |
| Fair Use & Refunds | 10% |
| Total | 100% |
Speed & Coverage carries the most weight because connectivity that does not work makes every other strength irrelevant. Pricing Value is next, because value — not raw cheapness — is what most travelers are really optimizing. The remaining dimensions each carry less individual weight, but together they decide most close contests. Each dimension has its own deep-dive page explaining what it measures and why it is hard to measure honestly.
How each dimension is scored
Every dimension is normalized to a 0–100 scale so they can be combined fairly. Where a dimension is naturally comparative — pricing value, for example — we score against the market median rather than an absolute, so a grade reflects how a provider stacks up against its actual peers in the same quarter. Where a dimension is a policy rather than a measurement — hotspot rules, refund terms — we grade consumer-friendliness on a defined rubric, reading the terms of service rather than the marketing page. The per-dimension method, including the exact source, is documented on each provider's own methodology page.
Why these eight dimensions and not others
The choice of dimensions is itself an editorial decision, so it deserves an explanation. We started from a simple test: what actually goes wrong for travelers, and what actually makes one eSIM better than another in practice? Every dimension on the list earned its place by being something a real traveler notices and something we can measure with a defensible source. Speed, price, and 5G access are the obvious performance axes. App quality and customer support are the experience axes — the parts that decide whether a good plan is usable when you are tired, jet-lagged, and standing in an airport. Plan flexibility, hotspot policy, and fair-use and refund terms are the fine-print axes, the places where a provider's real attitude toward its customers shows up.
We deliberately left some things off. We do not score brand popularity, marketing reach, or social-media following, because none of those tell you whether the product works. We do not score the number of countries a provider lists, because a long list of barely-supported countries is worth less than solid coverage where people actually travel — that quality is already captured inside Speed & Coverage. The list is intentionally short enough to stay meaningful: eight dimensions a reader can hold in their head, not forty that average into mush. If we add or remove a dimension in future, we will say so in the changelog and re-baseline every provider at once, never selectively.
Where the data comes from
A benchmark is only as honest as its inputs, so we are specific about ours. Speed and coverage are sourced from the measurement pipeline behind our sister site, howfastesim.com, which aggregates network-level speed data by country. Pricing comes from provider plan catalogs and published rate cards, normalized to a cost-per-gigabyte basis. App quality draws on the most recent public reviews from the iOS App Store and Google Play, analyzed for sentiment and recurring themes. Policy dimensions — fair use, refunds, hotspot — are extracted from each provider's terms of service and cross-checked against community reports. Customer support is tested directly: we send standardized queries through live chat and email each quarter and time a genuine response. The full list lives on our data sources page, and every score on the site carries a visible source and date.
Live scores versus quarterly snapshots
Some of our data updates continuously — plan prices change, app reviews accumulate, networks shift. Other data is gathered on a cadence, such as the quarterly support-response tests. To keep the site current without making it impossible to cite, we run two layers. Live scores update as fresh data arrives, so the site always reflects the latest picture. Then, once a quarter, we freeze every score into a dated snapshot and publish it as the State of eSIM report. A snapshot never changes after publication, which is what makes it citable; the live scores keep moving. When you see a grade on a provider page, it is the live score, labeled with its date. When you cite a report, you are citing a fixed moment.
Editorial independence
eSIMBench earns affiliate commissions when readers buy through some of our outbound links. That is how the site is funded, and we disclose it plainly. It also creates an obvious conflict of interest, so we have built specific safeguards rather than asking you to trust our good intentions.
First, scoring is mechanical. Grades are computed from the data and the published weights, not assigned by a person who knows which providers pay best. No one can nudge a score up because a provider has a generous affiliate program, because the score is not a matter of opinion in the first place. Second, the rubric and weights are public and fixed in advance of the data, so we cannot quietly re-weight the dimensions to favor a partner after seeing the results. Third, affiliate relationships have no input into the rubric, the weights, or any individual score, and a provider's commission rate is never a factor in its grade. Fourth, on scorecard pages we keep affiliate calls-to-action out of the way of the data — you see the grade and its evidence before you see any link to buy. We would rather lose a click than sell a grade.
Affiliate disclosure
Some links on eSIMBench are affiliate links: if you click one and make a purchase, we may earn a commission at no extra cost to you, and in some cases you may receive a discount. Affiliate links are marked as sponsored outbound links. The presence or absence of an affiliate relationship with a provider does not affect that provider's score, its rank, or whether it appears on the site. Every one of the providers we track is included on the same terms, regardless of whether we can earn anything from it. Our full disclosure also appears on the about page.
Corrections policy
We will get things wrong — a price will be misread, a policy will change the day after we read it, a network will be remapped. When that happens we fix the underlying data, let the score recompute, and note material corrections in the score changelog so the record is transparent rather than quietly edited. If you believe a score is wrong, tell us and point to the evidence; if you are right, we change it and say so. Providers do not get to edit their grades, but anyone can challenge the facts a grade is built on.
What we deliberately do not do
We do not publish estimated, provisional, or placeholder scores. Until a dimension has real, sourced data, it reads "Pending" rather than showing a number that might be wrong. We do not accept payment for a higher grade, for a better rank, or for inclusion. We do not let providers preview or approve their scorecards before publication. And we do not mark up structured data with ratings that are not actually shown to readers. These are not difficult rules to follow, but they are the difference between a benchmark and an advertisement.
How to read a scorecard
Each provider scorecard leads with the overall Bench Score and letter grade, then breaks the grade down into the eight dimensions so you can see exactly where a provider is strong and where it is weak. Two providers can share an identical overall grade for opposite reasons — one fast but expensive, the other cheap but slow — and the breakdown is where that difference becomes visible. Read the dimensions that matter most for your trip rather than fixating on the headline letter. A digital nomad and a weekend tourist should rationally pick different providers from the same scorecard, which is exactly why we also publish use-case rankings that re-weight the dimensions for specific traveler profiles. Wherever a dimension has not yet been measured, the scorecard shows "Pending" rather than a guess, and the overall grade reflects only what has actually been scored.
Questions about the method, or think we have a number wrong? See about & contact. The first full benchmark snapshot publishes in the Q3 2026 State of eSIM report.