How GradeBridge converts grades

Every number on this site comes from the method described below. It is written so you can check it rather than trust it — the worked example is produced by the live converter, and the counts are read from the same data the tool runs on.

Method last reviewed 2026-07-27 · covers all 53 grading systems

What does the converter actually do?

Three steps, in this order:

1. Read your grade against your own country's grade bands — the ranges its universities use to decide what counts as a First, a Distinction, a Gut — and work out where inside that band it sits.
2. Express that position as a US GPA (Grade Point Average — the American 0–4.0 scale, where 4.0 is the top). This is the shared currency, called the pivot throughout this page.
3. Read that pivot back against the destination country's bands, in the same geometry, to get the grade there.

Both directions use one linear map read forwards and backwards, which is what makes the conversion continuous (a small change in your grade produces a small change in the answer), monotonic (a better grade never converts to a worse one), and reversible.

Why go through a US GPA at all?

Two reasons, one practical and one structural.

Structural: with 53 grading systems there are 2,756 directed country pairs. Mapping each pair to each other pair directly would mean maintaining 2,756 sets of equivalences, and any inconsistency between them would be invisible. Pivoting through one scale means maintaining 53 mappings, and every pair is derived from them — so the whole system stays consistent by construction.

Practical: it is what credential evaluators do. WES (World Education Services — the service most US and Canadian universities require international applicants to use) converts a foreign transcript into US terms, not into some abstract neutral scale. Using the same pivot means our estimate and their report are answering the same question.

The cost of this choice, stated plainly: the US 4.0 scale has no headroom above an “A”. Where a country maps its whole top range to an A, every grade inside that range converts to the same pivot — an Egyptian 88% and an Egyptian 95% are not distinguishable through it. The result range is the one place that can say so, which is why it widens across a saturated top band instead of pretending to a precision the pivot does not carry.

How is a grade turned into a number? (a worked example)

Take an Indian 78% — the most common single conversion on this site.

India's top band, Distinction, runs from 75% upward and is anchored at a US 3.7 GPA at its entry. Its upper anchor is 4.0, and the grade it puts there is not 100% but 90% — the best mark Indian universities realistically award (see the next section). So 78% sits 20% of the way up that band, giving a pivot of 3.76.

Reading 3.76 back against the UK: it lands inside the Upper Second (2:1) band, which spans 60–69% between GPA anchors 3.30 and 3.85. Interpolating gives 68%, and applying the confidence tolerance gives the window the site actually shows:

🇮🇳 78% in India (Distinction) →
🇬🇧 66–71% — Upper Second (2:1)
US GPA pivot 3.76 / 4.0 · high confidence

Notice what the range does here: it crosses 70%, the First-Class threshold. The honest answer is not “68%” — it is “a strong 2:1, close enough to a First that some institutions will read it that way”. That is the thing you can actually plan around, and a single number would have hidden it.

What this replaced. The first version of this engine returned the midpoint of the nearest band, which meant a US 2.2 and a US 3.0 both converted to a UK 55%, and a 0.1 change from 3.5 to 3.6 jumped the answer 20 points. Interpolating within the band removed both problems.

Why doesn't a perfect grade convert to a perfect grade?

Because most countries never award the top of their own scale, and a converter that ignores that produces grades nobody receives.

UK marking above 85% is vanishingly rare — a 72% is already an excellent First. French marks above 18/20 are exceptional. So each system carries a realistic best alongside its theoretical ceiling, and that is what the top band interpolates against in both directions. Without it, a strong Indian student converting to the UK was being told they had an 85% — a mark that would raise eyebrows at any British university.

14 of the 53 systems have a realistic best below their theoretical top. Each one carries an inline reason in the source code for why that specific number was chosen.

Why is the answer a range, not one number?

A single converted figure carries the visual authority of a calculation that no credential evaluator would reproduce exactly. Two evaluators reading the same transcript can land a band apart, and institutions inside one country grade differently from each other. The window is the honest answer, and it is also the more useful one: “a solid 2:1” is something you can act on.

The width comes from the confidence of the pair:

These are declared uncertainties, not statistical error bars. There is no dataset of conversion outcomes to compute a confidence interval from, and inventing one would be exactly the false precision this method exists to avoid. They are our stated judgement of how much room to leave, and they are the same for every conversion at that confidence level.

A conversion is only as trustworthy as its weaker end, so the pair takes the lower of the two countries' confidence levels. Across the 53 systems: 32 high, 20 medium, 1 low.

Where does the grading data come from, and how is it checked?

Every country's bands were checked against a source that actually shows the bands — not a ministry homepage that mentions grading without publishing a table. In practice that is usually a Nuffic country module (Nuffic is the Dutch national body for international education comparison) or Scholaro's credential-evaluation database. The national education authority is kept alongside as a cross-check where one exists; 51 of 53 countries carry at least one.

Three rules govern this, and one of them is enforced by the build:

• A date means a document was opened. The build rejects a “verified” date unless the source link points at an actual page rather than a bare domain — so a check date can never be stamped on a homepage citation.
• No date is better than a false one. Sweden is deliberately left undated: it has no single national grading scale — universities variously use letter grades, ECTS, or plain pass/fail with no percentage at all — so there is nothing to verify against. Its page says “source-verification in progress” rather than claiming a check that did not happen.
• Two sources before moving a band. Where a second source disagrees with ours, the band does not change on one table alone — a single institution's conversion chart is evidence, not proof.

52 of 53 systems are verified, deep-linked and dated. Every country page shows its own source and check date, so you can go and read the document the numbers came from.

The pass found real errors, which is the point of running it: Egypt's “Excellent” band started ten points too low, over-crediting an entire country; Norway's pass mark was set a grade too high; Singapore, Saudi Arabia and Switzerland each had a top band that began below the published threshold. Those are fixed. The other 49 countries matched their source.

What about countries that use two grading systems at once?

Several do. A Nigerian transcript may carry a 5.0 CGPA (Cumulative Grade Point Average — an average across the whole degree) or a percentage; a Pakistani one may show a 4.0 CGPA or an older Division. The same number means different things on each, so where both are in real use the converter asks which scale your grade is on and reads it against that scale's own bands.

This is live for Nigeria and Pakistan. Other dual-scale countries are noted in a warning on their page until their second scale is researched — the warning is there because entering a percentage into a CGPA field is a silent, confident, wrong answer, and we would rather flag it than let it happen.

Where do the entry requirements come from?

The “where can I apply” comparison and the competitiveness bar measure your grade against published postgraduate entry requirements — a named institution or national body's own stated bar, quoted and deep-linked, on that country's own scale.

A requirement ships only when the page stating it was opened and read. That gives 12 destinations with a published bar. The remaining 41 are excluded from the comparison entirely rather than scored against somebody else's threshold — of those, 3 are listed with the reason no numeric bar exists (France admits on the dossier, Japan by entrance examination, Sweden has no scale to state one on), and 38 are simply not researched yet and are described that way.

The status shown beside each destination is read off the same range you are shown, not off a separately-computed figure. A range entirely above the bar clears it, a range straddling it is borderline, a range below it is below. Deriving the badge from the displayed number is what stops the badge and the number contradicting each other — which they did, at boundaries, when they were computed separately.

What can this tool not do?

The limits are real and worth knowing before you rely on any of it.

• It is not an official credential evaluation. No university will accept a GradeBridge conversion in place of a WES or ECE report. Use it to decide whether ordering one is worth it — see the WES guide.
• It does not know your institution. Each country is modelled as one national standard. Real universities vary within that, and some — Canada especially, where 4.0, 4.3, percentage, 9-point and 12-point scales all exist — vary a great deal.
• It converts one overall grade, not a transcript. It does not weight courses by credit or recalculate a GPA from individual results, which is precisely what a paid evaluation does.
• It does not cover secondary school. All 53 systems are university grading. A-levels, Class 12, the Abitur and US high-school GPA are separate scales, so the tool declines rather than scoring a school grade on university bands.
• It does not predict admission decisions. Grades are one input among references, statements, experience and interviews, and programmes set their own bars.
• It reflects current grading systems. Countries reform their scales; an older transcript may sit on a system that no longer matches what is modelled here.

Found something wrong?

Corrections are genuinely welcome, and a source beats an opinion — if you can point at a published table showing a band we have wrong, it will be checked against the two-source rule and fixed. Email debugt1000@gmail.com with the country and, if you have it, the link.

Frequently asked questions

Try the converter · All 53 grading systems · About GradeBridge