How this instrument works
A GTIN check digit is the last digit of a barcode number — the twelfth in a UPC-A, the thirteenth in an EAN-13 — and it isn't scanned as separate data. Every barcode reader recomputes it from the digits that precede it and rejects the read if the two don't match, which is why a barcode that scans cleanly is provably free of the single most common data-entry mistake: a mistyped or misread digit.
The formula sums the values sitting in odd and even positions separately, multiplies the odd-position total by three, adds the even-position total, and rounds the result up to the next multiple of ten. That asymmetric three-and-one weighting is deliberate: giving both sums equal weight would let some digit swaps and typos cancel each other out and pass as valid anyway. GS1's General Specifications fix this exact weighting for every GTIN length, so a scanner in Tokyo and one in Toledo run identical arithmetic on identical numbers.
A manufacturer assigns the check digit once, when a product's barcode is first printed, using a company prefix and item reference issued by GS1 or a national member organization; a warehouse clerk or point-of-sale developer later runs the same arithmetic to validate incoming scans or catalog records before they reach a database. What the arithmetic cannot do is confirm the number belongs to a real, registered product — only that the eleven figures entered are internally consistent with the twelfth.
- Type the first eleven digits of your barcode into Digit 1 through Digit 11, left to right, exactly as printed above the bars.
- Watch Sum of odd positions (1,3,5,7,9,11) and Sum of even positions (2,4,6,8,10) update — the two running totals the algorithm weighs differently.
- Read GTIN check digit (position 12) — that is the twelfth digit your barcode needs to carry to be scannable.
- Compare it against the digit already printed on the label or spec sheet; a mismatch usually means a transposed or mistyped figure somewhere in positions one through eleven.
Worked example — the Vicks VapoRub UPC-A barcode
Take the real UPC-A barcode 036000291452, printed on a bottle of Vicks VapoRub and used in countless barcode tutorials. Its first eleven digits, entered as Digit 1 through Digit 11, are 0, 3, 6, 0, 0, 0, 2, 9, 1, 4, 5. The odd positions — first, third, fifth, seventh, ninth, eleventh — read 0, 6, 0, 2, 1, 5 and sum to 14; the even positions — second, fourth, sixth, eighth, tenth — read 3, 0, 0, 9, 4 and sum to 16.
Multiply the odd sum by three and add the even sum: 14 × 3 + 16 = 58. The nearest multiple of ten at or above 58 is 60, so 60 − 58 = 2 is the check digit the instrument returns. Look at the twelfth character actually printed under the bars on that real bottle — it is a 2, confirming the arithmetic against a barcode anyone can find on a store shelf.
Questions
What does a GTIN check digit actually protect against?
It catches the two most common ways a barcode number gets corrupted: a single character typed or scanned wrong, and two neighboring characters swapped. The weighted mod-10 sum makes almost every single-character error and most swaps change the result, so a scanner or data-entry form can flag a bad number before it ships the wrong product to the wrong shelf.
Why do odd and even positions get different weights?
Multiplying the odd-position sum by three and leaving the even-position sum at one creates an asymmetric pattern that a simple swap or single mistake almost always disturbs. If both sums used equal weight, some transpositions would cancel out and slip through undetected; the three-and-one split, fixed in the GS1 General Specifications, is what makes the scheme reliable enough for point-of-sale systems worldwide.
Does this same arithmetic work for EAN-13 or GTIN-14 barcodes?
The weighted mod-10 idea applies to every GTIN length — EAN-13, GTIN-14, even the older 8-character code — but the count of figures and which positions land odd or even shift with the total length. This instrument is built for the eleven-figure input used in a twelve-figure UPC-A; longer formats need their own twelve or thirteen-figure input before the same three-and-one weighting produces a result.
What happens when the weighted total is already a multiple of 10?
The check digit becomes 0, not 10. Feed in eleven zeros and both position sums are zero, so the weighted total is zero — already a multiple of ten — and the formula's final mod-10 step turns that into a result of 0 rather than a two-character value. A real barcode never carries a two-character check position; the wraparound is built into the arithmetic for exactly this case.
My computed check digit doesn't match the one printed on the product — what's wrong?
Almost always one of the eleven entered figures was misread, mistyped, or placed in the wrong slot. Recount straight from the printed number above the bars, confirm nothing shifted left or right, and check that no leading zero got dropped — UPC-A numbers commonly start with a zero, and spreadsheets love to strip it on import.
Where does a business get the first eleven digits from in the first place?
GS1, or a national member organization such as GS1 US, assigns a company prefix, and the business itself assigns item-reference figures within that prefix; together they form the first eleven digits of a UPC-A. A manufacturer works out the twelfth figure with this same arithmetic before a barcode is ever printed on packaging or submitted to a retailer's item-setup system.
References
- GS1 — GTIN Standard, Structure and Specifications
- GS1 — Official Check Digit Calculator and Formula Reference
Read this first: This instrument shows arithmetic, not advice. Real offers add fees, taxes and terms that vary by lender and place — verify the figures against your actual paperwork before deciding anything.