Merchants searching for an HS code lookup usually expect a single official database to hand them one correct number, but the accessibility and browser data available for storefronts shows why the surrounding checkout and product pages that display those codes often fail before the code itself matters.
A shopper typing “hs code lookup” usually assumes the hard part is finding the right ten-digit number, so they expect a database and nothing else. The harder part, and the part this reference actually has data on, is whether the page that displays that number once they find it will load, render its text at a readable contrast, and label its input fields so a screen reader or a Samsung Internet user can actually use it. A correct HS code sitting inside a broken form helps nobody.
No dataset here contains a tariff schedule, a country code list, or a customs classification table, so nothing in this article names an actual HS code. What is measured is the state of the pages merchants build around that number: product pages with a customs note, checkout steps with a duty estimate, or admin panels where a merchant enters the code for a SKU. Those pages are ordinary ecommerce pages, and ordinary ecommerce pages carry the same accessibility and browser problems documented across a million homepages.
The count of detected accessibility errors across the measured homepages increased 10.1% over the period covered. At the same time, WCAG conformance rose from 94.8% to 95.9%. Both numbers come from the same scan of a million homepages, taken from the WebAIM Million dataset as loaded on September 9, 2026, and they describe different things: conformance is the share of pages clearing a baseline, while the error count is the raw volume of individual problems found, including on pages that still pass. A store that added a new customs disclaimer or a new duty calculator widget can pass the same automated checks while introducing new individual errors on that new element.
| Problem | Share of measured homepages |
|---|---|
| Low contrast text | 83.9% |
| Missing alternative text for images | 53.1% |
| Missing form input labels | 51% |
Missing form input labels affect 51% of the measured homepages, and any field where a shopper or a merchant enters or reads an HS code is a form input. If that field has no programmatic label, a screen reader announces nothing useful, and the customs step becomes a guess. Low contrast text, found on 83.9% of pages, tends to concentrate in exactly the kind of small print used for duty disclaimers and shipping notes, the same place an HS code reference often gets tucked. Missing alt text, at 53.1%, matters less for the code itself but matters for any icon or flag graphic used to indicate country of origin next to it.
Samsung Internet accounts for 0.614% of tracked browser usage, and an “Other” category not broken out further accounts for 0.467%. Together that is a small slice of total traffic, but for a store selling internationally, where customs and HS code fields matter most, that slice represents real shoppers on real devices. The figures come from Can I Use’s browser usage frequency table as loaded on September 9, 2026, and they exist to remind a merchant that testing a customs widget only in the two most common browsers leaves a measurable, named gap uncovered rather than an imaginary one.
Put together, these numbers describe a market where the visible headline metric, WCAG conformance, is improving, moving from 94.8% to 95.9%, while the underlying error count is getting worse, up 10.1%. Neither number is specific to customs or HS code fields, and no figure in these datasets names a tariff schedule or a duty rate. What they do establish is that a merchant relying on any form field, including one holding a customs classification, is building on a page environment where roughly half of homepages mislabel their inputs and a large majority carry contrast problems somewhere on the page.
A merchant who needs an actual HS code should still go to a national customs tariff database or a licensed broker, because that number does not live in any dataset cited here. Once that number is in hand, the next step is checking the specific field or page displaying it for a proper label and adequate contrast, rather than assuming the site-wide averages above do not apply to that one page.
This reference does not host a lookup tool itself, so a merchant needs a national customs tariff database or a broker's schedule to get the number itself. What this page does cover is the reliability of the pages ecommerce sites build around that number, because a code is only useful if the checkout or product page displaying it loads correctly and reads correctly for every visitor. The measured facts here come from accessibility and browser usage datasets, not from a tariff database, so treat this as a check on delivery, not a source for the code itself. If a store's customs field or duty calculator sits inside a form, the same input-labeling problems documented below apply to it directly.
Missing form input labels were found on 51% of measured homepages, and an HS code or customs field is a form input like any other. If that field lacks a proper label, screen reader users cannot tell what number they are supposed to enter or read, which turns a compliance requirement into a broken checkout step. The number comes from a scan of a million homepages, not from customs-specific pages, but the same templates and form components tend to carry the same labeling gaps across a site.
Low contrast text was found on 83.9% of the measured homepages, and duty disclaimers or HS code notes are often styled as small print, which is exactly the kind of text prone to contrast problems. There is no separate measurement for customs text alone in the facts here, so this is a reasonable inference from the general rate rather than a specific finding. A merchant who wants certainty should check the actual contrast ratio of that specific line of text rather than rely on the site-wide figure.
Samsung Internet holds 0.614% of tracked usage and an unnamed 'Other' category holds 0.467%, together a small but nonzero share of traffic. If a store's customs or HS code lookup depends on a script that behaves differently across browsers, that combined share is enough real visitors to justify a manual check rather than assuming universal support. Skipping it means a small, specific and known group of shoppers can hit a broken field with no fallback.
WCAG conformance moved from 94.8% to 95.9%, which is an improvement, but the count of detected errors rose 10.1% over the same period. Those two figures are not contradictory: a page can pass more of the automated conformance checks while still accumulating more individual errors elsewhere on the page. For a merchant, the practical takeaway is that overall trend lines look slightly better while the raw number of things that can go wrong on any single page has grown.