BITV-Test: Accessibility Audits in Germany

The BITV-Test is a standardised German testing procedure with 98 test steps, applied by trained testers against EN 301 549 and quality-assured by a second person. An audit is the general term for any expert accessibility review, with no fixed methodology. A VPAT is a template, and once filled in with documented results it becomes an ACR. If a German client asks for a BITV-Test, an audit report will not satisfy the request, and neither will an ACR.

You cannot run a BITV-Test yourself and then certify the result. That is the single most common misunderstanding among suppliers outside Germany, and it is worth knowing before you promise a deadline.

This is not legal advice. This page describes how the procedure works. Nothing here obliges you to use it; the BFSG requires accessibility and prescribes no particular test.

Key points

• The BIK BITV-Test covers 98 test steps assessing 88 requirements in the web variant. The figure of “around 60” that circulates in English articles belongs to a different variant, the WCAG test with 60 test steps for 50 requirements.
• It tests against the whole of EN 301 549, not only the web clause. Documentation, support channels and video capabilities are part of it.
• The points system was abolished on 15 March 2019. The seals “90plus” and “95plus” no longer exist. Every requirement is either met or not.
• You cannot run it yourself. It is coordinated by DIAS GmbH and carried out by trained testing bodies, including external quality assurance by a second person.
• The testing body picks the pages, not you: at least 3 for simple offerings, 5 to 10 or more for complex ones, each including complete processes such as login or checkout.
• Automated tools find a quarter to a half of all barriers. They check whether an attribute exists, not whether its content is correct.
• Price is driven by page types and processes, not by page count, and by how much preparatory work you have done.
• Since November 2024 there is a BITV-Test plus WCAG 2.2 variant covering the current WCAG level.

What the BITV-Test is

The BIK BITV-Test has existed since 2004 and is the established German testing procedure. It is coordinated by DIAS GmbH and carried out by a nationwide network of trained testing bodies. The name comes from the public-sector regulation described on BITV 2.0: German Public Sector Rules, but the procedure is used in the private sector too, because it is the most recognised evidence available on the German market.

Nothing in the BFSG requires it. Companies use it when they need evidence that survives a procurement process or a complaint, described on BFSG Enforcement: Status in Germany.

98 test steps, not 60

The web variant currently comprises 98 test steps assessing 88 requirements. The often-quoted 60 belongs to the WCAG test, a separate variant with 60 test steps for 50 requirements.

The difference is substantive. The BITV-Test assesses against the whole of EN 301 549, so it includes test steps for areas the WCAG do not cover at all:

• Clause 5, generic requirements: 3 test steps
• Clause 6, two-way voice communication: 12 test steps
• Clause 7, video capabilities: 9 test steps
• Clause 9, web, the WCAG core: around 55 test steps
• Clauses 11.7 and 11.8, user preferences and authoring tools
• Clause 12, documentation and support: 5 test steps

Clause 12 surprises most teams. Whether the help pages are accessible and whether support can be reached through an accessible channel is part of conformity.

How it is scored

This is where older English articles are still wrong. Until 2019 there was a points system with 100 available points, a threshold of 90 for “well accessible” and 95 for “very well accessible”, along with the “90plus” and “95plus” seals.

Since 15 March 2019 the points system is gone. The reasoning is consistent: WCAG conformance is all or nothing, since a page either meets a level in full or it does not. A score of 92 per cent could not be reconciled with that.

Scoring now runs in two stages:

1. Each test step is rated on a five-point scale: met, mostly met, partly met, mostly not met, not met.
2. A requirement counts as conformant only if all of its test steps were rated “met” or “mostly met”. Anything below counts as not met.

The practical consequence is that passing is demanding. There is no partial grade to hide a few open points behind. Claiming conformity means all 88 requirements on all tested pages.

Which pages get tested

You cannot choose the test pages. The testing body selects them independently and representatively. As a rule of thumb:

• at least 3 pages for simple, small, uniformly built offerings,
• 5 to 10 or more pages for complex offerings.

The selection is not about arbitrary single pages. It covers the typical page types and, importantly, complete processes. A contact form or a checkout is walked from start to finish, because the WCAG conformance conditions require it. A checkout that works on step one and breaks on step three is a failed checkout.

Procedure and seal

1. Individual assessment by a trained tester, step by step through the catalogue, with a finding and a screenshot for each defect.
2. External quality assurance by a second, independent person. This is the step that separates the BITV-Test from an ordinary audit.
3. Report, generated as HTML and as PDF/UA, so the report itself is an accessible document.
4. Seal. With a representative selection and full conformance, the offering may carry the “BIK-konform” mark. Without 100 per cent, you have a test report.

There is also a BITV self-assessment: the same catalogue as a tool to fill in yourself. It is a good preparation instrument. It is not evidence and it carries no seal.

BITV-Test or audit

My advice from practice: audit first, BITV-Test afterwards. An audit during development finds problems while they are still cheap to fix. The BITV-Test at the end confirms the result. Reversing the order means paying a premium for someone to write down a long list of defects in a standardised format.

VPAT, ACR and the German equivalent

A VPAT (Voluntary Product Accessibility Template) is a template published by the US industry association ITI. On its own it is an empty form. Filled in with documented test results it becomes an ACR (Accessibility Conformance Report). Saying “we have a VPAT” therefore means “we have a form”.

The template comes in several editions depending on the standard being documented:

• WCAG edition, the success criteria only
• Revised Section 508, the US federal standard
• EU edition, EN 301 549, which is the relevant one for Germany
• INT edition, all three combined

Each criterion is rated with one of four statements: Supports, Partially Supports, Does Not Support, Not Applicable. For the middle two the template requires “Remarks and Explanations”. An ACR with “Supports” everywhere and no remarks is a warning sign rather than a badge, and experienced German procurement officers read it that way.

In the European context the closest counterpart is the accessibility statement, or a conformance report against EN 301 549. Both exist for the same purpose: making clear what is met and what is not. What the German statement has to contain is on German Accessibility Statement: Rules.

What drives the price

There are no flat rates, and the cost drivers are predictable:

• Number of page types, not pages. Five thousand news articles from one template are one page type. Five hand-built landing pages are five.
• Number of processes. Login, checkout, form journeys. Each has to be walked in full.
• Component complexity. Data tables, maps, charts, media players and comboboxes consume testing time. Static text pages barely register.
• State of the preparatory work. An offering that still contains obvious defects spends testing time on findings you could have found yourself.

The last point is the lever you control. The more you find and fix in advance, the more of the testing budget goes into the difficult cases instead of missing alt text.

How to prepare

1. Run automated scans and fix everything a tool finds. That is the baseline.
2. Walk every important process with the keyboard only. In my experience this finds most of the serious defects.
3. List your page types and processes. The testing body needs this list to quote anyway, and having it ready gets you a firm price faster.
4. Document known gaps instead of hiding them. They will surface, and the list is the basis for your statement.
5. Settle who fixes the findings. A report with no budget behind it is money spent for nothing.

FAQ

Can we run a BITV-Test ourselves?

Not with certifying effect. The procedure requires a trained testing body and external quality assurance by a second, independent person. The official BITV self-assessment uses the same catalogue and is a good preparation tool, but it produces no seal.

Is a BITV-Test mandatory under the BFSG?

No. The BFSG requires accessibility and prescribes no test procedure. A documented audit plus an honest statement is enough for many companies. The test becomes relevant when a tender asks for evidence or a dispute needs a neutral finding, as described on BFSG: Germany’s Accessibility Act 2025.

Does an ACR satisfy a German client?

It depends on what the contract asks for. An ACR in the EU edition documents against the right standard and is often accepted. Where the contract names the BITV-Test specifically, only the procedure itself will do, because the seal depends on the second review.

How long does a test result stay valid?

Formally without limit, in practice until the next significant change. A relaunch, a new CMS template or a newly embedded widget can invalidate the statement, which is why an accessibility statement always carries an assessment date.

Do the testers work in German?

The procedure, the catalogue and the report are German. Reports can be discussed in English with most testing bodies, and the findings themselves are WCAG criteria that any accessibility specialist will recognise. Budget for translation if the report has to go to a non-German team.

Where do we find a testing body?

Through the BIK testing network, whose current list is maintained at bitvtest.de. If you want to talk through whether a test is the right step at all, the contact route is on About: German Accessibility Law in English.

Sources

• Description of the test procedure (web) - test steps, rating scale, page selection (in German)
• ACR/VPAT FAQ - editions and conformance statements, Section508.gov
• EN 301 549 V3.2.1 - the standard being tested against, free PDF