Tool AT-W14REV. 2026-A

BFSG check: must your website be accessible?

The Accessibility Strengthening Act (Barrierefreiheitsstärkungsgesetz, BFSG) has applied since 28 June 2025 to websites through which private individuals buy or book something or make a binding appointment. Pure information sites are not covered. Micro-enterprises with fewer than ten employees and a turnover or balance sheet total of no more than €2 million are exempt for services, but not when selling certain devices. The BFSG check clarifies your case in six questions, names the section for each step and then tests twelve common barriers.

  • 6 questions and 12 checks
  • Section cited at every step
  • runs in your browser only

Does the BFSG apply to your website?

Self-assessment · 6 questions · Self-test · 12 items

Not legal advice. The check matches your answers to the text of the law and the authorities’ guidelines and names the provision for each step. Whether a borderline case is covered is decided in a dispute by the market surveillance authority or a court; for a legal assessment of your case, a solicitor or your chamber is the right contact.

The evaluation runs in your browser. Your answers are neither sent nor stored.

BFSG check of

Aureum Technology GmbH · aureum-tech.com/tools/bfsg-check

Part 1 · Does the law apply to you?

Question 1 of 6Do private individuals also buy or book from you?

This means consumers: people who are not buying for their trade or self-employed work.

Question 2 of 6What can private customers do on your website?

Include an app too, if there is one. A calendar in which a binding appointment can be booked counts as booking.

Question 3 of 6Do you provide any of these services for private customers?

Telephone or internet connections, banking services, e-books, passenger transport by air, long-distance coach, rail or ship. The law names them individually, regardless of the website.

Question 4 of 6How many employees does your business have?

Counted in full-time positions over the year: two part-timers working half days make one position. Apprentices and employees on maternity or parental leave do not count.

Question 5 of 6Is your annual turnover or annual balance sheet total no more than €2 million?

It is enough if one of the two figures is below that. Both are in the annual accounts.

Question 6 of 6Do you sell devices that the law expressly names?

Computers, laptops and tablets including their operating system for consumers, smartphones, routers and telephones, TVs and receivers with an on-screen menu, e-book readers, payment terminals and self-service machines. Clothing, food or tools are not included.

Result, part 1

BFSG: still open

Answer the questions and this will show whether the law applies to you, with the section for each step.

6 questions still open.

Matching of your answers to the text of the law, not legal advice. Legal bases last checked on 13.11.2026.

Part 2 · Self-test: twelve common barriers

Regardless of the result above: twelve checks, each taking about a minute and needing no expertise. Open your website in a second window for this. It does not replace an expert assessment. In one comparison, automatic testing tools found only 14 to 38 per cent of violations Vigo 2013, and non-experts miss more than experts Brajnik 2011. The test shows where improvements are clearly needed.

Item 1 of 12 · Keyboard-only operationCan you reach everything using only the Tab key?

How to check: Put the mouse aside. Click into the address bar, then press Tab again and again; Shift + Tab goes back, Enter opens links and buttons. Can you get into the menu, through the form to the submit button and out again without getting stuck anywhere?

WCAG 2.1: 2.1.1 Keyboard (A), 2.1.2 No Keyboard Trap (A), 2.4.3 Focus Order (A)

Item 2 of 12 · Visible focusWhen tabbing through, can you always see where you are?

How to check: In the same run-through, watch the highlight: a frame, underline or colour change on the element currently selected. If it disappears anywhere, for example in the menu or on coloured buttons, that is a gap.

WCAG 2.1: 2.4.7 Focus Visible (AA)

Item 3 of 12 · Cookie banner by keyboardCan your cookie banner be operated using only the keyboard?
Item 4 of 12 · Sufficient contrastIs every text easy to read even in poor light?

How to check: On your phone, turn the brightness right down or step over to a window. Light grey text on white, white lettering on photos and pale button labels stand out immediately. You can measure it exactly with a free contrast checker: you need 4.5 : 1, or 3 : 1 for large type.

WCAG 2.1: 1.4.3 Contrast (Minimum) (AA), 1.4.11 Non-text Contrast (AA)

Item 5 of 12 · Zooming to 200 %Does the page stay fully usable at 200 % zoom?

How to check: On a PC, press Ctrl and + until the browser shows 200 %. Is all the text still there, does nothing overlap or disappear, and does the menu work? If you have to scroll sideways to read normal text, that is a gap.

WCAG 2.1: 1.4.4 Resize Text (AA), 1.4.10 Reflow (AA)

Item 6 of 12 · Descriptions for images and iconsDoes the screen reader say what is in images and on icon buttons?

How to check: Switch on VoiceOver on an iPhone, TalkBack on an Android phone (Settings › Accessibility), or the screen reader on a Windows PC with Ctrl + Windows + Enter. Then move over images, the logo and icons: do you hear what is shown or where the button leads? Or only “Image”, a file name or nothing at all?

WCAG 2.1: 1.1.1 Non-text Content (A)

Item 7 of 12 · Labelled form fieldsDoes every form field have a visible label that stays in place?

How to check: Click into the contact or order form and type something. If the only label disappears because it was just a grey placeholder inside the field, that is a gap. With the screen reader switched on you can also hear whether each field announces its name.

WCAG 2.1: 1.3.1 Info and Relationships (A), 3.3.2 Labels or Instructions (A), 4.1.2 Name, Role, Value (A)

Item 8 of 12 · Clear error messagesDoes the form say in words what is wrong?

How to check: Deliberately submit the form with errors: leave a required field empty, enter an e-mail address without an @. Does the field then say in words what is missing and how to put it right? A red border alone is not enough.

WCAG 2.1: 3.3.1 Error Identification (A), 3.3.3 Error Suggestion (AA)

Item 9 of 12 · CaptionsDo your videos with speech have captions?

How to check: Play each video on the website once without sound. Can you understand everything important from the captions? Proofread automatically generated captions before they go online.

WCAG 2.1: 1.2.2 Captions (Prerecorded) (A), 1.2.5 Audio Description (Prerecorded) (AA)

Item 10 of 12 · Page languageIs German given as the language in the source code?

How to check: Open the source code with Ctrl + U and search for lang= with Ctrl + F. If the first lines show lang="de", all is well.

WCAG 2.1: 3.1.1 Language of Page (A)

Item 11 of 12 · Meaningful link textCan you tell from each link text alone where it leads?

How to check: Skim a page for links only. “Here”, “more” or “read on” say nothing without the sentence around them, and icon links such as a telephone handset need a name that can be heard. Screen readers like to read links out as a list; that is exactly how they will then come across.

WCAG 2.1: 2.4.4 Link Purpose (In Context) (A)

Item 12 of 12 · Readable PDFsCan your important PDFs be read with a screen reader?

How to check: Open an important PDF, such as a price list, menu or registration form. Can you select the text with the mouse? If not, it is a scanned image that no screen reader can read. In Adobe Reader, File › Properties also shows whether the PDF is “tagged”.

EN 301 549, clause 10 (documents)

Result, part 2

Self-test: still open

Work through the twelve items and the traffic light will appear here with all the gaps.

12 items still open.

Self-assessment, not a test under EN 301 549. The traffic light turns red as soon as one barrier locks people out completely or three gaps come together.

Your answers

What the Accessibility Strengthening Act requires

The BFSG implements EU Directive 2019/882, the European Accessibility Act. For the first time, private companies must offer certain products and services in an accessible form. Accessible means that people with disabilities can use them in the usual way, without particular difficulty and in principle without help from others (§ 3 Abs. 1). For most small businesses, one single item in the law matters: services in electronic commerce. That means everything through which private customers conclude a contract online.

The provisions at a glance
StandardWhat it saysWhat it means for you
§ 1 Abs. 2 BFSGExhaustive list of products: computers and operating systems for consumers, payment terminals and self-service machines, terminal equipment for telephone, internet and television, e-book readers.Anyone who makes, imports or sells such devices is covered, even as a micro-enterprise.
§ 1 Abs. 3 BFSGServices for consumers: telecommunications, parts of passenger transport, banking services, e-books, services in electronic commerce.For most businesses only the last item counts: the contract concluded through the website.
§ 1 Abs. 4 BFSGExceptions: videos and office files published before 28.06.2025, online maps, third-party content, archives.Older PDFs and videos may stay as they are. New ones must be accessible.
§ 2 Nr. 17 BFSGMicro-enterprises: fewer than ten employees and annual turnover or annual balance sheet total of no more than €2 million.The threshold for the exemption. One of the two money figures is enough.
§ 2 Nr. 26 BFSGService in electronic commerce: via website or app, at a consumer’s request with a view to a contract.Shop, booking and binding appointment agreements count; a pure information site does not.
§ 3 BFSGAccessible means that people with disabilities can find and use something without particular difficulty and in principle without help from others. Para. 3: exemption for micro-enterprises for services.The measure is use, not a seal of approval.
§ 14 BFSG, Anlage 3Duties of the service provider: meet the requirements, publish accessibility information, keep it up to date, fix and report defects, give the authority information.The core of your duties if the law applies to you.
§§ 3, 12, 19 BFSGVState of the art; websites perceivable, operable, understandable and robust; in a shop, payment and sign-in as well.The technical requirements. The Federal Agency for Accessibility publishes which standards apply.
§ 17 BFSG, Anlage 4Disproportionate burden as an exemption, with assessment, documentation and repetition every five years.Not a free pass: anyone who relies on it must be able to prove it.
§§ 29, 32, 37 BFSGMarket surveillance through notices and prohibitions, applications from consumers and associations, fines of up to €100,000 or €10,000.Oversight by the joint authority of the federal states in Magdeburg.
§ 38 BFSGTransition until 27.06.2030 for products that a service provider was already using before, and for contracts from before 28.06.2025.There is no transitional period for the website itself.

Anyone affected has two main tasks. The website must meet the requirements, and there must be a public description of how it does so, naming the competent market surveillance authority (§ 14 Abs. 1, Anlage 3). This accessibility information goes in the terms and conditions (AGB) or in another clearly visible place, for example on a page of its own linked in the footer.

The ordinance does not set out in detail which technical rules apply. It requires the state of the art and leaves the list of standards to the Federal Agency for Accessibility (§ 3 BFSGV). For websites, that agency points to EN 301 549 and thus to WCAG 2.1 at levels A and AA. You can find the wording in the act and in the ordinance.

Provisions checked against the text of the law on 13.11.2026.

How the check decides

The check follows the path of the law in a fixed order. First: is your offer aimed at consumers? Anyone who sells only to companies, self-employed people or public bodies need go no further in respect of their service at this point. Then: is a contract initiated or concluded through the website? A shop counts, and so does a calendar in which an appointment is booked bindingly. As an example, the guidelines of the Federal Ministry of Labour and Social Affairs explicitly name a hairdresser whose customers book appointments online.

Only after that comes size. Micro-enterprises are exempt for services (§ 3 Abs. 3). These are businesses with fewer than ten employees whose annual turnover or annual balance sheet total is no more than €2 million. The check examines the devices under § 1 Abs. 2 separately, because there is no exemption for them.

For “I don’t know” it works through every possible answer. If the same result always comes out, it says so; otherwise it names both results and the detail they depend on. Guessing would be the worse answer to a legal question. You may know the same approach from the E-invoice check.

01Do private customersbuy from you?02Order or bookvia the website?03Micro-enterprise?under 10 employeesyesyesnononoyesnot covered§ 1 Abs. 3 BFSGnot covered§ 2 Nr. 26 BFSGexempt§ 3 Abs. 3 BFSGaffected§ 14 BFSG, BFSGVSelling devices under § 1 Abs. 2: affected,even as a micro-enterprise (§ 11 BFSG)

Four businesses, four results

Beauty salon without online booking

not affected

Three employees, appointments by phone only, and the website shows services and opening hours. No contract is concluded through the website, so it is not a service in electronic commerce.

§ 2 Nr. 26 BFSG

Beauty salon with online booking

exempt

The same salon now books appointments through a calendar on the website. That is covered. With three employees and under €2 million turnover, however, it is a micro-enterprise. With twelve employees it would be affected.

§ 1 Abs. 3 Nr. 5, § 3 Abs. 3 BFSG

Trade business with a contact form

probably not

15 employees; the website shows references and has an enquiry form. Such a form usually serves for first contact. It would be different if customers could use it to place binding orders.

§ 2 Nr. 26 BFSG

Online clothing shop

depends on the goods

Two people run the shop. As a micro-enterprise they are exempt for the shop. If they add smartphones to the range, retailer duties apply to those devices, because there is no exemption for products.

§ 1 Abs. 2, § 3 Abs. 3, § 11 BFSG

A “not affected” does not mean that accessibility does not matter. The ministry’s guidelines point to an ageing customer base, and in trials people without disabilities also reached their goal faster on more accessible sites Schmutz 2016. How we build websites that have this built in from the start is described under Web design; for local businesses, under Web design in the Vogtland.

What accessibility means in practice

The ordinance sums it up in four words: websites must be perceivable, operable, understandable and robust (§ 12 BFSGV). In everyday terms, that means, for example: what someone cannot see, they must be able to hear, for instance through an image description. Anyone who does not use a mouse can get everywhere with the keyboard. Error messages explain themselves without guesswork. And the page works together with assistive technology such as a screen reader instead of tricking it. How easy your texts are to read is measured by the Readability check.

The twelve items in the self-test cover the barriers that are noticed most often. An annual automated measurement of one million home pages found recognisable WCAG errors on 95.9 per cent in February 2026; most often too weak contrast (83.9 per cent of pages) and missing alternative texts (53.1 per cent), according to WebAIM Million 2026. The table links each item to its criterion.

The twelve check items and how they relate to WCAG 2.1
Check itemCriterion (level)In the check
Keyboard-only operation2.1.1 Keyboard (A), 2.1.2 No Keyboard Trap (A), 2.4.3 Focus Order (A)locks people out
Visible focus2.4.7 Focus Visible (AA)locks people out
Cookie banner by keyboard2.1.1 Keyboard (A), 2.4.3 Focus Order (A), 1.4.3 Contrast (Minimum) (AA)locks people out
Sufficient contrast1.4.3 Contrast (Minimum) (AA), 1.4.11 Non-text Contrast (AA)locks people out
Zooming to 200 %1.4.4 Resize Text (AA), 1.4.10 Reflow (AA)makes it harder
Descriptions for images and icons1.1.1 Non-text Content (A)locks people out
Labelled form fields1.3.1 Info and Relationships (A), 3.3.2 Labels or Instructions (A), 4.1.2 Name, Role, Value (A)locks people out
Clear error messages3.3.1 Error Identification (A), 3.3.3 Error Suggestion (AA)makes it harder
Captions1.2.2 Captions (Prerecorded) (A), 1.2.5 Audio Description (Prerecorded) (AA)makes it harder
Page language3.1.1 Language of Page (A)disrupts
Meaningful link text2.4.4 Link Purpose (In Context) (A)disrupts
Readable PDFsEN 301 549, clause 10 (documents)makes it harder

For contrast, WCAG requires a ratio of at least 4.5 : 1, and 3 : 1 for large type of at least 18 points (or 14 points bold). Zooming must work up to 200 per cent without losing content. And every image that conveys something needs an alternative text; purely decorative images get an empty alt="" so that the screen reader skips them.

What to do if the law applies to you

  1. Take stock. Do the self-test above and run two automatic testing tools with different test engines. Each covers only part of the criteria, and each a different part. Frazão 2020
  2. Start with the template. Menu, header, footer, forms and colours are all in the template. Anyone who improves them there fixes the fault on all pages at once.
  3. Go through the ordering and booking path. Once with the keyboard only, once with a screen reader, each time up to the confirmation. Payment and sign-in are part of it (§ 19 BFSGV).
  4. Publish accessibility information. Describe how your service meets the requirements and name the competent market surveillance authority (Anlage 3 BFSG).
  5. Keep it up to date. Create new pages, PDFs and videos in accessible form from the start. The law requires changes to be taken into account on an ongoing basis (§ 14 Abs. 3).

The cookie banner deserves a separate look. It sits in front of the whole page, and anyone who cannot close it with the keyboard gets nowhere. The simplest option is a website that does not need one; how that works is described in the guide Website without a cookie banner. Which third-party services your page loads today is shown by the Third-party check.

Who enforces it, and what happens in case of breaches

The federal states have set up a joint body for enforcement: the market surveillance authority of the federal states for the accessibility of products and services, MLBF for short, based in Magdeburg. Consumers can report barriers to it and apply for proceedings, as can recognised associations (§ 32 BFSG). The business concerned then gets an opportunity to comment.

The procedure is in stages (§ 29 BFSG). If the authority finds a defect, it first calls on the business to fix it within a reasonable period. If nothing happens, a second call follows, with a warning that the offer will be prohibited. Only after that can it order the service to be stopped until the business proves conformity.

There are also fines (§ 37 BFSG): up to €100,000 for offering a service that does not meet the requirements, and up to €10,000 for, say, a failure to notify the authority or a refused disclosure. These are maximum amounts. What is imposed in an individual case is not in the law, and we will give figures from practice only once reliable ones exist.

What the research says

The structure and weighting of the self-test rest on peer-reviewed studies. Six findings shaped the tool most; the full list is further down in the section Studies and sources.

Testing tools

14–38 %

of the actual violations were found by six widely used tools; they did not check half of the criteria at all. More recent comparisons show that each tool finds different things. Vigo 2013Frazão 2020Ismailova 2022

Guidelines alone

50,4 %

of the problems encountered by 32 blind users were covered by WCAG criteria. Blind and sighted test participants also found largely different problems. Power 2012Petrie 2007

Lost time

30,4 %

of their time online was lost to barriers by 100 blind users. At the top were a confusing page layout and unlabelled forms, with missing image descriptions close behind. Lazar 2007

Expertise

3 instead of 14

Three experts found all the problems in one evaluation; it took 14 non-experts. Their judgements were also less reliable. Brajnik 2011

Benefit for all

reaching the goal faster

People without disabilities also completed more tasks, faster, on more accessible sites. In one shop trial, however, no measurable differences appeared; only the free-text answers were better there. Schmutz 2016Schmutz 2017Vollenwyder 2023

Older people

Guidelines plus testing

On the site that followed the guidelines most closely out of three, older users completed the most tasks. They were not automatically faster or more satisfied there, though. Hart 2008

What follows from this in the tool: the self-test relies on manual checks that no tool carries out, and weights them by whether a barrier locks people out, not by the level of the criterion. The result states explicitly that it does not replace an expert assessment. A 14-year longitudinal study also shows that websites improved mainly when technology and construction methods changed Hanson 2013; hence the advice to start with the template. For apps, which the BFSG covers just the same, things are no better: of 479 Android apps, 94.8 per cent had violations, above all missing focus and missing descriptions Yan 2019.

What the check cannot do

  • It is not legal advice. Borderline cases, such as a form through which orders are accepted as binding, or bookings through a third party’s platform, belong with a solicitor or your chamber.
  • It does not replace a test under EN 301 549. Twelve items are an extract from 50 criteria. How well your website can really be used is shown only by a test with people who work with a screen reader every day.
  • It does not test apps. The BFSG applies to apps in the same way; the testing methods here are written for websites.
  • Sector rules only in outline. For banks, telecommunications and passenger transport the ordinance sets additional requirements that the check does not ask about individually.

How well your website is found on Google and in AI assistants is a different question; SpotLens answers it. What we measured on this website ourselves is in the audit report.

Using this tool from an AI assistant

Assistants such as Claude or ChatGPT that speak the Model Context Protocol (MCP) can call this page’s calculation directly instead of rebuilding it. The server is public, needs no sign-in and stores nothing. The answer always includes the link to this page, the date of the data and the limits of the result.

Endpoint
https://aureum-tech.com/mcp
Tool
werkzeug_bfsg_check
Transport
Streamable HTTP, read-only
Inputs
NameRequiredMeaning
kundennoDo private individuals also buy or book from you? Consumers are meant: people who are not buying for their trade or self-employed work. Values: verbraucher = Yes, private individuals too; nur_firmen = No, only companies, self-employed people or public bodies; weiss_nicht = I don’t know.
websitenoWhat can private customers do on your website? Include an app too, if there is one. A calendar in which an appointment is booked bindingly counts as booking. Values: info = Only find information (services, phone number, e-mail address, directions); kontakt = Find information and use a contact form (for general enquiries or a callback request); vertrag = Order, book or conclude a contract (shop, booking calendar, online registration, appointment booking); weiss_nicht = I don’t know.
branchenoDo you provide any of these services for private customers? Telephone or internet connections, banking services, e-books, passenger transport by air, long-distance coach, rail or ship. The law names them individually, regardless of the website. Values: ja = Yes; nein = No; weiss_nicht = I don’t know.
beschaeftigtenoHow many employees does your business have? Counted in full-time positions over the year: two part-timers working half days make one position. Apprentices and employees on maternity or parental leave do not count. Values: unter10 = Fewer than ten; ab10 = Ten or more; weiss_nicht = I don’t know.
umsatznoIs your annual turnover or annual balance sheet total no more than €2 million? It is enough if one of the two figures is below that. Both are in the annual accounts. Values: bis2 = Yes, at least one of the two; ueber2 = No, both are above; weiss_nicht = I don’t know.
produktenoDo you sell devices that the law expressly names? Computers, laptops and tablets including their operating system for consumers, smartphones, routers and telephones, TVs and receivers with an on-screen menu, e-book readers, payment terminals and self-service machines. Clothing, food or tools are not included. Values: ja = Yes; nein = No; weiss_nicht = I don’t know.
selbsttestnoOptional: result of the self-test of the twelve most common barriers, per item ok, luecke (gap), weiss_nicht (not checked) or entfaellt (does not apply). Keys: tastatur = Keyboard-only operation; fokus = Visible focus; cookie = Cookie banner by keyboard; kontrast = Sufficient contrast; zoom = Zooming to 200 %; alt = Descriptions for images and icons; formular = Labelled form fields; fehler = Clear error messages; video = Captions; sprache = Page language; linktexte = Meaningful link text; pdf = Readable PDFs. “entfaellt” only for cookie, formular, fehler, video and pdf.

Questions where an assistant should call it

  • “I have a beauty salon with three employees and online appointment booking. Does my website have to be accessible?”
  • “Does the BFSG apply to a tradesman’s website with a contact form?”
  • “We are an online shop with 12 employees. What do we have to do under the Accessibility Strengthening Act?”
  • “Is there an exemption from the BFSG for micro-enterprises, and does it also apply if I sell laptops?”

How the connection works and what MCP can do in your own business is explained on the page about MCP integrations.

Common questions about the BFSG

When does the Accessibility Strengthening Act apply?

Since 28 June 2025. It implements EU Directive 2019/882 and covers products placed on the market from that date and services provided to consumers from that date. Transitional periods until 27 June 2030 exist only for devices that a service provider was already using before then, and for contracts concluded before the cut-off date (§ 38 BFSG). There is no transitional period for the website itself.

Does a website with a contact form have to be accessible?

As a rule, no. The BFSG covers websites through which a consumer contract is initiated or concluded (§ 2 Nr. 26 BFSG). According to the German Confederation of Skilled Crafts (Zentralverband des Deutschen Handwerks), a form for general enquiries usually serves only for first contact. It may be different if orders are accepted as binding through the form. Online appointment booking, on the other hand, does count; the guidelines of the Federal Ministry of Labour and Social Affairs name it explicitly.

What is a micro-enterprise under the BFSG?

A business with fewer than ten employees and an annual turnover of no more than €2 million or an annual balance sheet total of no more than €2 million; one of the two figures is enough (§ 2 Nr. 17 BFSG). Employees are counted in full-time positions; apprentices do not count. The ministry’s guidelines give an example: a hairdresser with four employees and €1 million turnover is exempt for its online booking, but no longer with ten employees.

Does the micro-enterprise exemption also apply to online shops?

For the shop as a service, yes; for certain goods, no. The exemption in § 3 Abs. 3 BFSG covers services only. Anyone who sells devices named in the law, such as laptops, smartphones, routers or e-book readers, must, as a retailer, check even with two employees that they bear the CE marking and are accompanied by the manufacturer’s documents (§ 11 BFSG). Clothing and food are not on the list.

What happens if the BFSG is breached?

The joint market surveillance authority of the federal states (MLBF) in Magdeburg is responsible. It first calls on the business to put things right within a deadline, then a second time with a warning, and can finally prohibit the offer (§ 29 BFSG). For a service that does not meet the requirements, § 37 BFSG provides for a fine of up to €100,000, and up to €10,000 where reporting and information duties are breached. Consumers and associations can apply for proceedings (§ 32 BFSG).

Which standard applies to accessible websites?

Neither the act nor the ordinance names a standard; they require the state of the art (§ 3 BFSGV). For websites, the Federal Agency for Accessibility (Bundesfachstelle Barrierefreiheit) points to EN 301 549, clause 9, and that in turn to the WCAG 2.1 success criteria at levels A and AA. That is 50 criteria, including a contrast of at least 4.5 : 1 for normal text and full operation by keyboard.

Is an automatic testing tool enough for the BFSG?

No. In one comparison, six widely used tools covered at most half of the WCAG criteria and found 14 to 38 per cent of the actual violations (Vigo et al. 2013). In tests with 32 blind users, only 50.4 per cent of the problems were covered by WCAG criteria at all (Power et al. 2012). Tools are a good start; keyboard operation and a run-through with a screen reader belong with them.

Are my answers stored or transmitted?

No. The check runs entirely in your browser; the page sends no answers to our server and stores nothing. Once you close the window, everything is gone. If you want to keep the result, print the report or save it as a PDF; “Copy result” puts it on the clipboard as text.

What the tool is based on

The guide values and notes above are based on these 16 scientific papers. Each one has been checked via its DOI or the publisher’s page (most recently on 13 November 2026): it exists, it has not been retracted, and its findings still hold for this question today.

  1. Vigo, M.; Brown, J.; Conway, V. (2013): Benchmarking web accessibility evaluation tools: measuring the harm of sole reliance on automated tests. Proceedings of the 10th International Cross-Disciplinary Conference on Web Accessibility (W4A 2013), pp. 1–10

    Finding Six widely used testing tools covered at most half of the WCAG 2.0 criteria at all and found 14 to 38 per cent of the actual violations; the more thorough ones reported more false alarms in return.

    In the tool Basis for the note in the result that an automatic test finds only part, and for the check items that are deliberately done by hand (keyboard, focus, cookie banner).

  2. Frazão, T.; Duarte, C. (2020): Comparing accessibility evaluation plug-ins. Proceedings of the 17th International Web for All Conference (W4A 2020), pp. 1–11

    Finding Eight browser extensions for accessibility testing on ten large websites: each covers only part of the criteria, and which part differs greatly. The authors advise using more than one tool and testing by hand.

    In the tool The section “What to do” recommends two tools with different test engines plus manual testing.

  3. Ismailova, R.; Inal, Y. (2022): Comparison of Online Accessibility Evaluation Tools: An Analysis of Tool Effectiveness. IEEE Access 10, pp. 58233–58239

    Finding Six online testing tools run on 41 government websites gave different results for the same page each time; only in combination do they add up to usable coverage.

    In the tool Confirms, with newer tools, why a single green test report is no proof.

  4. Brajnik, G.; Yesilada, Y.; Harper, S. (2011): The Expertise Effect on Web Accessibility Evaluation Methods. Human–Computer Interaction 26 (3), pp. 246–283

    Finding 19 experts and 57 non-experts evaluated the same pages. Without expertise, validity and reliability dropped; three experts found all the problems, while it took 14 non-experts.

    In the tool That is why the self-test is limited to checks that work without expertise, and says openly that it does not replace an expert assessment.

  5. Power, C.; Freire, A.; Petrie, H. et al. (2012): Guidelines are only half of the story: accessibility problems encountered by blind users on the web. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI 2012), pp. 433–442

    Finding 32 blind users encountered 1,383 problems on 16 websites. Only 50.4 per cent of them were covered by WCAG 2.0 criteria, and some techniques from the guidelines did not solve the problem.

    In the tool Supports the recommendation to try the page yourself with a screen reader and not just tick off criteria; hence the testing methods using VoiceOver, TalkBack and the Windows screen reader.

  6. Petrie, H.; Kheir, O. (2007): The relationship between accessibility and usability of websites. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI 2007), pp. 397–406

    Finding Blind and sighted test participants encountered problems on two websites that overlapped by only about 15 per cent. The ranking from the guidelines did not match the severity experienced.

    In the tool Explains why a website that seems fine to sighted staff can still have barriers; the weighting in the self-test is based on locking people out, not on level A or AA.

  7. Lazar, J.; Allen, A.; Kleinman, J. et al. (2007): What Frustrates Screen Reader Users on the Web: A Study of 100 Blind Users. International Journal of Human-Computer Interaction 22 (3), pp. 247–269

    Finding 100 blind users kept time diaries. Biggest annoyances: confusing page layout, unlabelled forms, missing image descriptions, misleading links and inaccessible PDFs. On average, 30.4 per cent of the time was lost.

    In the tool Forms and alternative texts are therefore weighted as “locks people out”; link texts and PDFs have check items of their own.

  8. Hanson, V. L.; Richards, J. T. (2013): Progress on Website Accessibility?. ACM Transactions on the Web 7 (1), pp. 1–30

    Finding Over 100 large and government websites in the USA and the UK, followed from 1999 to 2012: high error rates throughout, with improvements mainly where technology and construction methods changed.

    In the tool Background to the advice to fix barriers in the template rather than page by page: whatever is in the template is repeated everywhere.

  9. Yan, S.; Ramachandran, P. G. (2019): The Current Status of Accessibility in Mobile Apps. ACM Transactions on Accessible Computing 12 (1), pp. 1–31

    Finding Of 479 Android apps, 94.8 per cent had clear violations, mainly missing focus, missing descriptions, too weak contrast and controls that were too small. The testing tool found around 67 per cent of the problems.

    In the tool Neighbouring research on apps: the same types of error top the list as on the web, which is why focus, descriptions and contrast come first in the self-test.

  10. Hart, T. A.; Chaparro, B. S.; Halcomb, C. G. (2008): Evaluating websites for older adults: adherence to ‘senior-friendly’ guidelines and end-user performance. Behaviour & Information Technology 27 (3), pp. 191–199

    Finding 40 websites for older people, checked against senior-friendly guidelines, three of them tested with older users: the site that followed the guidelines most closely produced the most completed tasks, but not automatically more speed or satisfaction.

    In the tool Supports the advice to complement guidelines with a test by real users, and the check item on zooming.

  11. Schmutz, S.; Sonderegger, A.; Sauer, J. (2016): Implementing Recommendations From Web Accessibility Guidelines: Would They Also Provide Benefits to Nondisabled Users. Human Factors 58 (4), pp. 611–629

    Finding 61 people without disabilities each used one of three websites with different levels of accessibility. On the most accessible one they completed more tasks faster and rated usability, appearance and trustworthiness higher.

    In the tool Evidence for the note in the “not affected” result that an easier-to-use website is worthwhile even voluntarily.

  12. Schmutz, S.; Sonderegger, A.; Sauer, J. (2017): Implementing Recommendations From Web Accessibility Guidelines: A Comparative Study of Nondisabled Users and Users With Visual Impairments. Human Factors 59 (6), pp. 956–972

    Finding 55 people each with and without visual impairment: higher conformance with WCAG 2.0 improved performance and ratings equally in both groups.

    In the tool A second piece of evidence for the benefit beyond the obligation, this time with both groups in the same trial.

  13. Vollenwyder, B.; Petralito, S.; Iten, G. H. et al. (2023): How compliance with web accessibility standards shapes the experiences of users with and without disabilities. International Journal of Human-Computer Studies 170, Art. 102956

    Finding 131 people with and without visual impairment shopped in an online shop that did or did not meet WCAG AA. There were no measurable differences in usability and experience; in the free-text answers, both groups reported a better experience on the conforming shop.

    In the tool An honest assessment on the page: conformance with the standard is mandatory, but on its own does not guarantee a good shopping experience. Testing with people is part of that.

  14. Gernsbacher, M. A. (2015): Video Captions Benefit Everyone. Policy Insights from the Behavioral and Brain Sciences 2 (1), pp. 195–202

    Finding Overview of more than 100 empirical studies: captions improve comprehension, attention and memory, in children and adults, not only for deaf and hard-of-hearing people.

    In the tool Reasoning in the captions check item.

  15. Bargas-Avila, J. A.; Oberholzer, G.; Schmutz, P. et al. (2007): Usable error message presentation in the World Wide Web: Do not show errors right away. Interacting with Computers 19 (3), pp. 330–341

    Finding Two trials with 77 and 90 people: error messages worked best when they appeared at the relevant field after filling in. Immediate messages while typing were often ignored.

    In the tool Reasoning and recommendation in the error messages check item.

  16. Hall, R. H.; Hanna, P. (2004): The impact of web page text-background colour combinations on readability, retention, aesthetics and behavioural intention. Behaviour & Information Technology 23 (3), pp. 183–195

    Finding 136 people read web pages in four colour combinations: a higher contrast ratio generally gave better readability; the choice of colour also influenced how appealing the page seemed and whether there was an intention to buy.

    In the tool Reasoning in the contrast check item; the self-test gives the thresholds 4.5 : 1 and 3 : 1 from WCAG 1.4.3.

Standards, authorities and manufacturer data

Mon–Fri 8:00–17:00

We check whether your website meets the requirements using a keyboard and a screen reader. You get a list of the barriers and what fixing them costs. If a rebuild is planned, we plan accessibility in from the start.

+49 3744 4399760[email protected]

Mon–Fri 8:00–17:00 · Straße der Jugend 5, 08228 Rodewisch

Have your website checked
+49 3744 4399760Mon–Fri 8:00–17:00
CallCallback

PARTS LIST · SHEET 1/1

What this page is made of

Values measured at build time. With JavaScript, your browser measures them again live.

ITEMPARTTYPESIZE
01index.htmlDocument≈ 29 kB
02styles.cssStylesheet≈ 20 kB
03scripts.jsScript≈ 13 kB
04archivo-variable.woff2Font≈ 58 kB
05ibm-plex-mono.woff2Font≈ 30 kB
06favicon.svgImage< 1 kB

Advertising or analytics cookies: none. Without your consent, the page loads nothing from third-party servers apart from Cloudflare, our delivery service. Fonts, scripts and images are hosted by us. That is how we build your IT, too.

OUTSIDE OFFICE HOURS

Feel free to call anyway

Our regular office hours are Monday to Friday, 8:00–17:00.

In the evening and at weekends the number is forwarded to us, and often someone will pick up. We cannot promise that, though.

If nobody answers, leave a short message on the voicemail: your name, your number and what it is about. We will call back as soon as we can, on the next working day at the latest.

The call costs nothing. If someone has to start work straight away outside office hours, the surcharge in the price list applies.

Call now+49 3744 4399760

Prefer to write? Use the contact form or e-mail [email protected]. We will read it on the next working day.

DIAGNOSIS · AT-K2

What do you need help with?

↵ SELECTESC CLOSECALL: +49 3744 4399760

Custom settings

Which services may load?

Strictly necessary

The choice you make here and your settings for colour scheme and motion. These are kept as small entries on your device and are never sent to a server. They are needed for the site to work and therefore cannot be switched off.