Most people who think about accessibility think of a small group: blind people using a screen reader, wheelchair users, sign language. In an online shop, a large share of the people affected sit somewhere else. They are customers over 65 who are well informed and ready to buy, but who forgive small print, pale grey, tiny buttons and password prompts less readily than younger people. Many of them would not describe themselves as disabled, use no special assistive technology and still abandon the purchase when the shop slows them down. Using figures from January and July 2026, this article shows how large this clientele is, translates the typical limitations of older age into testable WCAG 2.2 criteria and ranks them by where they hold up the purchase most. How an accessible online shop is built as a whole is described elsewhere. This article is about the customers for whom it pays off most clearly.
Key takeaways
- 74 per cent of people over 65 use the internet, up from 48 per cent five years ago (Bitkom, 2026). Of these internet users, 76 per cent shop online (Bitkom, 2026).
- People over 65 rate their own digital skills at only a school grade of 3.2, and those over 80 at 3.7 (Bitkom, 2026). Every unclear spot in the shop meets this uncertainty.
- Disabilities mostly arise later in life: 34 per cent of people with severe disabilities are 75 or older, which is 2.7 million people (Destatis, 2026).
- Test zoom first: text must be resizable to 200 per cent (W3C, WCAG 2.2, SC 1.4.4), and at a width of 320 CSS pixels no horizontal scrolling may be needed (W3C, WCAG 2.2, SC 1.4.10).
- Quantity buttons, close icons and filter checkboxes need a target area of at least 24 by 24 CSS pixels (W3C, WCAG 2.2, SC 2.5.8). For an unsteady hand, this area decides whether the tap lands.
- Sign-in without a memory test: pasting and password managers must work (W3C, WCAG 2.2, SC 3.3.8). Where help and contact options are offered on several pages, they sit in the same place there (W3C, WCAG 2.2, SC 3.2.6).
How large the clientele over 65 online is
In September and October 2025, Bitkom surveyed a total of 1,004 people aged 65 and over by telephone and published the results in January 2026 (Bitkom, 2026). The result corrects a common picture: 74 per cent of people over 65 now use the internet, compared with only 48 per cent five years ago (Bitkom, 2026). Within the age group, the share drops markedly. Among 65 to 69 year olds, 98 per cent are online, among 70 to 74 year olds 85 per cent, among 75 to 79 year olds 64 per cent and among those over 80 just under half, at 49 per cent (Bitkom, 2026). Someone who is 68 today and orders online as a matter of course will usually want to keep doing so at 78, only with weaker eyes and a less steady hand.
People who are online also shop there. 76 per cent of internet users over 65 shop online, and 80 per cent do their banking online (Bitkom, 2026). This places the age group not at the edge of the target audience but at its centre. For a shop selling household goods, garden supplies, books, travel, mobility aids or groceries, it can make up a significant part of the regular customer base. Losing these people in the basket does not mean losing a fringe group, but customers who would come back.
The second figure from the same survey matters more for design than the first. Respondents rate their own digital skills at only a school grade of 3.2 overall; in a Bitkom study from mid-2025, 16 to 64 year olds gave themselves an average of 2.8 (Bitkom, 2026). The German school scale runs from 1, the best grade, to 6. Self-assessment continues to fall with age: from 2.6 among 65 to 69 year olds through 3.1 and 3.5 to 3.7 among those over 80 (Bitkom, 2026). For the shop, this means these customers are willing but unsure. They try, hit an unclear spot and are then more likely to doubt themselves than the shop. An abandoned purchase caused by uncertainty usually shows up in no complaint, only as an abandoned basket.
Help is explicitly wanted
Why disability is largely an age issue
At the end of 2025, a good 7.8 million people with severe disabilities lived in Germany, which is 9.4 per cent of the population (Destatis, 2026). The Federal Statistical Office precedes this with a sentence that matters more for shop design than the total: disabilities rarely exist from birth and mostly arise only at an advanced age (Destatis, 2026). 34 per cent of people with severe disabilities, or 2.7 million, were 75 or older; a further 45 per cent belonged to the 55 to 74 age group (Destatis, 2026). Only 3 per cent of disabilities were congenital or occurred in the first year of life, while a good 91 per cent were caused by illness (Destatis, 2026).
These figures only cover people with an officially recognised severe disability. Many age-related limitations lie below this threshold: declining near vision, reduced perception of contrast, a hand that is no longer quite steady, hearing that gives way against background noise, a working memory that no longer retains a customer number in passing. For hearing there is a reliable order of magnitude: according to the World Health Organization, more than a quarter of people over 60 are affected by what the WHO classifies as disabling hearing loss (WHO, 2026).
For the shop, this has a practical consequence. These customers usually do not use a screen reader or a braille display. They zoom the page with a keyboard shortcut or two fingers, set a larger system font on the tablet, hold their glasses closer to the screen and type more slowly. The criteria that matter for this are part of the scope of any WCAG audit at level AA anyway, but are often worked through only after the screen reader findings. For customers over 65, this order should be reversed, not out of consideration, but because this is where the revenue depends.
No senior mode as a way out
From limitation to testable criterion
Age-related limitations do not have their own chapter in WCAG, but they map largely onto success criteria that can be measured in an audit. The following table maps the four typical areas of vision, hearing, fine motor skills and memory, plus uncertainty in the process, to the criteria on which a shop can fail or pass, each with the test step according to WCAG 2.2 (W3C). All criteria listed belong to level A or AA. Criteria 2.5.8, 3.2.6 and 3.3.8 are among the nine criteria that WCAG 2.2 added to WCAG 2.1 (W3C, 2023), the others were already part of WCAG 2.1.
| Limitation with age | What happens in the shop | Criterion according to WCAG 2.2 | Test step |
|---|---|---|---|
| Near vision declines | Prices and product details are too small, zooming breaks the layout | 1.4.4 Resize Text (AA) | Enlarge text to 200 per cent, nothing may be cut off |
| Strong zoom needed | The page scrolls in two directions, the basket button slides out of view | 1.4.10 Reflow (AA) | Shop at a width of 320 CSS pixels without horizontal scrolling |
| Contrast is perceived more weakly | Light grey notes, struck-through prices and placeholders disappear | 1.4.3 Contrast (AA), 1.4.11 Non-text Contrast (AA) | Measure text at least 4.5:1, controls at least 3:1 |
| Hand trembles or is stiff | Quantity buttons and close icons are missed | 2.5.8 Target Size (AA) | Measure click areas at no less than 24 by 24 CSS pixels |
| Hearing declines | The product video explains handling only in speech | 1.2.2 Captions (A) | Check captions for every prerecorded video with sound |
| Working memory declines | Password forgotten, field hint gone while typing, session expired | 3.3.8 Authentication (AA), 3.3.2 Labels (A), 2.2.1 Timing (A) | Sign-in with pasting and password manager, persistent labels, extension before expiry |
| Uncertainty in the process | The order is sent without a review step, help is somewhere else on every page | 3.3.4 Error Prevention (AA), 3.2.6 Consistent Help (A) | Order summary with correction option, contact route in a fixed place |
Two additional points belong with this table. First, the rows are connected: setting text to 200 per cent also reduces the effective window width, and that is exactly when reflow errors appear that are invisible at normal size. Second, shops rarely fail on a single criterion. In a test by Aktion Mensch, only a third of the shops examined were accessible, and only 20 of 65 tested sites could be operated by keyboard alone (Aktion Mensch, 2025). Working through the findings for older customers therefore removes barriers for many other groups at the same time.
Zoom without detours: SC 1.4.4 and 1.4.10
Criterion 1.4.4 requires that text can be resized without assistive technology up to 200 per cent without loss of content or functionality (W3C, WCAG 2.2, SC 1.4.4). In a shop this rarely fails in the product description, but at the edges: prices in fixed boxes, variant buttons with a fixed height, truncated menu items and notes on delivery time and shipping costs that have to fit on one line and overflow their frame at double text size. There are two ways to enlarge a page: browser zoom via keyboard shortcut or a larger default font in the operating system. A shop should cope with both.
Criterion 1.4.10 goes one step further. Content must be presentable at a width of 320 CSS pixels without scrolling in two directions (W3C, WCAG 2.2, SC 1.4.10). This corresponds to a desktop window 1280 CSS pixels wide at 400 per cent zoom (W3C, Understanding SC 1.4.10). Anyone shopping at this zoom level essentially sees the mobile view on their laptop. Responsive shops already have that view, so the rework is often smaller than feared. The errors lie in elements that escape the mobile view: wide size charts, fixed headers that take up half the screen at high zoom, consent banners without a reachable close button and basket drawers whose checkout button sticks below the visible area. Which font sizes, line spacing and typefaces support readability is covered in our article on accessible typography and readability.
- Set browser zoom step by step to 200 and 400 per cent and run through the purchase from the search field to the order confirmation, in line with SC 1.4.4 and 1.4.10
- Set only the browser font size to the largest level and check whether prices, variants and delivery notes remain readable
- Measure headers, basket bars and chat buttons at high zoom: how much space remains for the actual content?
- Check whether consent banners and dialogs keep a reachable close or confirm button at high zoom
- Make sure the viewport meta element does not block pinch zoom on the phone
- Check size and measurement charts for horizontal scrolling and allow it only where the two-dimensional layout is needed
/* Text follows the person's setting, not fixed pixels */
body { font-size: 1.125rem; line-height: 1.5; }
/* Price and delivery notes may wrap (SC 1.4.4) */
.price, .delivery-note { white-space: normal; overflow-wrap: anywhere; }
/* No fixed height for controls with text */
.variant, .button { min-height: 2.75rem; height: auto; padding: 0.5rem 1rem; }
/* Quantity buttons: AA requires 24 x 24 CSS pixels (SC 2.5.8), here with headroom */
.quantity button { min-width: 44px; min-height: 44px; }
/* Header only sticks while there is enough height (SC 1.4.10) */
@media (min-height: 37.5em) {
.site-header { position: sticky; top: 0; }
}
/* In the HTML head: do not block pinch zoom */
/* <meta name="viewport" content="width=device-width, initial-scale=1"> */Contrast and text: what holds up as sight declines
As eyesight declines, light grey on white is the first to disappear. Criterion 1.4.3 requires a contrast ratio of at least 4.5:1 for normal text (W3C, WCAG 2.2, SC 1.4.3). For input field borders, checkboxes and unlabelled icons, at least 3:1 against the adjacent colour applies (W3C, WCAG 2.2, SC 1.4.11). How widespread the problem is shows in the annual analysis of the most visited home pages: WebAIM found low-contrast text on 83.9 per cent of them (WebAIM Million, 2026).
In the shop, the typical findings sit where the purchase decision is made: unit prices and struck-through prices in light grey, delivery times in pale green, placeholder text as the only label of input fields, buttons that look disabled but are in fact active, and return notes in small, light footer text. With older customers, it may also be that they use the screen at reduced brightness, against the light by a window or on an older device. A contrast that barely passes on the design monitor is not enough there. What to consider when measuring, and why the threshold is a floor rather than a target, is covered in our article on colour contrast and accessibility.
Hitting instead of missing: target size under SC 2.5.8
Fine motor skills decline with age, through arthritis, tremor or simply slower movement. On tablets and smartphones, the finger then hits the neighbouring element. WCAG 2.2 sets a minimum size for this: targets for pointer input must be at least 24 by 24 CSS pixels or have enough spacing to neighbouring targets (W3C, WCAG 2.2, SC 2.5.8). The enhanced level AAA specifies 44 by 44 CSS pixels (W3C, WCAG 2.2, SC 2.5.5). For customers over 65, it pays to take the higher value as the benchmark for purchase-critical elements, even though the lower one is sufficient for conformance.
In shops, the critical targets sit in recurring places: the plus and minus buttons of the quantity selector, the close icon of dialogs and consent banners, the checkboxes in filter lists, colour swatches for variants, arrows in product galleries and the bin icon in the basket, placed right next to the quantity selector. Tapping the bin instead of the plus is an annoyance for younger people. For unsure users, it can be the reason to give up the purchase. How quantity selection, variant selection and the gallery become operable on the product page is shown in our article on accessible product detail pages with variants, gallery and basket.
Label, help, catch errors
Forms are where uncertainty weighs most. Criterion 3.3.2 requires labels or instructions whenever content requires user input (W3C, WCAG 2.2, SC 3.3.2). In practice, this means every field has a visible label that stays in place while typing. Placeholder text in the field disappears with the first character, and anyone who is no longer sure at the third field whether the house number belongs here or in the next field has to delete everything to check. Format requirements for date of birth, phone number or voucher code sit before the field, not only in the error message. How labels, required field marking and error messages work together is described in our article on accessible forms and validation.
Criterion 3.2.6 was added with WCAG 2.2 and addresses exactly the need described in the Bitkom survey. If a website offers help, such as contact details, a contact mechanism, a self-help option like frequently asked questions or an automated contact mechanism, and this help is repeated on several pages, it must appear there in the same order relative to the other content (W3C, WCAG 2.2, SC 3.2.6). The criterion does not require help to exist. It requires that existing help does not move from page to page. For older customers this makes a considerable difference: anyone who found the phone number at the top right of the product page will look for it in the same place during checkout.
For people with declining hearing, a phone number alone is not a sufficient contact route. A contact form, an email address or a call-back option next to it closes this gap. The same applies to product videos that explain assembly, care or handling only in speech: for prerecorded videos with sound, WCAG requires captions (W3C, WCAG 2.2, SC 1.2.2). What matters for captions and transcripts is covered in our article on accessible media with captions and transcripts.
Criterion 3.3.4 applies to pages that cause legal commitments or financial transactions, which means every order (W3C, WCAG 2.2, SC 3.3.4). At least one of three safeguards must apply: the submission can be reversed, it is checked for errors and the person can correct them, or before submission there is a summary in which everything can be reviewed, confirmed and corrected. An order summary with the option to make changes is the usual solution in a shop, and for customers who are unsure of their entries it often does more than an extra explanation. How the entire checkout is built without barriers is shown in our article on the accessible checkout. How an order can then be withdrawn just as easily is described in the article on the accessible withdrawal and cancellation button.
Signing in without a memory test: SC 3.3.8
A customer account is convenient for regular customers, but can become a hurdle for older people: a password with special characters set months ago, a puzzle against bots and a confirmation code that expires after a short time. WCAG 2.2 introduced criterion 3.3.8 for this. A cognitive function test, such as remembering a password or solving a puzzle, may not be required in any step of the authentication process unless the step offers an alternative or a mechanism to assist (W3C, WCAG 2.2, SC 3.3.8). As examples of such a mechanism, the guideline explicitly names password managers that fill in the password, and copy and paste so that nothing has to be retyped (W3C, WCAG 2.2, SC 3.3.8).
In the shop this means: the password field blocks neither pasting nor autocomplete, sign-in via a one-time link to the email address is available as an equivalent alternative, and guest checkout remains possible without detours. Puzzles protecting against automated requests are replaced by methods that set the person no task. Time limits come on top. If a session expires during checkout, the person must be warned beforehand and be able to extend the time limit with a simple action, with at least 20 seconds to do so, unless the limit can be turned off or adjusted (W3C, WCAG 2.2, SC 2.2.1). More on sign-in methods without puzzles is in our article on accessible authentication and CAPTCHA alternatives.
SC 3.3.2 Labels
Every field has a visible label that stays in place while typing; format requirements sit before the field (W3C, WCAG 2.2, SC 3.3.2).
SC 3.2.6 Consistent help
Phone number, contact route and frequently asked questions appear in the same order on all pages of the purchase journey (W3C, WCAG 2.2, SC 3.2.6).
SC 3.3.4 Error prevention
Before submission, an order summary shows all details, and each can still be changed there (W3C, WCAG 2.2, SC 3.3.4).
SC 3.3.8 Authentication
No memorised password as the only hurdle: pasting, password managers and one-time links are allowed (W3C, WCAG 2.2, SC 3.3.8).
SC 2.2.1 Timing
Before a session expires, a warning appears, and the time limit can be extended with a simple action (W3C, WCAG 2.2, SC 2.2.1).
SC 1.2.2 Captions
Product videos with spoken instructions get captions so that declining hearing can still follow along (W3C, WCAG 2.2, SC 1.2.2).
Where older customers abandon the purchase: an order
A full audit report contains many findings. For customers over 65, the order of remediation can be derived from two questions: how close is the barrier to the checkout button, and can the person work around it on their own? Text that is too small in the product description can be offset with zoom. A sign-in step that demands a forgotten password can hardly be worked around without outside help. This results in an order that serves as a starting point and should be adapted to each shop.
| Order | Barrier | Criterion | Reason |
|---|---|---|---|
| first | Sign-in requires a memorised password or a puzzle | 3.3.8, 2.2.1 | blocks the purchase, often with no way out without outside help |
| next | Checkout button or required field cannot be reached when zoomed | 1.4.10, 1.4.4 | last step before revenue, abandonment without feedback |
| next | Quantity selector, bin icon and close icons are too small | 2.5.8 | mis-taps change the basket unnoticed |
| next | Field label only as placeholder, no order summary | 3.3.2, 3.3.4 | uncertainty about one's own entries leads to abandonment before submission |
| next | Help and contact change position | 3.2.6 | anyone who looks for help and does not find it is more likely to give up |
| last | Grey text on prices, delivery times and notes | 1.4.3, 1.4.11 | makes the purchase decision harder, but can partly be offset with zoom |
This order does not replace a full audit of all level A and AA criteria. It answers a different question: where does a shop start when time and budget are limited and older customers account for a large part of revenue? Which findings can be fixed quickly and which need a rebuild is set out in our article from audit report to remediation plan.
How we audit a shop for this clientele
An automated scan finds some of these findings, such as missing labels or insufficient contrast. Whether the checkout button remains reachable at 400 per cent zoom (W3C, Understanding SC 1.4.10), whether the quantity selector is large enough to hit and whether sign-in allows pasting and password managers only shows when every step of the purchase journey is checked in every view. Our WCAG audit therefore focuses on zoom and operability and assesses each step of a purchase individually, in line with the level A and AA criteria:
- Search and navigation at 200 and 400 per cent zoom in line with SC 1.4.4 and 1.4.10, in the desktop and smartphone view
- Product page with variants, quantity selector, gallery and video: target sizes measured, contrast on prices and notes measured, captions checked
- Basket and checkout: labels, format requirements, error messages and an order summary with correction option
- Customer account and sign-in: pasting, password manager, one-time link, guest checkout and time limits
- Help and contact: position on all pages of the purchase journey and contact routes besides the phone
- Result: findings per step with criterion, location and an order based on the effect on purchase completion
Expert review and tools find many findings, but not all. How to involve affected people meaningfully in the audit is described in our article on usability testing with disabled participants. For customers over 65, we recommend inviting test participants aged 70 and over and aged 80 and over, because self-assessment of digital skills continues to fall with age (Bitkom, 2026).
Legally, the situation is clearly defined. The German Accessibility Strengthening Act (BFSG) applies to certain services provided to consumers after 28 June 2025, including e-commerce services (BFSG, Section 1). Infringements can be punished with a fine of up to one hundred thousand euros (BFSG, Section 37). Which requirements apply in detail and which exemptions exist for micro-enterprises is summarised on our page on the BFSG requirements. For many shops, however, the stronger reason for implementation lies not in the fine but in the customers: three in four people over 65 are online, and most of them shop there (Bitkom, 2026).
A customer who puts on her glasses and zooms the page has already decided to buy. Whether she completes the purchase is then up to the shop.
The effort is often manageable because the building blocks already exist: a mobile view that holds up at high zoom, a stylesheet in which target sizes and contrast can be raised centrally, and a sign-in process in which pasting and one-time links only need to be enabled. Anyone who tackles these points in the order described is not building a special shop for older people, but a shop that grows with the age of its customers. How we go about this is described on our page on the accessible online shop.
Sources and studies
Related Articles
Accessible Product Pages: Variants, Gallery, Add to Cart
A flaw in the product template appears on every item page. How to make variant choice, gallery, price line and add-to-cart feedback accessible.
Accessible Withdrawal and Cancellation Buttons for Shops
Since 19 June 2026, § 356a BGB requires a withdrawal function. How it and the cancellation button work with keyboard, screen reader and zoom.
Accessible Product Search and Filters in Online Shops
Building search, autocomplete, facet filters and results lists accessibly: the combobox pattern, WCAG 2.2 and a practical test routine for shop teams.