Skip to main content

Work sample / DEMO-01

A journey through
Nordljus Demoaffär.

Compare two checkout barriers before and after code changes.

Fictional work sample. The shop and product are made up. No customer data, real purchases or payments are used.

The demo shop itself is in Swedish. Its deliberate barriers and before/after behaviour are preserved. This overview and the report are available in English.

One product, one cart, one delivery choice

Try the same journey
in both versions.

The cart is preset with one lantern. You do not need to enter any details. The before pages contain two deliberate barriers.

The product page is the same in both versions. The payment section only marks the boundary of the work.

Two specific code changes

What changes?

01 / Delivery options

Choose without a mouse

Before: The options look like buttons but respond only to clicks. A keyboard cannot select a delivery method.

After: Standard radio buttons let you select with Space and the arrow keys. If you continue without a choice, an error appears and focus moves to the group.

Read the delivery steps →

02 / Cart

Read on a narrow screen

Before: The cart keeps two columns. At 320 pixels wide, the checkout button falls outside the viewport.

After: Content stacks in one column when space is limited. Long product codes can wrap, and the button follows the items.

Read the cart steps →

The evidence

See the code changes
and test results.

The report describes the barriers, code changes and repeatable checks. It separates older observations from checks of this website.

Read the full report →

Testing and limitations

We use Codex to run scripted checks of specified keyboard interactions, focus, accessibility-tree data and layout, plus axe checks. This is not a screen-reader test.

Layout is checked at a viewport width of 320 CSS pixels and other specified narrow widths. These checks do not test browser zoom or text resizing. Checks must detect the deliberate before barriers. Results apply only to the checks run.

There are no measured customer results. Manual testing, screen-reader testing, testing with users and production testing have not been carried out.

Payment and identity checks are outside the scope. There is no such component here to test or change. This boundary says nothing about a real provider’s accessibility or legal responsibilities.

Have a similar barrier?

Tell us about it and who handles your development.

Email us about your shop