Skip to content

Accessibility Reachability Test

Reachability audit — instantiates Target-Complete Mapping Design

Checks that each required target can actually be reached and used in practice by the people it is meant to serve — not just that a route exists on paper — and flags targets reachable only inequitably.

Accessibility Reachability Test verifies that each required target is reachable and usable in practice by the population it is meant to serve — closing the gap between "a route exists" and "a real user can traverse it." Its defining move is testing effective, equitable access: a witness that exists on paper but is unusable for part of the target population — too costly, too complex, blocked by a barrier — counts as a gap, and the test explicitly flags targets that are reachable only for some. It converts nominal coverage into experienced coverage, probing usability and fairness rather than the abstract existence of a path.

Example

A government service claims every required task in its benefits portal is "reachable" — every page is linked and loads. An accessibility reachability test drives each required user journey the way a real constituent would, including with a screen reader and keyboard alone, against WCAG success criteria. It finds that the "upload proof of income" task — nominally reachable, and green on the link-level report — cannot be completed with a screen reader, because the file-upload control carries no accessible label. The path exists; for blind users the target is unreachable. That inequitable-access gap goes on the register even though link-level coverage read 100%. (WCAG is a real standard, referenced here for context; the specific defect is illustrative.)

How it works

What distinguishes it is that it is empirical and population-aware, not a check that a path exists. For each target it simulates the actual intended user — including the least-served — attempting to reach and use the witness, and checks whether the task genuinely completes, at what cost, and against what barriers. Unequal reachability across the population is itself recorded as a gap, so "works for the median user" is never mistaken for "serves everyone required."

Tuning parameters

  • Population coverage — test the median user versus the least-served, edge, and assistive-technology users. Testing the hardest cases is what surfaces equity gaps a happy-path pass hides.
  • Criteria strictness — "the page loads" versus "the task completes for everyone." Stricter criteria, such as full WCAG conformance, catch more and cost more.
  • Assistance level — unaided versus with reasonable accommodation. This dial sets where the line of "reachable" is drawn.
  • Automated versus human — automated scans catch a share of failures; human and assistive-technology testing catch the rest (see below).

When it helps, and when it misleads

Its strength is that it converts "a path exists" into "the target is actually served, for everyone required" — catching equity gaps that aggregate counts and abstract reachability models both miss. Its weakness is that real-world access is expensive to test exhaustively, so sampling the wrong users misses the very group that is excluded. The classic misuse is running only automated checks, or only the happy path, and declaring the target accessible — the overlay-badge trap, where a green automated scan hides journeys real users cannot complete. The discipline that guards against it is to test with the actual least-served population and their assistive tools, not just an automated pass, and to treat any inequitable reach as a gap rather than a rounding error. The pattern that access built for the excluded tends to help everyone is the well-known curb-cut effect.[1]

How it implements the components

Accessibility Reachability Test realizes the real-world-access side of the archetype — the components an empirical usability audit can fill:

  • target_access_and_equity_check — its core: does every required target reach the whole intended population, fairly, and not just a convenient majority?
  • feasibility_and_capacity_check — whether the witness is practically usable — reachable at acceptable cost and effort — for those who actually need it, not merely present.

It does not compute whether a path exists in the abstract model — that is Graph Reachability Analysis — nor stress a source's throughput at scale, which is Source Capacity Load Test; this test measures experienced, equitable reach.

  • Instantiates: Target-Complete Mapping Design — the test is what confirms coverage that holds for the real population, not just on paper.
  • Consumes: Graph Reachability Analysis — a path must exist in the model before its real-world usability is worth testing.
  • Sibling mechanisms: Graph Reachability Analysis · Source Capacity Load Test · Service-Area Gap Analysis · Witness Validation Test · Coverage Counterexample Search

References

[1] The curb-cut effect — accommodations designed for a specific excluded group (curb cuts for wheelchair users) routinely end up benefiting many others (people with strollers, carts, luggage). It is why testing for the least-served target is not a niche concern: the barrier it exposes usually degrades the experience of the majority too. ↩