Usability.ie runs five core engagements: accessibility audit, heuristic / expert UX audit, usability testing, user research and conversion-rate optimisation (CRO). Most people who contact us don't arrive already knowing which one they need — they arrive with a symptom: weak enquiries, an argument at a stakeholder meeting, a compliance deadline, a redesign nobody's confident about. This page starts from those symptoms rather than from the methods, then compares the methods directly once you've narrowed it down.
There is no universal order to run them in. Each method answers a different kind of question and produces a different kind of evidence, and the right starting point depends on the decision you're trying to make and the gap in evidence you're trying to close — not on a fixed sequence that's supposed to apply to every project regardless of what's actually going on.
The five methods, in plain terms
Accessibility audit
A scoped evaluation of a website, app or digital service against an agreed WCAG version and level, combining automated tooling with the manual checks and assistive technologies stated in the proposal. It produces a findings register mapping each issue to the applicable criterion, with severity, evidence and remediation notes. It's an evaluation of a defined sample, not a certification or a legal guarantee that an entire service complies with a piece of legislation.
Heuristic / expert UX audit
An experienced auditor walks the site or product as a target user would and compares what they find against an established set of usability principles (Nielsen's ten heuristics). It's the fastest and cheapest of the five methods, and it's expert judgement — informed and structured, but not the same as watching a real user attempt a real task.
Usability testing
Moderated sessions in which recruited target users attempt realistic tasks on a live service or prototype, while a moderator observes what they understand, attempt and get stuck on. Where a heuristic audit shows what an expert thinks, usability testing shows what a set of real users actually did — evidence bounded by the participants, tasks and context of that particular study.
User research
Broader discovery work — stakeholder interviews, customer interviews, surveys, persona work, journey mapping — used to understand who the users are, what they need and why a current approach isn't working. It's typically the starting point at the beginning of a redesign or when a product isn't converting and nobody yet knows why, rather than an evaluation of a specific interface that already exists.
Conversion-rate optimisation (CRO)
A data-led engagement for sites that already have traffic but weak conversion. It works from analytics (Matomo or GA4 access is required), identifies friction points in the funnel, proposes specific changes, and tests them where the traffic volume supports it. It's the method most tied to ongoing measurement rather than a single up-front report.
A symptom-based decision framework
If more than one symptom below sounds familiar, say so on the brief call — the scoping conversation exists to recommend one route rather than a bundle of every possible activity.
Traffic but nobody's converting
Start with CRO if you already have enough traffic and analytics access to build a baseline funnel and test changes. If you're not sure why visitors are dropping off, a heuristic audit focused on the primary conversion path — messaging, trust signals, forms, mobile layout — is usually the cheaper first step, with CRO or usability testing following once you know what to fix or test.
About to spend money on a redesign
User research at the start, so the redesign brief is based on what users actually need rather than assumption. Pair it with a heuristic audit of the current site to capture what's already known to be broken — that combination produces a defensible brief: what must change, what can wait, and what risk each decision addresses.
Compliance pressure just became urgent
Go straight to an accessibility audit. It has a specific compliance evidence trail — findings mapped to an agreed WCAG version and level — that a general UX review doesn't provide. Accessibility findings can overlap with usability findings, but the audit itself is the method with a defensible paper trail for procurement or regulatory questions.
The team disagrees about what users need
Usability testing can replace an abstract argument with observed task evidence. The participant groups, number of sessions and rounds should follow the decision and risk; the report should state what that sample can and cannot support.
Just launched and want to validate it
Usability testing is the direct route — post-launch validation is one of its standard use cases, especially where analytics show drop-off but not the cause. If you also want a fast sanity check of obvious friction before booking participants, a short heuristic audit first means testing sessions aren't spent on issues an expert could have flagged in a day.
How the methods compare
| Method | Evidence it produces | What sets the timetable | What it can't tell you |
|---|---|---|---|
| Accessibility audit | Findings mapped to WCAG success criteria within an agreed scope and sample | Pages, states, technologies and evidence required | Whether the whole service is legally compliant — a sampled evaluation isn't a certification |
| Heuristic audit | Expert inspection against stated heuristics, prioritised and severity-scored | Journeys and interface breadth | How real users actually behave — it's expert judgement, not observed behaviour |
| Usability testing | Observed task behaviour and the agreed session evidence | Recruitment, rounds, tasks and analysis depth | Whether a redesign will convert better — findings are bounded by the participants and tasks tested |
| User research | Interview or survey themes and the agreed decision artefacts | Participant access, methods and analysis depth | Whether an existing interface is usable — discovery research may not evaluate a specific design |
| CRO | Funnel analysis, prioritised hypotheses and test results where volume supports testing | Baseline data, traffic and test cadence | Why users struggle, in their own words — that gap is usually closed with usability testing or research |
The written proposal should turn these dependencies into an actual timetable before work starts.
None of the five needs to run alone. A heuristic audit and an accessibility audit pair well on rebuild projects, since one covers general usability and the other a specific compliance evidence trail. Usability testing is often stronger once an audit has already cleared out the obvious friction, so sessions aren't spent on issues an expert could have flagged in a day. And user research at the front of a project tends to make everything downstream — the audit, the test plan, the CRO hypotheses — sharper, because the team is working from an understanding of users rather than assumption. The brief call exists to work out which of these combinations, if any, fits your situation, rather than defaulting to a bundle of every service at once.
Read next
- Heuristic audit vs usability testing: which one first? — a closer look at sequencing those two methods specifically
- 9 questions to ask a usability testing supplier in Ireland — what to check before you commission any of these methods
- All UX services — full pricing bands and engagement details for all five
Still not sure which one fits?
Use the brief call to describe the symptom, not the method. We'll tell you which of the five engagements fits and at what price band — a written scope follows with deliverables, timeline and fee.