Altronics - e-commerce UX and IT audit
Customers asked me at the counter what the website should have told them. Nobody asked for an audit, so I wrote one: 26 findings, triaged by fix effort, with a working prototype beside the report.
The brief
I work as a Key Holder at an Australian retail electronics business. After the new online store went live, customers kept coming to the counter with questions the site should have answered - which store had the part, whether the return window applied, what a product actually was.
Internally the concern existed but only as a sentence: the new store feels harder to use. A sentence like that cannot be prioritised, funded or argued with. It needed to become specific.
My role
I initiated this myself alongside my Key Holder role. It was unpaid work: I produced the audit and prototype and sent them to management. Implementation, and any measured improvement, were outside this piece of work.
What I did
I reviewed the site from both sides at once: what the interface actually did, and what customers standing in front of me were trying to find. I covered navigation, homepage behaviour, interaction states, content structure, product pages and search visibility. Every finding states the problem, where it appears, and what I recommend doing about it.
Then I grouped the findings by fix effort rather than by page. Grouping by page produces a list nobody can act on; grouping by effort lets management ship the cheap half without a meeting.
The output
26 numbered findings - 5 high, 12 medium, 9 low - sorted into six workstreams: quick CSS and copy fixes (10), interaction and usability (7), content and structure (7), SEO and discoverability (2), plus 2 open ideas for discussion and 2 items I considered and deliberately set aside as intentional.
The substance included homepage “buttons” that were only images, which costs both accessibility and search visibility from one decision; buttons with no interaction states; a hero carousel image rendering blank on some machines; return-window text that contradicted itself across comparable products; and part numbers with no copy-to-clipboard - a finding you only get from standing on the shop floor watching people retype them.
Alongside the report I built a working prototype of the proposed experience, covering the key journeys, structure and interactions end to end, so the recommendations could be walked through as an interaction rather than imagined from prose.
What happened next
The report and prototype went to management on 19 August 2026. No response has been recorded, and nothing from it has been implemented as far as I know.
One reflection
The column I would keep in every audit I write from now on is “considered and set aside”. Two findings looked like problems and turned out to be deliberate, so I said so. A report that only contains the things you were right about reads as a sales pitch, and the reader can tell.