// WORK 002
I was brought in to write product pages. I left having redesigned how the company sells.
Full-stack UX redesign for a Toronto-based B2B industrial cleaning supplier.
A Toronto B2B supplier of industrial cleaning consumables was competing against broadline distributors many times its size, while hiding the three advantages that could beat them. I was hired to write Amazon product pages, and followed the content problem upstream into a UX one: the site they owned was a flat catalogue doing none of the selling. I ran a competitive audit and primary research with facilities managers and EHS officers, then rebuilt the experience around how B2B buyers actually purchase: self-serve, long before they contact a vendor. Four structural moves came out of it, including a new customer-facing reorder product.
Client withheld under NDA, visuals brand-stripped
Where it started
My first job for this client wasn’t UX. It was Amazon.
I was building their A+ content, a branded storefront that told buyers what the company actually stood for, and a set of product pages that did one job well: educate the person reading them. What does this wiper absorb. Where do you use it. Why this one and not the cheaper one next to it.
Somewhere in that work the real problem got loud. I was pouring brand story and product education into Amazon (a platform the company doesn’t own), while the website they do own did none of it. Their own site was a flat product grid. No context, no education, no reason to trust them over a distributor ten times their size.
So I took the instinct behind the A+ work (explain the product, communicate the brand, remove the reason to call sales) and moved it upstream, to the surface that actually decides whether this company wins or loses a buyer. That became a full-stack redesign of their digital experience.
The company, and the three things it kept hiding
A Toronto-based B2B supplier of industrial cleaning and safety consumables (wipers, absorbents, spill kits) competing in a market dominated by broadline distributors many times its size.
On paper it should lose. In practice it had three real advantages the giants couldn’t match:
Canadian
Headquartered and sourcing domestically, which, under Canada’s new Buy Canadian procurement policy, is worth real money to institutional buyers.
Sustainable, verifiably
An in-house textile-reuse programme that turns post-consumer cotton into industrial wipers, backed by audited numbers, not a marketing claim.
SMB-priced
Reachable for the small and mid-size buyers the premium distributors price out.
The site communicated none of the three. It behaved like an electronic catalogue in a market that had moved on from catalogues.
What the research actually found
I ran a competitive audit against the four distributors the redesign had to beat, and triangulated primary research with the people who actually buy this stuff: facilities managers and EHS officers in the Greater Toronto Area.
The finding that reframed the whole project: B2B buyers now complete most of their purchasing journey before they ever contact a vendor. They spend serious money self-serve, they switch suppliers the moment the digital experience fails them, and they consult more channels than they used to. The website isn’t a brochure the sales team hands out. The website is the sales team.
Measured against that, the old site failed at exactly the moments that decide a sale: it didn’t prove the company understood the buyer’s industry, it didn’t surface the certifications and safety documentation buyers vet on, and it treated a repeat order (the thing that drives the revenue) exactly like a first-time purchase.
I built three personas to hold the research steady: a facilities manager drowning in a weekly inventory spreadsheet, a healthcare compliance approver who won’t call for a safety sheet she should be able to download, and a public-sector procurement lead who needs provable Canadian eligibility on the page. Read apart they want different things. Read together they want the same thing: buy without talking to a salesperson, and see the evidence on the page instead of gating it behind one.
The reframe
The brief looked like a website redesign. The research turned it into a positioning exercise that happened to ship as a website.
The design hypothesis I committed to: if the site communicates the three moats credibly and prominently, curates the catalogue by buyer industry instead of by SKU type, and replaces the spreadsheet reorder cycle with a real tool, then the company converts more first-time buyers and keeps more of the repeat ones.
Four structural moves
Every one is tied to a research finding, not an aesthetic preference.
Industry-led navigation
Buyers said they default to vendors who speak their language. So industries got promoted to top-level navigation. A healthcare buyer never lands on a generic grid again. Nine industry pages, one click from home, each curating products and surfacing the compliance considerations that sector actually cares about.
Product pages that teach, not spec-dump
Lead with use-case framing and “where to use it,” industry tags, certifications above the fold, and one-click access to safety data sheets. Every question answered on the page is one fewer email, and the buyers who’d email instead just leave.
A dedicated Buy Canadian page
The procurement opportunity is invisible if it’s buried in a footer. So it became a standalone, indexable surface that cites the policy by name with a registration number, offers downloadable eligibility certificates a procurement officer can drop straight into a vetting file, and routes to a named institutional contact instead of a generic sales inbox.
ReStock: the part that isn’t a website at all
The biggest churn driver wasn’t the storefront. It was the reorder. Buyers track consumption in spreadsheets that are correct on Monday and wrong by Wednesday; stockouts mean expedited fees or a compliance miss. No vendor offered anything better. So I designed one: a customer-facing inventory dashboard (live stock, threshold alerts, auto-reorder drafts, consumption analytics) gated behind a verified business account. The public site is the acquisition surface; ReStock is the retention surface. Same customer record underneath.
That’s the through-line I’m proudest of: I didn’t just redesign a catalogue. I identified where the company was quietly losing customers and designed a new product to stop it.
How I know it works: without pretending it shipped
This was a concept redesign, so I’m not going to hand you a conversion lift I didn’t measure. Instead I benchmarked the design the way the strategy demanded: against the four competitors, on the heuristics that decide B2B purchases.
5/5
On the three B2B-specific heuristics the strategy concentrates on: industry relevance, compliance visibility, self-service depth. No competitor scores 5 on all three; the closest scores 4/4/3.
4–5
Across all ten Nielsen usability heuristics, matching or edging the strongest competitor on each, by design. The redesign consolidates proven patterns rather than inventing new ones.
2
Heuristics I scored a 4, not a 5 (power-user efficiency and depth of help documentation), written up honestly as priority directions for a v2, not quietly rounded up.
The part that taught me the most
Halfway through, I audited my own research against its original sources and found several statistics I’d been confidently citing were misattributed or imprecisely worded. I corrected them. It was uncomfortable and it was the right call: in a research-led project, one soft number undermines every decision that leans on it. Research-led design doesn’t mean “we did some interviews.” It means every visible choice traces back to evidence that survives being checked.
The other lesson was scoping. Redesigning five hero product pages instead of all 546, nine industry pages instead of fifteen, one geography of primary research: those weren’t failures of ambition. They were the decisions that made the work completable to a standard I could defend.
Why it’s in my portfolio: it’s the clearest proof of how I work. I followed a content problem upstream into a UX problem, questioned what the brief assumed, and let the logic of every decision show its work. Good UX isn’t decoration on a product. It’s the product.