The catalogue three agencies gave up on – and how we made it shine

Rebuilding the catalogue around who is buying

ServicesGoogle Ads · Microsoft Ads · Feed Management

DataFeedWatch was already in place when we took over, but the feed was doing half a job. We reworked colour mapping, shipping, GTINs and descriptions, taking disapprovals from 1,162, around 11% of the catalogue, down to 23.

Then we split the campaigns by end user, so home cooks and professional chefs stopped competing for the same budget. Shopping was rebuilt around buying intent, with low, medium and high intent separated and commercial products pulled out from the rest. On top of that we added product bucketing: custom labels that segment the catalogue by audience and by performance, so budget goes to the products that actually do well.

Feed rules in DataFeedWatch, where the attribute work was done.

Tracking needed rebuilding too. Enhanced conversions were not firing, so we wrote the code, briefed their developers and shipped it on the thank you page through Google Tag Manager. That unlocked the bidding strategies the account had been locked out of.

The same structure went onto Microsoft Ads, where we added Dynamic Search Ads as well. Microsoft skews toward business users, which makes it a strong B2B channel on this account. Performance Max launched a few weeks before we took the account on, and we started testing segmentation on it straight away.

Taking over an account that was spending without a structure

ServicesMeta Ads · Creative · Feed Management

Meta was live but unmanaged. We audited what was running, cut the campaigns doing no job, fixed the catalogue campaign taking most of the budget, then rebuilt the structure from scratch.

We started the rebuild with research: the market and audience, then their own best-performing brands and categories. That analysis set both the structure and the creative.

The top of the funnel runs on a broad, always-on Advantage+ campaign working from the full consumer catalogue, with no manual interest layering, so the algorithm has room to find buyers. Sitting next to it are category-focused sets for electrics and knives. Those ranges are high value and visually strong, and in a broad catalogue campaign they get drowned out by cheaper items, so they each get their own budget and impression share.

Meta ads in rotation: the Factory Seconds promotion, the electrics category ad, and the Le Creuset Bleu Riviera unboxing.

Then there are brand sets built around moments rather than always-on noise: Le Creuset for the Bleu Riviera launch and the limited-edition Petal collection at Mother’s Day, All-Clad for bakeware promotions, Wusthof around the Classic Ikon chef’s knife. Each one is excluded from the others so we are not paying twice for the same person.

At the bottom, one retargeting pool covers product viewers, add-to-cart and checkout starters, prioritised in that order, with catalogue ads showing people exactly what they looked at. Alongside it runs a brand video built around the things that actually make people trust this business: Canadian, family-owned since 1968, free delivery over $99, used by home cooks and restaurants alike, gift registry available.

Creative follows the brand’s own editorial plan, so the ads look like Williams Food Equipment. Part of it is produced in-house: statics built in Canva, video cut in CapCut. Formats are mixed on purpose – Reels-first video, dynamic catalogue carousels, collection ads – so Meta has enough variety to match a creative to the right person. Local campaigns for in-store sales and chef demonstrations run on their own budget and connect all of it back to the shop floor.

Restructuring 7,000 categories

ServicesTechnical SEO · On-Page Optimization · Content & Copywriting · Off-Page & Link Building

We opened on the technical layer. Product pages were missing JSON-LD, caused by parent-child data flow issues and a theme bug, which kept products out of Google’s SERP features. We removed the old Microdata, standardised the JSON-LD, validated it and had the pages re-indexed within 24 hours. We also found images and elements loading from a sandbox environment rather than the main domain, and fixed those with their development team.

Alongside that we rewrote meta titles and descriptions, built content around the highest-converting categories and brands, and set up blog content. Screaming Frog and Semrush crawl the site on a regular cycle so nothing new creeps back in, and the backlink profile is monitored as it shifts.

The bigger project, still running, is restructuring years of over-categorization into a structure no deeper than three levels, to clear the duplicate category and brand pages and stop crawl budget going to pages that get very little traffic.

Manual cleanup is not realistic at this scale, so we built tooling for it. Products are exported from the site and reorganized with AI so none are left orphaned. Keyword data comes through an SE Ranking connector to decide which categories to keep, merge or remove. We then check the proposed structure against both search demand and how easy it is to navigate.

The workflow runs on a set of custom Claude skills built in-house: one writes on-brand content with internal linking, one reads the live category structure and proposes its replacement, one handles SEO audits, and one takes brand pages end to end, writing the content and setting up the redirects from old slugs to new URLs so nothing 404s. BigCommerce has no bulk editing, so we wrote our own script: it pulls every category into a Google Sheet, we edit titles, descriptions and meta data at scale, and it pushes the changes back to the site.

Rebuilding the pages that do the converting

ServicesConversion Audit · UX & Design · Testing & Implementation

The bulk of the work sits on the product page, where we redesigned the above-the-fold with Williams Food Equipment’s development team. Every change was mocked up in Figma, built on established CRO research, and walked through with the reasoning behind it. Conversion rate improved, and the page is still being worked.

The home page has since had a full redesign: mockups, feedback sessions with the developers, now in implementation. It moved fast, helped by a dev team that ships.

We are also rebuilding the commercial category pages we send traffic to, bringing together product information that is currently spread across the site. These are the pages the paid campaigns land on, so the work feeds straight back into the other three channels.

The changes come from Clarity recordings, heatmaps, click and scroll maps, on-site surveys and heuristic UX analysis. On the analytics side we also keep an eye on the data itself, checking GA4 for tracking gaps and filtering out bot traffic, so the numbers everything else is judged on can be trusted.