R.M. Williams challenge century-old opinions with testing

For a brand that's been making boots for over 90 years, R. M. Williams has strong opinions about what it stands for. So does everyone in the head office. The challenge for Maija Frankovich, Digital Experience Manager, wasn't a lack of ideas — it was proving which ones were right.

Four years ago, R. M. Williams partnered with Deloitte and Optimizely to build an experimentation program from scratch. The goal wasn't a specific KPI. It was something harder: creating a culture where data, not the highest-paid person's opinion, drives decisions. For a heritage brand with passionate internal stakeholders, that meant using experiments to prove or disprove long-held beliefs about the brand, the customer, and what actually works online.

The program has since evolved from culture change into a full-funnel testing operation. The team tests everything from homepage messaging and campaign concepts to major product page redesigns — monitoring engagement, conversion, and completed orders across every experiment. When R. M. Williams launched its boot discovery project, a complete redesign of its product and listing pages, they used Optimizely to run a guided tour that brought customers along the journey — driving positive engagement and uplift in completed orders where previous big redesigns had tanked KPIs.

Now, with Opal AI helping with result summaries and idea generation, and conversational AI-powered search on the horizon, R. M. Williams is pushing the program further. But Maija's advice stays grounded: don't be afraid to fail, because even a losing test means you haven't spent all that money on dev for something that doesn't work.

Building a Culture of Experimentation at a Heritage Brand

0:00

Maija uses customer insights and data to drive the development roadmap at R. M. Williams — a proudly Australian-owned brand that's been manufacturing boots for over 90 years, with stores across Australia and the UK alongside its digital channel
Four years ago, R. M. Williams partnered with Deloitte to launch an experimentation program — starting not with a KPI, but with a research study into customer data that laid the foundation for questioning long-held internal beliefs about the brand
The early focus was culture over conversion — using experiments to prove or disprove assumptions and change mindsets within a business where strong, passionate opinions about what the brand means had traditionally driven decisions

Our main focus was really about creating a culture of experimentation and questioning beliefs that we had within the company. We had a lot of ideas of the brand and what the brand meant, and we were able to use the experimentation program to change some mindsets within the business, which was really exciting.

Maija Frankovich — R. M. Williams

From Brand-Led to Customer-Led, One Test at a Time

0:00

The experimentation program started without a concrete KPI — it was about culture first — but has since evolved into a focus on improving the customer experience, particularly how R. M. Williams translates the craft of its boots into the online journey
Every test forces the team to be more customer-focused rather than brand-focused — a significant shift for a heritage brand where everyone in the head office has strong, passionate opinions about what the brand means
The boot discovery project, a complete redesign of product and listing pages, used Optimizely to launch a guided tour that brought customers along the journey — a direct response to previous big redesigns that had tanked KPIs because customers weren't brought along for the change, resulting in positive engagement and uplift in completed orders

Full-Funnel Testing and the Value of a "Losing" Result

0:00

The team tests both big launches and small iterative changes — from homepage messaging and banner positioning to the broader personalization program, now exploring how Opal can finesse copywriting across the site
Every test monitors the full funnel, not just the primary metric — a recent homepage banner redesign showed increased engagement but lower downstream conversion, proving that a lift at one level can mask a problem further down
Maija's advice to retailers starting or scaling an experimentation program: don't be afraid to fail, because a losing result still means you haven't spent all that money on dev for something that doesn't work — you're learning, even when it looks like you're losing

Opal, Conversational Search, and What's Next for Boot Discovery

0:00

Opal is already helping with result summaries and idea generation for experiments — with the learning that providing more specific context produces higher-quality test ideas
The next frontier is using Opal to generate code for experiment variations, which could accelerate testing velocity significantly — Maija's dev team can build variations, but the opportunity to test more and faster is what excites her most
On the digital roadmap, R. M. Williams is progressing the boot discovery project further with AI capabilities on site — upgrading search and merchandising tools and introducing conversational search to help customers find the right boot for them

Opal has been really helpful in terms of the result summaries that it offers and also the idea generation as well. We have a dev team, but obviously it'd be great to be able to accelerate that and test more and faster.

Maija Frankovich — R. M. Williams

A Heritage Brand That Lets the Data Decide

Maija Frankovich's experimentation program at R. M. Williams didn't start with a revenue target or a conversion goal. It started with something harder — changing how a 90-year-old brand makes decisions.

Four years in, that culture shift has taken hold. The team now tests everything from homepage banners to full product page redesigns, monitoring the complete funnel and learning as much from the experiments that don't win as the ones that do. The boot discovery project proved that bringing customers along the journey — rather than surprising them with change — drives both engagement and completed orders. And when a homepage test showed more engagement but less conversion, that wasn't a failure. It was exactly the kind of insight that stops the business investing dev resources in the wrong direction.

Now, with Opal helping generate test ideas and summarise results, and conversational AI-powered search on the horizon, R.M. Williams is pushing the program into its next phase. But the foundation Maija built hasn't changed: use data to prove or disprove what the business believes, and let the customer — not the highest-paid person's opinion — have the final word.