Preaching conversion (rates)

Dan Croxen-John is chief executive of AWA digital, an agency that specialises in ‘Conversion Rate Optimisation’ for retail websites, and here he explains how AWA helped Lyle & Scott transform its online operation and see a 49 percent increase in revenue per visitor

Throughout its 140-year history, Lyle & Scott has built a global reputation for distinctive, beautifully designed, high-quality knitwear, combined with the brand’s solid authority and presence in the golfing sector. The company is committed to developing new and exciting products, and to making them available to customers across multiple channels. The Lyle & Scott website -www.lyleandscott. com – now retails in over 50 countries worldwide, making e-commerce one of the major components of the company’s sales structure.

Having an international, online retail offering is one thing, but optimising its conversion rate is another, and Lyle & Scott’s e-commerce team wanted to augment online sales by increasing revenue per customer. They had built a substantial customer database and were already mining the data, optimising it by creating new segments, but they felt there was room to improve conversion once those customers reached the website.

Lyle & Scott chose AWA digital for the conversion rate optimisation project because of our proven track record and understanding of the retail sector.

The objectives for AWA digital were as follows:

  • To convert more customers online
  • To increase average order value
  • To deliver on AWA’s guaranteed 20% online sales uplift

Furthermore, these objectives needed to be delivered while respecting the established Lyle & Scott brand guidelines.

Screen Shot 2014-07-21 at 15.58.21

The Solution 

AWA digital implemented its 5-step Conversion System. This is a proven methodology we have created that comprises five key stages:

1. Tool set up

2. Diagnostics

3. Triage

4. Optimisation plan

5. Creative execution.

As a full-service conversion optimisation and analytics company, AWA digital handled the project in its entirety, developing a complete test plan – including hypotheses, wireframes, graphic design, copy and web development. Very little input was required from Lyle & Scott throughout the process.

Stage 1 – Funnel analysis and tool set up 

Each time a customer attempts to make a purchase on the website there are a number of steps they need to take in order to complete the order.

Funnels helped AWA see these processes easily, by giving a visual representation of the conversion data between each step. AWA analysed 6.8m user sessions and segmented over 50 million data points. This gave us a deep understanding of visitor flow through the site, specifically looking at the difference in behaviour between people who bought and those who didn’t. The analysis allowed AWA digital to:

  • Determine what steps were causing customer confusion or trouble
  • Highlight any bugs, browser issues and other technical nuisances
  • Decide what research would be required to understand why visitors were dropping out of the sales process

AWAStage 2 – Diagnostic review of website 

We used a range of voice-of-customer tools – including interviews with store managers, online surveys and moderated usability testing. Exploring the psychology of a Lyle & Scott customer gave the team a better understanding of why they weren’t converting. As an established and globally recognised golf brand, the AWA team drilled down on specifics which included a survey of golf club members chosen to test drive the site. The results revealed clear differences between multi-buyers and one-time buyers.

As part of the deep-dive analysis, the team also visited Lyle & Scott’s London stores to observe customer behaviour and interview managers about perceived brand positioning. They needed to understand why customers preferred Lyle & Scott over competing brands.

Stage 3 – Identification of conversion killers 

The next step was to deploy our ‘prioritisation’ process (Triage), to identify the key conversion killers. These included a category page with too many choices, poor location of the search filters, a high checkout drop-off rate and too little emphasis given to order-builders.

Stage 4 – Testing strategy 

Our team prioritised the various sections of the website for the first round of optimisation. The evidence pointed to the category and basket pages being key optimisation opportunities, rather than the home page. Having gathered and compiled the information, the team made its recommendations. We then created an optimisation plan that was agreed and signed off by the Lyle & Scott e-commerce team.

Stage 5 – Split testing 

Before launching split-tests, the team got feedback from usability testers previously recruited in the diagnostic phase. This process enabled the team to gather necessary feedback on proposed wireframes and designs before launching the split-tests. Incoming traffic to the website was then distributed between the original site and the different variations without any of the visitors knowing they were part of an experiment.

The Results 

At the time of writing, AWA’s split-tests have shown an increase in revenue per visitor of over 48 percent. This came from just two split-tests, with the first delivering over 19 percent increase, and the second over 30 percent.

Return on investment: £19.55 of additional sales for every £1 spent with AWA digital. 

Lyle & Scott has extended its contract with AWA digital for a further round of customer research and optimisation.

www.awa-digital.com 41

Lyle and Scott