Recommendations

Background

To reach our corporate goals, we needed to increase average order size on the web. We also discovered that users were unaware at the breadth of product we have to offer, this was noted organically through unrelated user tests.

Key performance indicators (KPIs)

Increase Average Order Value (AOV): Recommendations have the potential to increase the average amount spent by users in a single order. A rise in AOV signifies that the recommended items are effectively motivating users to make higher-value purchases.

My Role

UX Researcher
UX Designer
UI Designer

Problem Statements

Howdens

Steve - Builder

John - Builder

As Howdens I want the opportunities to up-sell, cross-sell by showing relevant best-seller products or missing items to logged-in builders and builders to add to lists when logged in

John - Builder

Steve - Builder

John - Builder

As a user, I want to see related products to support my main purchase

Howdens

Steve - Builder

As Howdens I want to provide product recommendations that are relevant to the page that Howdens.com users are viewing on the website

Steve - Builder

As a user I want to see the amount of stock so I can collect from my selected depot

Prototype

Hypothesis

We believed our category-based suggestions approach was more relevant and less disruptive to users when compared to traditional single product suggestions. For example, when a user adds a door to their basket, traditionally you may recommend a single handle which is a guess whether this is the users preferred style. By suggesting a category which is then linked to an on-page execution you give the user more power to choose more tailored product. We also think it will be beneficial to suggest quantity matching based on the number of linkable products the user has in their basket.

View Axure RP Prototype

I think it’s very useful! (…) you get all the bits that you need in one go, and you’ve not had to leave the page.