The brief

The brief for this project was the following:

"Navigating ASOS's large product mix can bring challenges. We are therefore looking to personalise the onsite experience to help our customers find the products that are right for them."

The deliverable required was a feature for the mobile app to help encourage discoverability of products in the ASOS mobile app.

With this as my starting off point, I began my research.

User interviews

The first step in the process was to write open-ended user interview questions which I could ask participants. The questions I came up with were the following:

  • What do you like about shopping online vs in stores?

  • What makes you shop at ASOS?

  • What kinds of things are you looking for when shopping at asos?

  • How could shopping on ASOS feel more personal?

  • What kinds of difficulties do you have finding items when online shopping?

  • What kinds of things are you thinking about when buying items for an event?

I carried out user interviews with five particpants. I then synthesised the results. I found four major groups:

  • People who didn't know about existing features, and would have benefited from their personlisation

  • Ideas for personlisation (participants wanted items suggested to them)

  • The fact that there are too many options in the app, which are unorganised

  • The fact that almost every participant mentioned shopping for events.

During the interviews, I asked some follow up questions which ended up being repeated in all interviews:

  • What types of events are there? And themes for these?

  • What kinds of weddings are there?

I ended up synthesising these responses into new categories:

Types of events

Types of weddings specifically

Ultimately, three things stood out from my user interviews:

  • Customers decide to shop when they need to attend an event.

  • Customers have too much choice so can't find what they need.

  • Customers want the items they are shown to be tailored to them

User persona

Based on my user research, I created the following persona:

Competitor review

I spent some time reviewing potential competitors to the functionality I was working on. I researched both traiditional competitors in the clothing space, as well as technolocy companies who have solved the problem of simplifying user choice for huge amounts of options.

One interesting experience to review was netflix. Netflix has a range of highly specific categories of their products, which were initially created by humans but then went on to be generated, using machine learning.


I created some sketches of possible solutions to the brief. I focussed on ways to suggest clothing for events as that was shown to be a key concern in my user interviews and persona.

Adding items that go together for an event while checking out.

An upsell feature in the existing bag.

Adding specific suggestions to search.

Creating a board which suggests items as you create it.

A stylist feature where user can get suggestions based on specific events.

I carried out guerrilla testing of these sketches on some ASOS customers, and got the most positive reviews for the last sketch, a stylist feature. This is the feature I therefore chose to take further in the design process.


The idea I selected for further fleshing to present as a solution to the initial brief was "Your Event Stylist".

This is a feature that would predominantly live on the homepage of the app, giving it huge visibilty to users (hopefully to avoid the issue of findability for previous customisation/suggestion features in the app).

The designs below show the entire user journey for the Event Stylist feature. At each level, the specificity of the event can be drilled down to suggest a list of items which are relevant for a very specific event. Initially, the categories can be human curated, but eventually (like the netflix model) we can use machine generated categories to make it relevant to the millions of ASOS customers around the world.

This design also shows on the Event Stylist feature could appear in other relevant touchpoints in the app. In this example, a user has created a board with clothing for a wedding. We can unobtrusively prompt the user to try the Event Style to find other relevant items.

This works well as the user is already trying to find items like this, a similar pattern would be Google search displaying adverts for items relevant to what a user is already searching for.

Data to measure

As part of this project I thought about potential data which would be measured to judge whether it had been a success or not. This includes:

  • Usage of the event stylist feature (measurements of opening the initial flow, which stage of the flow they reach, how many items are added to bag/boards from the event stylist suggestions, which types of items are more likely to be added, which types of categories are more popular, whether the items are converted into sales).

  • Usage of the search functionality for related terms (i.e. if searches for 'boho summer wedding outfits' increase or decrease after the introduction of this feature).

Measures of success

Measures of success for this project could be:

  • Sales through the event stylist feature

  • Items added to boards through the event stylist feature

Next steps

  • Further user interviews to determine categories, card sorting to decide categories.

  • Eventually the category data can be generated by machine learning, no need for human input (and categories can be very specific).