Case Study: Shopping for clothes with confidence
If it doesn’t make me feel good, I won’t buy it.
UX RESEARCH CASE STUDY | FASHION AND CLOTHING | HYPER ISLAND
A recent survey of 12,000 people revealed that women choose their clothes based on how they make them feel. According to the study, commissioned by Comfort, most women decide what to wear based on whether the clothes make them feel confident or happy.
Our clothing choices have also been shown to help us perform better at work.
How easy is it for women to feel confident that the clothes they choose will meet their expectations?
This case study, which forms part of my UX research assessment for the UX Upskill programme at Hyper Island, will use qualitative and quantitative insights gathered from user surveys and interviews to answer the following questions:
What expectations do women have from their clothes ?
What are women’s priorities when shopping for clothes?
What key issues do women experience when clothes shopping?
How can we enable women to shop with greater confidence?
At the end of the article, I’ll bring together insights from the UX research findings to deliver a set of recommendations, which will outline the next steps required towards developing a feature to help women shop more confidently.
Article Index
- Interpreting the Brief
- Research Process
- Recommendations and Results
- Reflections and Conclusion
1. Interpreting The Brief
The challenge was to come up with recommendations for a feature to help deliver confidence when shopping.
Meet our persona, Anna:
Understanding User Needs
To create empathy with Anna, understand her life situation and her needs, I created a kick-off board to explore what she needed to shop for clothes with confidence.
Next, I collected some inspiration using a Pinterest board, thinking about where she liked to shop, styles and colours, where she drew her fashion inspiration from.
Feeling like I ‘knew’ the target user a little better, I began the research process with greater confidence that I could create hypotheses that were grounded in empathy and understanding of her needs.
2. Research Process
Having completed a workshop in which I identified Anna’s current and future needs, these were plotted onto a prioritisation matrix. The matrix helped me to work out what we knew about Anna already, and what we didn’t know, and to see more clearly what would be priorities as part of our research.
I selected the 3 most important of these to generate the research hypotheses, and a set of open questions for each hypothesis. These are shown below.
Screening Survey
In order to select participants for the interviews, I designed a screening survey using Google Forms. As we were time-limited, I also considered how I might maximise the benefit of this survey through obtaining some quantitative data. I then shared the survey through my network and also on social media.
User Interviews
From the respondents who volunteered to be interviewed, I selected 5 interviewees based on the following criteria:
Fully completed the survey questions
Identified as female
Aged between 30 and 55
Report that their body shape has changed in recent times
Mix of users who enjoy clothes shopping, and users who don’t.
I then carried out interviews remotely using Zoom. Each interview took between 30 and 45 minutes. During the call I asked the questions outlined below. Video calls were recorded with participant consent, and then transcribed using Otter.ai.
Card Sorting Activity
In order to obtain some additional quantitative data, I also designed a simple card sorting activity. I carried out some competitor research to discover what features already existed to help women shop for clothes more confidently.
Because I intentionally wanted to use a ‘traffic light’ system, I made sure to run this through a colour blindness simulator first as you can see on the video.
I asked participants to sort and then rate different shopping features based on whether they thought the feature would help them to shop for clothes with greater confidence. The video below explains how the activity worked:
3. Recommendations and Results
Key Recommendations
The recommendations to support the research goal of more confident shopping are:
- Create an AI Stylist and online moodboard feature, which would allow users to upload their favourite items at home to mix and match with stylist-recommended ones.
- Improve search and filter functions on the website/ app, to allow users to tailor their own experience more effectively.
- Improve product descriptions in order to support users to make informed decisions and reduce the need to order multiple different sizes and styles.
- Consider a customer loyalty programme if one does not already exist.
- Review current return/ change policy to ensure it is matching market leaders.
Discussion
The research showed that users have a clear idea of what colours and styles flatter their shape, and as a result often shop with specific items in mind.
However, despite being sure about what they want, many users will still order multiple items, in multiple colours and multiple sizes to try at home, which represents a significant cost to the business in terms of logistics, transportation costs, and returns processing.
By improving the information available to users about clothing products, and making it easier for them to find exactly what they’re looking for — plus at the same time offering them ways to explore new styles in a safer or guided way — we may be able to improve both user experience, but also potentially support the business to reduce some of their costs.
There may also be some value in reviewing key brand elements to ensure that brand messaging is aligned to the idea of how great clothes make you feel, rather than how they make you look. This is worth exploring further.
What next steps are required?
- Carry out a journey mapping exercise to understand pain points throughout the user journey.
- Carry out further user research to gain further insights into how users search and filter items of clothing when shopping.
- Carry out competitor research to explore and analyse any existing AI Stylist and moodboard features and to evaluate their returns policies.
- Evaluate AI stylist and moodboard features through further user research and testing, to choose a priority design option.
- Work together with website and marketing teams to obtain additional quantitative data (search data, CRM data) to support decision-making.
What are the insights and principles underpinning the next steps?
The insights and principles shown below were driven by the research findings, which are discussed in more detail in the next stage.
What are the research findings that produced the insights?
The quantitative and qualitative research data shown below underpins the insights in the table above. I would like to discuss each hypothesis in turn.
Hypothesis 1: We believe that women do not know how to dress to flatter a changing body shape
This hypothesis was disproved. The qualitative data suggests that most users know what items of clothing are flattering to their figure.
This was supported by quantitative data from the survey. 26 of 29 (90%) considered it important that their clothes were flattering. Although 72% of respondents reported that their body shape had changed recently, only 10% (3 of 29) did not know what clothes flattered their shape.
Hypothesis 2: We believe that choosing the right clothes makes women feel good about themselves.
This hypothesis was proved. The qualitative data suggests that when users wear clothes that are ‘right’ for them — whether that’s based on comfort, colour, or style — they feel confident and happy.
This sits alongside quantitative data from the survey, which showed that 93% (27 from 29) agreed that their clothes can make them appear more confident. However, despite this almost universal agreement that confidence was important, only 72% of respondents (21 from 29) considered this to be one of their top 5 considerations when buying clothes.
This was quite an interesting result — perhaps users expect to feel confident as a consequence of their clothing choices, but don’t actively seek out confidence when choosing their clothes?
Hypothesis 3: We believe that women want clothes shopping to be easier.
This hypothesis was neither proved nor disproved. The qualitative data suggests that users experience a challenge when trying to find flattering clothes. Issues mentioned vary widely, from variations in sizing between brands, to poor product descriptions online.
Users also reported turning to online reviews about the item for product information that they feel is ‘missing’ in the main description. This was an interesting discovery, because the majority said that they don’t feel the need for their clothing choices to be validated or approved by anybody. However, this objective feedback from other purchasers is still clearly relevant and important to users.
However, the quantitative results were a little more mixed. Even though every user interviewed reported challenges with shopping, ease and speed were not ranked as top 5 considerations. In fact, only 48% surveyed (14 of 29) ranked ‘clothes shopping should be easy’ as one of their top 5 most important factors, and even fewer (7%) ranked ‘clothes shopping should be fast’ as an important factor.
It would be useful to conduct some more user research and journey mapping to unpick exactly what their pain points are, and at which stage in their journey they occur.
Other Findings
The survey and interviews also highlighted some other important insights that have relevance to any project aiming to improve user experience when shopping for clothes.
Priority Features
During the interview, users were asked to sort real-life features into order of priority for helping them to shop more confidently. The results show that an AI stylist and the ability to upload own items ranked as the most popular features, with a Brand Matcher feature ranking as next most popular. Two of the five interviewees suggested that it would be most helpful to them to combine features — for example, an AI stylist with an online moodboard.
Brand loyalty and user experience
When users find a brand that delivers a great user experience, they tend to be loyal to that brand. As the quotes show, this isn’t exclusively linked to clothes that suit them, but other customer service-related factors including range of products, quality of materials, relatable models of a similar age or body shape, returns policy, and delivery times.
I went back and carried out another iteration of the hypotheses, as I felt that this was an interesting observation that was a consistent finding between all users.
Hypothesis 4 focuses on brand loyalty, and examines what factors are important to users when choosing which brand to shop with.
Quantitative data suggests that although 62% of users tend to stick with certain brands that they know ‘suit’ them, 86% (25 of 29) are willing to give new brands a try, especially if they can offer something that can give them a more individual look.
This is an important insight for any brand looking to retain customers in the longer term.
4. Conclusion and Reflections
Reflections
This was an extremely interesting project to work on. I resonated with the target persona’s dilemma, and so it was so interesting to talk to real users and to understand their different perspectives. I felt honoured that these smart and thoughtful women had given up their time to talk to me.
I think that I have a new appreciation of how important clear research goals are for any research project. I uncovered so many interesting insights that it would have been very easy to get side-tracked on any number of them. However, keeping focused is very important if the research is to have any meaningful impact for the business.
Conclusion
From clothes shopping lovers to those who absolutely hate the process, users still seemed to want the same things from their experience — to shop with confidence.
It is great news for the sector that users are willing to embrace technology in order to receive a better user experience. The challenge now is to understand what difficulties users face when shopping for clothes, and how we can use technology to deliver the experience users would like to see.
