to find a perfectly-fitting bra online.
A traditional brick-and-mortar European lingerie company (specializing in lingerie for women with large cup sizes) hired McKinsey to help it build its online business from scratch. As a lead designer, my role was to create an outstanding customer experience - from online marketing, to best-in-class website, to simple purchasing and eventually - unpacking your new lingerie set. You can see the final project here.
As with all my projects, I structured the solution to this problem around the principles of Human-Centered Design (via iterative loops of research -> ideation -> implementation). Research was the first step, and in this project, probably due to being a man and designing something so feminine, I assumed a complete beginner’s mindset. The brand has already developed some “Personas” (based on marketing demographics, which is not ideal, but very common), and I was expecting a considerable stakeholder pushback for a lengthy research phase (the brand is over 100 years old and it quite set in its ways and vision). So I put forward a case for a 2-week research, presenting its value in terms of adding a “why” to the “who”. I’ve suggested that such information would be helpful not just to the website team, but also to marketing, branding and product design teams. Even the CEO got curious. The client was honest to admit that they didn’t know “the why” or the inner motivations why a certain Persona picked our brand. I got a green light for a 2-week research phase and a small budget. Over years I’ve learned that it’s much easier to get buy-in for projects that are presented as improvements to existing work, rather than something completely new (organizational infamous “not-invented-here-syndrome”).
Once the jungles or organizational politics were behind me, I could spend my time planning the actual work. Around that time, I was reading about the methodology of Jobs-to-be-Done and I was eager to try it. A research colleague and I have set up in-person interviews with 15 women (in UK, Germany and Belgium) with 3 main goals (formulated literally from instructions in “Competing Against Luck” by Clayton Christensen).
Check and enhance existing Personas
Identify functional, emotional and social motivations for buying lingerie
Identify demand-generating “push” and “pull” factors, as well as demand-reducing “inertias” and anxieties.
For the interviews, we prepared a semi-structured questionnaire that would give us some structure, yet be flexible enough to explore new and unexpected themes. I have gave my colleague a little lesson in in-depth interviewing techniques like probing, uncomfortable silence and others.
Those interviews produced a lot of mind-boggling and surprising insights. Here are some of them:
- For large-breasted women, comfortable fit was an absolute must. For that reason, women often stuck with brands that they knew and were very reluctant to change to a different brand. There was a lot of intertia and brand loyalty. This gave us insights that the most difficult part for us will be acquiring new customers who are not yet familiar with the brand.
- Motivations for purchasing a bra were highly personal. After years at the service of others (husband, children, careers) for many women in their 50s (target audience), an expensive bra was like a gift to themselves. “I deserved this” was a common sentiment.
- With a deep emotional motivator, feeling like they “deserved a little luxury”, we were able to start seeing competition in a different light. It was not other lingerie brands that we were competing with, but also other brands and services that made the women feel that way - spa treatments, bags and shoes - were all our competition now.
- Most women had serious reservations about buying a bra online, afraid that the fit would not be perfect. They wanted to try a bra before buying one. This changed our expectations about conversion rates for first-time site visitors. Instead, later on we decided to make the website convert women to signing up for a free personalized “in-store measuring”. That session would then become one of the highlights in their respective journeys, where women would be treated like queens, getting a fitting and familiarizing themselves with the brand down the road. Once we knew the woman’s size, we could use website more effectively for re-ordering (and not first-time purchases).
Common stakeholder thinking was “we need a website with a pay button”. However, armed with knowledge generated from the research phase (especially about loyalty and reservations to buy bras from unknown brands online), I organized an ideation workshop, trying to push stakeholders to think of the new offering as a journey that would combine physical store, the website, and the delivery experience. What came out of it was the beginning of a cross-channel brand and product strategy, populating product backlog with numerous short- and long-term ideas.
I wished to use the outputs of the workshop to create a “Journey MVP” - or to formulate our MVP deliverable as an experience, of which a website would be just one. Yet, due to internal ownership questions that this idea raised, the client decided to start with the website first.
There were a few “modules” that a modern eCommerce website has to get right:
- Outstanding landing page, communicating the best of what the brand stands for, to capture hearts and minds of first-time visitors
- Browsing experience that could be both - efficient and inspirational
- Content, as most modern marketing strategies blend content to enhance branding and help promote the products (indirectly) featured in that content
- Smooth checkout experience. Checking out has to be fast, easy and accomodate for users with various payment preferences (e.g. are we going to support ApplePay?)
- Supporting confirmation emails and feedback loops
- Optional: “My Profile” section, letting the user manage orders, see order history, edit sizes and preferences, etc.
On top of those, ours also had to include an elaborate “Fit Finder” - a tool that would allow women to find their own perfect size based on answers to a few questions. The tool was already in development by the client. It’s overall quality was good and it was backed by scientific papers and various data models, but we had to give it a little “boost”.
Numerous iterations were created for each section and tested every week on Thursdays. Remote usability tests were scheduled well ahead with Usertesting.com and later on - with its more affordable competitors.
To start exploring aesthetical options, I’ve set up another workshop inviting stakeholders to bring their favorite designs (digital or physical, even art), and then to present each design saying what they liked about it. From there, I’ve put together a list of things people liked. I later used that list to inform the general aesthetic of the site, which was centered around the creative tension of simplicity and sophistication and the idea of being “classy with a twist”. It’s main elements included:
- Abudnance of white space
- Sophisitcated typography
- Outstanding typography
- Transitions and animations
A lot of inspiration was drawn from the beautiful Aesop.com .
To help the front-end development team, I have worked with several developers to help convert the Sketch symbols into React.js components. Later on, we created a custom UI kit (similar to many others, e.g. Audi), and published it on an internal site. That way, future iterations happened much faster, and the workflow was optimized.
Overall, after a 4-month effort, the company was already selling its products online, converting higher % of its business into online channels every month. As always, it wasn’t always comfortable for everybody involved, but with the right communication, appropriate stakeholder management and time spent on doing design “right”, the team achieved a significant impact.
Thank you for reading! Here are other projects you might find interesting to learn about:
The Race of Gentlemen
McKinsey & Company
McKinsey & Company
McKinsey & Company
McKinsey & Company
The Race of Gentlemen
Copyright (c) 2021. Martin Ahe