This rather excellent if rude quote comes from a conversation with Phillip Attenborough of AmberLight fame. And it got us thinking about how we approach UX, strategy, design and projects in general, especially when it comes to startups and grow-ups.
Let us start by unpicking this quote:
"We are going to tell you your baby is ugly. And you are going to thank us for it."
Could be translated as:
We are going to look at the evidence, and might tell you how people think differently about your product than you do.
We'll explain how to match their needs rather than your own.
On the surface this is the same as any designer, or UX designer might say. We all start by looking at the user groups. We create personas and we imagine what they may need. We match the user journeys, the marketing and experience to those personas. And it is true that most of the time this approach is quick, easy and pays off immensely.
For example, within the work we did for the NHS in Gloucestershire, we initially sat down with the client and within 10 minutes we came up with 10 personas for who would be using the website. From this we created user journeys for each and created a highly successful strategy and website. Many years later this strategy is still in place and those very same personas are being used throughout the organisation. User-centric design was an alien concept to the NHS trust, but now this way of thinking influences not all aspects, but certainly more than it ever did before.
This is a good example where the clients already knew the audience and the groups. We gave them a simple push in the direction of thinking about users’ needs first. The NHS is notorious as an organisation that has evolved around services and locations, not the end users. The ugly baby in this case is the fact that the services offered bear little relation to how people get ill and search for help.
However sometimes this quick approach does not work, and a more evidence based user-centric view needs to happen. This is generally when there is a new product or proposition which is untested in a market, or the market is unknown.
At this point we move from desk-based research – which if you are an experienced user-centric thinker works most of the time – to evidence-based UX. I haven’t got stats on how often this happens, but I am tempted to invoke the 80/20 rule (See diagrams to live your life by). We need to get primary evidence from our target users, and there are several methods to do this.
- Analytics is the best, most efficient way of seeing user behaviour. Cold hard stats rarely lie. But interrogating and asking the right questions of the data is a skill to be learned.
- A/B tests are also a great evidence-based way to show how users react and respond. There are two ways to make this happen: the proper way where users are randomly served different options, or when this is not possible, trying different things a month at a time.
However UX design is of the most importance before a product or service is ever released, so how do you gather stats and tests without anything being built?
- A user questionnaire is a good place to start, following up with some targeted face-to-face recorded interviews. These are doubly useful as they give you user quotes about your product even if it isn’t live yet, create some real buzz and a willing set of first customers.
- Another great way to efficiently gather feedback is to show someone a walkthrough or boards, and judge their reaction. Preferably by standing behind them. You judge when they hesitate, are confused or express joy.
- A prototype-based sprint, as outlined in the wonderful book of the same title is again a really efficient way to test the hardest aspects of ‘product verses your target users’. This process should take five days but I have conducted one day and even four hour cut-down versions. Fail fast is always the motto on these sprints. The prototype here might be a poster with a strapline, a simple mockup lashed together with PowerPoint. The aim is to test something simple, not spend ages making something perfect. On this I highly recommend the book, or perhaps the short version.
- There is one method for getting user feedback that in my experience has never delivered efficient results. This is the focus group. There are companies that specialise in this, whereby a group will sit around a table and discuss aspects of a product. Not only is this expensive, but the group is guided by the group, a bit like the movement on a wicca board.
The evidence-based approach also works when there is disagreement in the team about direction. And most of all it is when the founder remains unconvinced of change. And this is where the ugly baby quote comes in. Telling that founder that their vision is not quite right is an almost impossible task, especially when that user feels they have succeeded as far as raising money, perhaps gaining publicity. To them this success enforces their vision. But most founders do not represent their users, and the ego necessary to be a great founder does not generally translate to empathy for their users. In such cases the desk research and logic of personas may well need to be augmented by some hardcore evidence-based UI.
It is also useful to bring in an expert to guide the process. Cold hard evidence is the way to be sure you are on the right track, but also a UX expert will guide the process and diffuse egos, passion and closeness to the subject.
When I worked at prop tech start-up HiP, we required an outside expert – one Stuart Eggleton, also of Amber Light – to help us. We were far too close to the project and committed to a commercial solution. Stuart had the evidence and confidence to convince all the founders that a consumer solution, though in many ways harder, would bring better rewards and an instant marketplace. Seeing this process from the other side has helped my journey and strategic output.
The moral of this short article?
Don’t be scared to get in outside help, especially when it comes to strategy and UX. It’s hard to empathise with your potential customers all of the time; it’s hard to stay focused. Evidence-based design is even harder when you are skewing questions and answers to your own idea.
Most of all I would say continually test everything, and do not be upset with the answers.
Oh, your baby. They are beautiful. *As is the one in the top picture.
This post was originally posted to LinkedIn.