This article first appeared on the carwow blog on Medium.
Gut feel and opinion go a long way, but it can also take you down the wrong path if you don’t listen to your audience. At carwow, we’re strongly opinionated but it doesn’t rule our decision-making. We use data and user feedback to inform our opinions and decisions about product development and we’re better off for it.
Where do the Hippos come in?
‘HiPPO’ is a term I first heard 5 years ago. It an acronym for ‘Highest PaidPerson’s Opinion’ and was used to describe how many companies make decisions about their product development and marketing. When faced with a decision, the choice was made based on the opinion of a senior figure, with an absence of data or a conscious decision to ignore it.
My view of a HiPPO is less about the seniority and pay grade of an individual and more about the mindset. It’s very easy to believe we know best, or to assume “I could easily be our target customer” but it can be a dangerous habit. I’ve been working on websites since the last millennium (1999?—?yes, I’m that old!) and I’ve lost track of the number of conversion optimisation tests where I’ve guessed the wrong winner. You just don’t know. You need more information.
So if guessing is not an option…?
If you can accept that you don’t know all the solutions to your business problems, then the answer is very simple. You listen.
There are cryptic clues, hidden messages, explicit complaints, enthusiastic endorsements all over your business and from a wide variety of sources. Sometimes you just have to shut up and listen for them.
It’s a trait that’s strong in those who contribute to developing the product at carwow.
So how do we listen?
The information you seek comes in many forms and so you need to listen in different ways. This list isn’t extensive but does highlight a few of our methods.
To see what is happening we use a variety of data reporting sources?—?primarily MixPanel, Google Analytics and our own internal customer data. We can see who is using our website, how they interact and convert, and identify bottlenecks that are hindering growth.
On a more granular level, we can review pages using clickmaps, heatmaps and (one of my favourites) video recordings of users interacting with specific functionality on the site.
Analytics is great at the ‘what’ part, but pretty lacking when it comes to understanding the ‘why’. Why are users bouncing from your homepage? Why are they dropping out on your signup form? Why are they unsubscribing from your beautifully crafted emails?
Listen. They’re telling you why.
User testing sessions
These are fun, try it. With user testing, you ask a participant to carry out tasks on your website, whilst you sit next to them and observe. They think out loud, telling you what they’re doing, why they’re doing it and how it makes them feel. All the while, you’re screaming inside your head:
“Click the big green button! It’s right there in front of you!!”
It’s an enlightening experience that really underlines the (potentially incorrect) assumptions you’ve made about how your product should function.
Don’t pass over this method thinking you don’t have the capability or experience to run these sessions. Instead, pick up a copy of Steve Krug’s book ‘Rocket Surgery Made Easy’. It’s a quick and immensely valuable read that will have you testing within the week.
Unlike user testing sessions which can be done in your office or in a testing lab, you can also learn a great deal by observing your customer in their own environment. carwow is a two-sided marketplace, matching car buyers with dealers, so we often visit the dealerships to see how our business customers incorporate our service into their working lives. You quickly spot the holes you can’t see from your desk.
Customer feedback loops
One of my favourite elements about the structure of carwow is that the Customer Service team are part of Product Development. If our goal is to build a product that enables the best car buying experience you’ll ever have, anyone needing to interact with Customer Service is evidence that we must continue to find ways to make it better. And if you’ve ever looked at the information handled by a Customer Service team then you’ll know it’s an embarrassment of riches for anyone looking to improve their service.
First places to look? Dive into your helpdesk enquiries, chat session data and your phone call transcripts. You’ll find problems, solutions, ideas and emotions.
When you look at analytics data, the most depressing moment is when a visitor leaves, task incomplete. Why? What put them off?
We’ll never know, is the natural assumption, they’ve gone. Well, true, but we can always ask them as they head out the door. Add an exit survey to your problematic page?—?a little slide-in single question survey box that appears as the user’s mouse moves perilously close to the browser menu bars?—?and ask them what’s wrong.
We did this on a signup form on carwow, asking:
What made you decide not to sign up today?
The results were illuminating, giving us some quick-fix issues to address and food for thought for some conversion tests to run. Handily, they’re also a great tool for capturing the voice of the customer, so you can use the same terminology they use in your web and email copy and not the usual jargon-laden marketing babble.
Insights as far as the eye can see. Now what?
I think you’ll agree, if you’ve looked at your analytics, spoken with customers and observed their actions, then you’re in a better, much more informed place to make decisions.
At carwow, all this information feeds into our design and development process. Our product managers, UX designers and developers assimilate this knowledge into our designs and code.
The proof of the pudding is in the eating, as they say, so we A/B test major changes with our audience, ensuring that the choices we made based on their feedback were the correct ones.
And if not, we go back to the information, we listen some more and we iterate.