To Top

Flud. A case study in user data blindness.

If you are like me there are a handful of apps on your phone or tablet that are an integral part of your daily routine. Every morning my alarm clock goes off, I roll over to turn on the TV and grab my iPad. My first ritual is to check my email to make sure nothing caught on fire overnight and then I launch the news aggregator app Flud. I love Flud because my time is always limited so it lets me quickly pour through content from over 30 different self curated sites. Flud launched a redesign of the app last Thursday where they made a number of large changes to the interface and functionality but the reaction from me and the user base at large has been horrible causing their four star iTunes App Store rating to go crashing down to two stars. In an act of damage control the founder took to social media to explain their new direction and why they made the decision that lead them to the new design. I love Flud and would love to see it return to its former glory so I thought I would a little time to break down what happened and what we all can learn from Flud who seems to have been blinded by user data to the point where it is destroying what they were trying to create.

The problem centers starts with the previous version of the app the content from each site was laid out in a series of long pages. This meant that to see all the article from one site you wold scroll vertically but to move from one site to another you would have to swipe horizontally (you can see it in action in this video). It was this horizontal swiping that which lead them to their new direction as they cited the fact that 10 months of user data showed them that users did not want to swipe more than 7 or 8 times through the app. The new interface now completely removes the horizontal swiping to access to the content from individual sites in favor of a single page that mixes all the content from all the sources together. Let’s look at the real problems and how I think they should have handled their user data and new design.

Problem 1 – Control issues

I think the root of the problem everyone has with the new design has little to do with the actual design but has everything to do with control issues. The previous version of the app let me choose the sites where the content came from, let me control the order in which they appeared in and let me flip through them to decide which ones I wanted to read. The new version of the app takes all that control and customization away from the user and instead collapses everything down to one content stream that is controlled by the app. This is a huge user experience design mistake because they mis-read the data and over reacted to it.

Apps like Flud are just interactive tools and you build a tool you have to understand that there will be different types of users who want to accomplish different things and the design has to account for that. You need to have one path through the app for task focused users are seeking to accomplish tasks quickly through quick and simple interactions. These task focused users  have very definite behavioral limits and little patience for anything that distracts them from their desired task. They want control in the form of being able to customize the interface to make things more efficient. You need to have another path through the app for content grazing users who want to take their time and look around. These content grazers are much more open to new experiences, to experimentation and distraction.  They want control in the form of being able to add or remove functionality or content to the app so they can experiment and see how they like it. Both task focused and content grazing user are invested in the app and want some form of control over it.

I think that Flud looked at the data showing that there was a behavioral limit to the current interface and let it blind them to other larger behaviors of their users. In this case fundamentally changing the functionality so it took control away from those users after they have grown use to it and forcing them into a predetermined experience caused very bad reactions.

Problem 2 – Content consumption isn’t one-dimensional

The other problem that I think went unrecognized is that the way a person consumes content and the type of content they want to consume isn’t always the same. I use Flud in the morning to read through what was going in the worlds of technology, social media, mobile and digital design so when I walked into work I would be sure I was up to date on the latest developments and stories that would apply to my job. But when I got home at night I would use Flud to read through content about personal interests like industrial design, fashion, video games, interior design and movies. This means that for me as a single user I would use the same app for two very different purposes at two very different times but a behavior like that would never be reflected in just the user data. This goes back to my first point about control but now with the new design its impossible for me to replicate this behavior since all the content is now all mashed together. You can’t assume that information consumption is one-dimensional and all user will act exactly the same way all the time. I will say it again, Flud is a tool and no successful tool can be one-dimensional.

My two cents on using user data

There is an art and a science to how to react to user data. If you are talking about something straight forward and transactional where you have data that you will sell 20% more products with a certain verbiage on a button or fewer steps in the check out process then the path forward is clear. But if you are talking about data from large experiences or interactive tools like web sites, mobile apps or interface designs then the user data is only the starting point. Use it to understand literal behaviors like click-through or where there are behavior limits – like in this case where users wouldn’t swipe more than 7 or 8 horizontal times to see content. But then you have to be able to use art and science to read between the lines starting with the user data but add in an understanding of the psychology of your users then create the vision of how you want to evolve the experience to accomplish the goal of the re-design.

If you keep this type of information in vacuum without understanding user behavior and a clear concept then you get over reactions, things fall out of balance and you ultimately get very bad user reactions. This lack of balance also happens with more than just user data. Just look at the redesign of the Gawker platform last year where they let their need for increased online advertising dollars lead their re-design which killed the experience for the user and lead to their site traffic dropping from 5 million views a day to under 500k. For any digital experience to be success the final result has to be a negotiated blend of brand, business and customer goals. You have to look at each project and determine how those three needs to be weighted to achieve the desired end result and no single goal can completely take over the process.  If it does then you get what we see with Flud or Gawker if information is misunderstood or if one of the goals dominates so completely that it gets in the way of the customer goals and creates a bad user experience.

You must be logged in to post a comment Login

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

More in Creative direction