Personalization is a huge topic in design right now. Big data, IoT, and technological autonomy are all buzz topics right now, and it feels super cool, daunting and tiresome, all at the same time. The talented souls at BKWLD reveal what they have learnt from their experiences
If you put the bullshit aside, what it really comes down to is that users are expecting, and brands are (sometimes) delivering better experiences.
We have a growing kit of tools and knowledge at our disposal, and user-centered design has been demonstrated to be good for the bottom line. A few years ago, CLIF bar approached us about designing a new flagship website to house their various brands, catering to various demographics. We had an idea that seemed a little crazy at the time – what if when I come to the site, it’s smart enough to just show me the content that I’m interested in? If I’m a runner, show me running products, and other content about running. If I’m a mom, show me stuff for kids. Even crazier, how do we do that without any creepy third party services that have been tracking years of customer data.
For a brand like CLIF, it was a powerful tool. It would allow users to find what they’re looking for, but also allow the brand to showcase related products and content, in a smart way. “Oh, you’re a runner? Did you know we make energy gels? And recovery bars? And here are some training tips for achieving a PR.” More time on site, greater brand awareness, increased household penetration.
We had to think realistically, from a technology place, a content place, and a user experience place. It took some time scratching our heads, time spent outside the office with real users, and a lot of time on whiteboards, but eventually we figured it out. We learned a lot, and have been able to apply what we’ve learned to other projects. It’s not uncommon for the sites we work on now to have some sort of “smart” content recommendation aspect to them.
Customized web experiences are a growing trend, and it makes sense. After working through a few of these projects over the past couple of years, here are some of the things we’ve learned, and some of the things that are important to keep in mind.
1) Customized experiences should help the user, not control them.
For every article that extols the virtues of Netflix’s recommendation algorithm, I hear 10 real people bitching that Netflix’s recommendations are awful. Most of the sites we make are funded by marketing dollars, so there obviously needs to be a balance of brand messaging and “real” content, but don’t let it stand in the way of what brought users to your door in the first place. On that note…
2) Let people go where they want.
Coming to a page that’s pre-filtered in an intelligent way can be a great time-saver and help your users find what they’re looking for faster. Awesome. But don’t put so much faith in your smart filter that it can’t be easily undone. Always have an easy way to “show me everything”.
3) Be honest. Do your users stand to benefit from this?
See number one. This is supposed to help the user. Creating a customized web experience isn’t easy, and there are a lot of pitfalls. How are real users interacting with your content now, and would the experience benefit if it were served up in an intelligent way?
4) Be honest. Does your content support this?
Do you have a variety of content that appeals to the various niches of your demographic? That isn’t easily sortable at a glance? Again, be honest. Effective content creation and
planning takes a lot of work, and shouldn’t be taken lightly. That said, brands like Red Bull and GoPro have shown that there is a huge appetite for branded content.
5) Being smart doesn’t mean being creepy.
There are two big cultural movements happening right now. On one side, we have the age of big data, where every digital footprint is cataloged. On the other hand, people are
clamoring for their privacy. What’s a *well-meaning* brand to do? Look at the points of data you can realistically collect without being intrusive. There are probably more than you think. Then look for ways that data can inform your experience. Do you have localized content? Are people coming in from a known url? What did they click on when they came to your site. All of these things can tell you something about your user.
6) Don’t judge people’s devices.
One of the easiest things to detect is a user’s device. The old assumption that users on mobile devices are strictly task-oriented, and need a truncated experience has been shot
down time and time again. People on mobile devices read long-form content, and there are crisis-users on desktop. Shoot for parity across platforms, and make optimizations where it makes sense. Again, it’s about helping the user, not controlling them.
Obviously, these are all really high level things. Every project is different, and every user group
is different. Ultimately it comes down to taking a realistic look at your users and yourself. Look
for patterns, and think about what you can do to deliver the best experience possible. Mind your
manners. What’s good for the user, will ultimately be good for your brand.