Using Psychology to Create a Killer Sticky Website

Posted May 28th, 2007 at 3:42pm

In thinking about current and future projects, I have done my best to be as conscious as possible about the reasons for doing things a certain way, rather than doing things a certain way because that seems like a good idea. That is, I want every feature and option to have a darn good purpose. I realize this could seem a bit obvious, but it’s so easy to just stick something here or there or add a “cool” feature just because you can.

So I set out recently to understand what makes some web apps irresistibly sticky and why others can’t seem to gain traction. It didn’t take me long to get totally obsessed with understanding this stuff because the deeper I got, the most interesting it all became.

I kept finding this great material from people who had the same thoughts I had (of course - you know there are ALWAYS people who have already had whatever question you currently have, and chances are that they have already blogged about it). Here is some of what I found.

“…the service automatically gets better the more people use it.” (O’Reilly)

First of all, if you want to build something that really sticks, you need to understand what motivates us. In reality, it all comes down to not only making something fun, but understanding what we think is fun and why.

I suppose it’s not hard to make the leap that game developers are the kings and queens of sticky, but that’s not something I had really wrapped my head around until now. Amy Jo Kim is pretty much THE source for this discussion, and she’s been at it for quite a while.

She lays out 5 components that aid creating sticky app or game - and she argues that the most sticky sites make use of all of them well. (Ebay being a prime example.) The idea is that you can actually shape the behavior of your users (a better way to say this is to say that you can get them to do what you want them to do to further the goals of the site and improve user experience). By developing concepts around these 5 components, you can shape the way your site/app is used.

• Collecting things.
• Earning Points.
• Getting/giving feedback.
• Exchanges/Gifting.
• User Customization.

How many sites do you know that incorporate these features? The ones that do are pretty darn sticky. Ebay, MySpace, Facebook, etc. That’s not to say that any app with these 5 things will work - it’s far more complex than that - but it does mean that these features matter.

If you want to learn more about what each of these means so that you can start incorporating them into your own sites, here are some great resources:

1. Amy Jo Kim’s E-tech 2006 PowerPoint
2. How Game Mechanics Can Make Your App More Fun
3. Game mechanics applied to social media: “Easy to Learn, Hard to Master”

Jay Gould of Bolt.com adds a few more to the list:

• Syndication.
• Competition.
• Tipping Point.

Make sure to check out the full post for his interesting thoughts.

“…users pursuing their own “selfish” interests build collective value as an automatic byproduct.” (O’Reilly)

If you want to build a web app that’s totally irresistible and undeniably sticky, you have to follow these golden rules:

• Your app needs a goal - some ultimate achievement - and that has to be obvious the moment that you hit the page. But once you figure that you, you need to forget about how cool that goal is because that’s not enough to create a site that sticks. You must focus on all of the steps to achieve that goal and make them FUN, EASY, and ENJOYABLE. Remember: it’s not about the goal, it’s about the PROCESS to get to the goal.
• Feedback makes even the most basic tasks fun. Allow people to leave one way message AND back and forth conversations. Think Ebay feedback, think MySpace messages, think blog comments, think forum private messages. Think about how people communicate and make those kinds of communications available to your users. Allow for ‘gifting’ between members. Gifts create a sense of reciprocity and people generally want to outdo the giver. “Gift economies” are very interesting, and I think we will see a lot more of this in the future - beyond Hot or Not, Facebook, et al.
• Customization abilities create a sense of ownership and belonging. Being able to make something represent who we are at the core makes us more likely to share it (perhaps that’s an all black page with gray text, or a pink page with sunflowers), and ostensibly makes it harder to abandon because it’s really a part of who we are. MySpace is fugly, but it works.
• Take customization a step further by customizing data and suggestions for each person. If you know someone loves Metallica because their (now required on all UGC sites) says so, why not suggest other Metallica fans to them as friends, or groups that talk about Metallica? If you aren’t talking about a more social networking site, you can still make suggestions based on what we know for sure about a user and all other users. If a you are always giving users the next click, that click will be far more useful if it’s actually relevant to their stated tastes (or surfing habits). Study Amazon’s ability to suggest… Have you clicked before? Were you interested in why or how they might know you so well?
• Everything is ranked and sorted and everything has a number attached. If you attach a number to someone, we are naturally inclined to want to see that number rise. Depending on the site, it could be the number of friends I have, or it could be the number of photos I have submitted, or it could be the number of posts I have made. It instinctively feels good to see “our” number increasing. Anything that’s important to the quality of your site should have a number. (Hint: Don’t make it about friends.) Perhaps there is also a general experience number that ties in all sorts of aspects that be called Experience or Karma. this number could be affected by submitting content, adding comments, community moderation, etc. Once you have numbers, everything is ranked and those ranking are public. Just watch how people suddenly start increasing their activity levels…
• Give your users a reason to keep returning every day. If your users don’t have a reason to visit daily, give them one! Give daily log-ons weight in your overall points system. Have a robust user notification system that nudges them about how many new posts/reviews/friends/etc. have been added to the site since they last visited. Depending your site, this will be different, but do what you can to give a REASON to come back.
• Sit down and figure out your EXACT goals and then build around them. If you are a review site, you want reviews. Every single game component should be built around that one goal. Don’t let your mind wander and don’t start adding things that aren’t a part of that one goal. Add features that encourage THAT behavior.
• Much of user-generated content is crap. I love UCG, but the reality is that a lot of it isn’t that great. That’s okay, though, because your incredibly brilliant app will be able to bring the cream to the top with very little user interaction. You know what people are clicking on, what people are commenting on, what people are sharing, etc. You will do this well because you know that it’s your job to locate the best content, while also letting users dig as deeply they want into the site. Have analytical tools in place that bring the good stuff to the forefront for your users. Never make them look for it unless they want to.
• You are not creating the content, you are creating ways to USE and DISPLAY the content. That’s your job. Focus on ways to use the data that you have that will improve on user experience. Make your data available to outside developers through an API. Don’t horde it! If you have data, it can get used by someone. Think about it: Google’s charm has nothing to do with the actual sites they index, but it has everything to do with the DATA they collect to create their rankings and how they manipulate that data. That’s what creates the listing of websites in their results pages. Never forget - it’s not so much about what you collect as what you DO with what you collect.

Let me leave you with this:

The uniting theme in all of this is that you need to understand what makes people act and react in order to develop something that they want to use again and again. This is about creating interfaces that understand the way we consume and create content. To me, the future of social applications on the web is fully understanding how we can better use data to create userfulness and relevancy. For some apps it’s about how people are connected, for others it’s about how things are connected… I think we have barely scratched the surface of intelligent, evolved, niche social networks and applications.

Filed under General

Tags: No Tags