Brainspew on Killer App Theory

by phil on Tuesday Jun 3, 2003 12:11 PM

It's always been a dream of mine to develop the next big killer app. So anyways, I've been writing a little bit, to myself, formulating my thoughts on how killer apps work. Very unfiltered, I'm sorry, but hey, at least it's something!

The following are products that took off rapidly and developed their own niches. Here's an analysis of how and why they succeeded.

Winamp
When mp3 first came out, there were already quite a few mp3 players out there, but they all had like two or three flaws. So there's this, lack of flaws characteristic that prevents killer apps from taking off. If you have like a few bugs or hic-ups in the UAT (User Activity Train) that derails him, he will lose confidence and motivation. If an intention is created and then fulfilled through manipulation, then the user is satisfied, and continues to use it. The usage is the ability to play one's mp3s, sort playlists, without any problems. Other programs crashed every now and then, or had cumbersome interfaces. Winamp was simple. Even musicMatch, I don't use it because it includes other features that are "thoughtful" but don't speed the UAT.

UAT scenarios:

- Seek, User-driven idea or intention.. The user comes with a desire or idea, and then resolves themselves to implement. You present the most likely options straight up, and get them to execute their activity with the most ease and speed possible. Double-clicking on an mp3 should begin to play it, not open the program and then "ask you to add more items to the list."

- Wander, Passive user creativity.. The user notices features or other things he can do with the software, and then starts to "fiddle" with it to explore other features. This gives him more ideas or intentions that he'll do later, therefore adding to the utility of the product. i.e. Google's selling point is NOT the I'm Feeling Lucky, but those that stumble into it then take advantage of it and THEN it becomes the value of it. Google's intention

- React, Prompted for action, and then response.. first the user has to be receptive to a particular action. A pop-up on my screen is quickly rejected because I'm not expecting it. So if there is an expectation created, maybe by the placement of the program somewhere, then people are receptive to receiving messages from their system tray. Then, upon the prompt, the improvisational response to it is then required since the user wasn't thinking about how to respond when the response came, he was prompted, so the interface for this has to be extremely simple and passive.

- Shelve, Products are just tools, so then a part of the user experience is in the retrieval and replacement of the tool. Opening the program or product must be easy, or it must fold away properly, and coexist with all the other tools. Tools that interfere with other tools then cause the user to react by throwing the product away. You want the user to be comfortable with the way the tools are shelved away. i.e. we don't want a HUGE start menu in our face, we need it in it's right place. A program should only occupy the space that it needs to occupy to allow them to Seek, Wander if necessary, react.

Programs are either given to us as in bundles, so retrieving it is not a problem. But in most cases, you want a product that people first want to seek out, and then have things they seek by using it, something they're looking for, in order to get the utility. If there are other things that they can wander and do, that's fine, but people rarely take a product just to wander unless they're early adopters.

The UAT's in all cases should be as simple as computationally possible. The least number of steps should be made in order for the seek's to be fulfilled, ports for wandering should be easily accessible and not hidden, like easter eggs, shelving should be appropriate, as should the reactive systems.

Multiplicative Effects of small features, viral marketing, etc. Hotmail and Napster are intriguing for me because of the multiplicative effect of very small features.

First, early hotmail and napster both had no interruptions in any of the UATs. You sought something, you got it, no bugs, no hick-ups, everything was in its right place. The key to hotmail was placing that link at the bottom of their e-mail to the hotmail service. This seems commonplace now, but this was, I think, their innovation. This then made it so that people would infect other people with the utility they found in the program. So if bill gates sends me an e-mail and at the bottom it has an endorsement for hotmail, he's adding his name to the credibility of hotmail, and therefore, word-of-mouth spreads quickly.

The small thing that Napster did was extremely simple and clever. When you initially sign-up to the program, it scans your hard drive for mp3s, allowing you to contribute value to the system automatically. Then, when you close the program, it just sits in your system tray. Most people have no idea that this happens, while unbeknownst to them, they're increasing the mp3 pool size. They can, of course, if they seek, easily turn this off, but most people don't. Most people are usually seekers and just want to shelve things as quickly as possible. Then, when they're seeking an mp3 a few weeks later, boom, it's there. There was no wandering in napster either to distract from the main purpose of pirating mp3s, it was right there.

Efficient UI cannot be stressed more and more. A designer should prioritize the activities that the user is seeking to accomplish by being on that page. This is what distinguishes ICQ, Google, and Early Yahoo.

With ICQ, they realized that you don't want your contacts to be right in your face, but rather in a tray that you can invoke when you need to and shelve when you don't need it. So they made the floating contact list that could dock on the side or be minimized. They also realized that needs to be ever-present, so they made it minimize into the system tray, easily accessible, not occupying the whole taskbar at the bottom. These things seem like common place now, but I remember early chat programs, everytime you'd start a chat, a whole window would pop-up. To access to your contact list you'd have to go through a drop down menu. When you closed the program, it went away completely so you couldn't receive any more messages, and when you did, it would just pop-up in your face. Otherwise, if you minimized the program, you'd have this program in the background, sitting on your taskbar, annoying you. In addition, ICQ worked with consistency. Other programs, I had to take 3 or 5 steps in order to make simple contact with my friend, and even then, it might not work. The next time around too, it was also inconsistency.

People take solace in consistency whenever they use programs. They're seeking something every time, and they want the program to operate the way they envision it to, otherwise it throws their whole seek train out of whack.

The thing that ICQ did, when you shelved it, was also make messages pop-up in the system tray with a small flashing notification. This is brilliant because it doesn't make a desperate alert for your attention, but allows you to react in the way you want to. Double-clicking on it instantly gets you in response mode to that message without any more finagling.

Now, back to Google and early Yahoo. These both were simple and designed to help the user achieve the one or two things that they were going there for in the first place. I use Google to search 99% of the time, so that's the only thing I really see there. The logo is there to make it more friendly and seem usable, but other than that, Google is just a small text form. The complexity is hidden from the user, as it should be. Other search engines I've noticed before tended to shove a bunch of features onto the front page as if they were impressing the user about all the great things this site did. Users are only impressed by the speed and effectiveness by which what they seek is found and executed. People feel utility as they utilize, not by observing, "wow, this site has a lot of neat items on it, so it must be good." These products again are tools, and people use them to do something in particular. If you interfere with what the user is intended to do, then he will move onto something more efficient.

Early Yahoo was also like this, all there was was a directory and a search field to search the directory. I sought some particular website, I knew Yahoo would take me there without interruptions. Yahoo now is a portal that appears on people's start pages, and so they have to bank on the wandering factor. Although, I wonder how often do people look at their IE start-page and just surf around on there, clicking on "Rock n' Roll" or "Philosophy" just for kicks. So then current Yahoo, as with the MSN start page, they focus on reaction and prompting the user. MSN will always have a catchy headline like, "Fight Terrorism at Home" or "25 things to do this weekend" People open up IE and then are then prompted to wander around MSN. Clever.