If it ain’t broke… it can probably still be optimized
Any time I change anything on SearchTempest (or any of the Tempest sites for that matter), I get two broad categories of responses. Some people write to thank me for the new feature, let me know how useful it is, etc. (We like those people.) Others write to inform me that I must be most mind-numbingly stupid person on the face of the planet to think that anyone, ever would use this ridiculous new feature. (OK, they’re not all that extreme, but that’s the gist.) I admit I enjoy the comedic effect when both varieties of emails arrive at the same time. 🙂 To an extent, I can understand where they’re coming from. It’s the same reason people hang on to ancient web browsers or other software when new, better, free alternatives are available. When you’ve found a solution that works for you, there’s little motivation to take the time out of your life to learn something new. With a website though, you don’t have a choice, which can be frustrating. On the other hand, we don’t have to look far to see what happens when you take “If it ain’t broke, don’t fix it” to extremes. You get craigslist. Craigslist is fantastic, but I’m guessing you wouldn’t be here if you thought it was perfect. So what’s the solution? Well, first I’ve had to accept that I can’t please absolutely everyone. Instead, I have a rough algorithm for any changes I’m considering:
Likelihood of change = difficulty level * (users helped – users annoyed) / total usersDifficulty factors in because it’s better to do a bunch of things that each help 10 people and take a day than one thing that helps 20 but takes a week. Second, when I’m contemplating any change, I think on how I can minimize the annoyance factor. Mainly, this involves making things optional. For example, the SearchTempest results originally showed just one craigslist city per page, with Previous and Next links to switch between them. Nowadays the default mode shows 20 cities per page, which lets you flip through them more quickly. The original mode is still available though – just click the Display Options tab on the results page. About 10% of searchers still use it, so clearly it’s worth keeping around. That’s not always the case. In theory, making everything optional sounds perfect. Everyone gets exactly what they want! There are two significant drawbacks though. First, each additional option means additional development and maintenance time. And every minute I spend keeping a legacy feature working is a minute I’m not spending on a cool new feature, or even a cool new site, like MovieTempest.com. Second, every additional option makes the site incrementally more complex, which makes it appear harder to use. Even if all these extra options can be safely ignored (which they can due to intelligent defaults), a new user doesn’t know that. And losing new users to other sites because they have less features is an irony I prefer to avoid! 🙂 So in the end it’s all about balance. I will always keep innovating, because if I didn’t, we’d still be stuck with craigslist. I’ll make changes optional whenever possible, because I want the sites to be useful for as many people as possible. But occasionally I will likely do things you don’t like. I hope you can find it in your heart to forgive me. 😉 And if not, I at least hope your email’s funny.