Friday, August 29, 2008

"Make Something People Want"

I came across this article in the Wall Street Journal regarding issues that sellers have been having with eBay. This reminded me of a recent conversation I had with another software developer friend of mine regarding how craigslist could be improved with...something else.

If a potential competitor of eBay were to read about its users' woes then it would definitely be a nice blueprint on how to build something better. Why? Because it is "something people want" but are not getting.

I don't know if eBay's problems could necessarily be solved via technology (although some looked to be that way). But, it serves to show that just when you think something has been "solved" then think again. Has online auctioning really been "solved"? Has online "classified ads" been solved? Has "[fill in the blank]" been solved?

Classic example is Google. When they first entered the search engine market everyone believed that search engines have been "solved" and it could not possibly succeed against the more "mature" search engines available at the time...well, we all know how that story ends. But even with Google what it is today, still, you can not assume that search engines have been truly solved. For example, there are folks that do feel it is not solved and therefore are attempting to take on Google. But given Google's own beginnings it does not sound strange at all and should serve as an illustration that it can be done (or at least attempted).

If you do want to build something people actually want then keep some of following things in mind in respect to your competitor:
  • Keep your features list smaller- More features != better product. Most people think this makes for a better product but that is a complete fallacy that has been proven time and again. Keep it simple and focused on features that people actually would use. It does not matter how many features your competitors has in their product. It only makes their product worse and harder to use. Check out the book 'The Inmates Are Running the Asylum' to get a better idea as to why this is true.
  • Make it easier to use- Sounds "easy" but it is not. You really need to cut down on the amount of friction it takes in using your software as compared to your competitor. Make it goal oriented and not task oriented. Check out the previously mentioned book as well as 'Don't Make Me Think!' for infinitely better explanations than mine.
  • Stay smaller and leaner- Your main goal is your product. Try not to worry about building an empire. Do not focus on other unrelated products that you could make. Do not aim to take the whole market because in some markets even as low as a 5% share can be quite lucrative (you might be serving some niche that is not being adequately satisfied with whoever is the current market leader). Do not hire extraneous folks which only serve to drag down your costs. And so forth...
  • Ignore your competitors- Seems like a contradiction to the previous ones but not really. Who cares what your competitor is doing? Care about what they are not doing and what that means for the users.
Finally, as a software developer/engineer/programmer/hacker what can you do? I suggest you start here.

No comments: