Before you read the pre-launch, listening to customers, and development advice, though, look at the checklist below to see if you’re destined for success.
Your New Software Venture Will Succeed If:
Approaching the Launch Pad Before software exists, there is a business problem. You have this business problem and you realize that software could solve this problem. From your intrepid band of programmers you or your COO assembles a team. Because software is code, and because code comes from programmers, the quality of the software you build depends almost entirely on the quality of your programmers. Selecting your team members is the most crucial aspect of building your solution.
John Marshall envisioned a web server log analysis tool that was not database driven. He had the technical acumen to realize that this was possible, but he didn’t know how to pull it off. During interviews he asked potential employees how they would build this particular software, and they all said that they would use a database.
He wanted programmers who were willing to move past the fundamental tenets of software architecture and build something that would work more quickly and efficiently. After a “stand up blazing row,” he convinced his potential crewmembers that his idea was possible. Then he hired two of them out of pocket, even though he knew the company might not be profitable for at least a year. After six months he received private funding, and he expects to be fully revenue supported by the end of the year.
If you’re assembling a team, or even if you’d like to screen the programmers you already employ for a special assignment, read Joel’s article called “The Guerilla Guide to Interviewing Programmers.” It’s a great way to find out who’s a shining star of programming and who’s merely above average. Once you’ve assembled your team, it’s time to set your launch date in stone. This date is immovable. John launched ClickTracks before a major trade show, in which they had a booth, forcing them to ship the best product they could in the time they had. Setting this launch date is a crucial step in the development process, whether you’re just starting up or bringing your in-house tools to market. Your nonnegotiable launch date is an acceptance of one of the basic facts of software development: it’s never done.
John stressed that first-time software vendors “just have to get something out there.” Your team could program until doomsday and still not perfect the code. In addition, you don’t know how to improve, or even market your software until you get some user feedback. User feedback is so valuable to the company that John spends almost half of his eleven-hour days talking with customers and potential customers.
In another of Joel’s articles, he declares that software vendors must “kill known bugs” before launch. His reasoning is that, especially if you’re beta testing something, you don’t want people to spend their time catching bugs that you already know about. He also has an article that states, “Fixing bugs is only important when the value of having the bug fixed exceeds the cost of the fixing it.” To fix or not can be a complicated issue.
I asked John about resources that he found useful as he brought his software to market and he began to laugh. “My wife,” he said, “has by far been the most valuable resource. She was there at the end of everyday offering support.” Let that be a lesson to those of you planning to launch software – marrying someone who supports your chaotic entrepreneurial lifestyle could be your smartest business decision. Now You’re Ready for… Lunch Now that your prototype is up and running (except for the bugs that aren’t cost-effective to fix), you’ve got to find some people to try it out. John offered a unique approach. He applied a distinctive sandwich-based design criteria to the 30-day evaluation (eval) version of ClickTracks available online. He challenged his team to make the software easy enough for a hungry executive to download, install, and test, all while eating a sandwich during lunch break. To make the eval copy function this way, his team made sure that there was little-to-no configuration time. This meant keeping typing to a minimum, and making intelligent configuration decisions for the user. Oh, and ClickTracks offers a free McDonalds’ burger to anyone who tries out their software and offers critical feedback (some restrictions may apply). They have never lacked for customers, or feedback.
In regards to collecting email addresses, John requires a valid email address for anyone who downloads the software by sending out a serial number key. This keeps away anyone who’s not serious about using the software, and begins the customer/developer dialogue that’s so crucial to ClickTracks’s success. On Programming and Programmers: To reiterate, programming is never finished. Especially when, in the case of ClickTracks, business requires a constant conversation with customers and potential customers. Your end users will always have surprising suggestions and ideas for your software that you never even considered. That’s the de facto of development. One technique John uses to keep the team fresh is to talk about changes they’ll be making to the software six months down the road. Programmers prefer intellectual challenges to the tedium of, say, graphic design. By focusing his programmers on the future, they’ll be ready to make it through the boring aspects of programming. If you’d like insight into the social world of programmers, consider reading “Communicating with Einsteins.” Your in-house software could be more than just a useful tool – it could be your next business venture.
Garrett French is the editor of murdok’s eBusiness channel. You can talk to him directly at WebProWorld, the eBusiness Community Forum.
