Lessons in building for the customer…
Posted by pallid on November 25th, 2008That’s the one thing I’ve learned over the roller coaster ride that was the last two years, building software for clients both for consumer-facing properties and back office applications, and with my fair share of both success and failures on both ends. When I first started out on this, the word “Web 2.0″ has just started it’s long slide into meaningless-buzzword land, and hand-sculpted AJAX was a pain in the butt without frameworks like JQuery and Yahoo UI to help you along. And as we bopped along, I’ve learnt these lessons first hand:
- Engage the end user, BUT the customer (end-user) is almost never right about what they need. Don’t listen to them tell you what they want to put into the software. Listen to what their problem is. Easier said that done, that. Especially with deadlines and an opinionated customer!
- When you are done with the application/site/software, when all the features have been coded. Pat yourself on the back: You are only 20% to 40% complete! QA and problem solving is the other 80%. Silly me… I had to learn that the hard way. Of course there are ways to mitigate this, but you can just Google it.
- Always, always build the HTML first. Most devs don’t design (nor should be expected to). Designers don’t code (nor should be expected to). Which leads to the following point;
- Always prototype in usable, semantically correct as possible HTML code. The Devs will love you for it.
- Don’t re-invent the wheel. Obvious? Well, how about this: You better make damn sure that other wheel fits first before deciding not to re-invent it. Nothing is a fit all solution.
- Don’t re-invent the wheel. No not a typo. You’ll be surprised how difficult that decision is, whether to reinvent or not! It’s not as clear cut as you might think.
I don’t think the learning ever stops… 
