We launched a big website last week. The site soft-launched as a beautifully designed site that was fully functional but some features didn’t make it across the finish line.
The site is pretty big, roughly 35,000 lines of code. A few hundred more and those few final features will dive across the finish line. Launch was an anti-climatic mess that was more of capitulation than celebration. The development team had been running at full-speed for the final two weeks and was running on fumes. Nerves were raw, eyes were bleary and brains were foggy.
As the guy that is suppose to make that run a bit smoother than that, I felt like I had let my team down. What could I have done better?
Working at an agency has it’s pros and cons. At an agency a website is rarely just a website. It’s usually tied in to a plethora of other moving parts so moving a deadline once it’s been set is rarely an option, especially late in the game.
That means we must get the length of a project right months before we start, and often without all of the details. One the rare occasion that may happen, sometimes that schedule get’s over ruled by a clients own timeline that needs to be met.
At the point where the schedule gets compressed, we should have asked what functionality they would like to remove from the substantial list of features. I didn’t do that, instead we decided to just work a little bit more.
My friend Karl Sakas told me about a good analogy he had run across that compared development to a hike from SF to LA by Michael Wolfe, CEO, Pipewise. It’s one of the best analogies I’ve come across about why website scoping goes wrong.
The other insight I heard this week was about Blizzard Games and their mantra of “we’ll ship it when it’s done”.
A website is not a book. It’s never really “done”. It can be “launched”, but that’s different from done. There is always more you can do to a website. For many, a year after a launch you need to start thinking about updating the site.
How did your last website launch go?
Karl Sakas says
Most website launches are stressful — the client has high expectations, and the agency just wants to get it launched, especially if there’s been a lot of last-minute scope creep or delays in loading content.
In my experience, I find that the website the client wants at the end of the project is rarely entirely the same as the website they thought they wanted at the beginning of the project — hence the “never really ‘done'” idea you mentioned. Working via Time & Materials (as opposed to Fixed-Bid) makes it easier and less confrontational to adapt along the way, without constantly having to say “No” to every new request.
For the interactive agency project manager, the opportunity is reassure the client that things will be OK, and to help them understand the value of moving non-scope and “there’s no time” items to Phase 2. It’s important to demonstrate that you’re listening to their concerns, but it’s important to acknowledge that adding everything they now realize they want means the site will never launch. And of course, having kept the client updated along the way so there aren’t as many last-minute surprises.
Phil Buckley says
I agree Karl that it’s the client management that makes all the difference.
webb says
I could not agree more.