What does February 1st have to do with testing, you might ask?
Well, people are great with signing up for a gym membership the day after New Year’s. But the only statistic that really counts is, “how many people are still at the gym when February starts?”
Recently one of our many new clients signed up with us, and, clients being clients, this one was gung-ho to get started. “Throw as much at me as you can!” was the challenge.
Now, I’ve heard that before. Usually it turns out that our definition of “throwing a lot” is way more than the client can handle and pretty soon “Uncle!” gets called. So we have a process for ratcheting down the volume to something the client feels is manageable.
This particular client enthusiastically jumped into making some improvements on some landing pages week one, but by weeks two and three other important issues — all legit — were holding the tech team from implementing planned changes.
The question here is how that priority list got put together. If the test you just postponed a month ends up lifting your conversion rate from 2% to 3% (a 50% increase by the way!) and your small company currently sells a million bucks of product a quarter, that one month push-back cost you $167,000 ($1,000,000/3*.50) in sales, which is the Opportunity Cost of having delayed that test.
Even at American developer prices, something like $125/hour for really good people in this economic environment, that’s equivalent to about 1300+ techie man-hours that money would have paid for. Are you sure you don’t have the resources to do both regular development and testing, too?
Put it another way, if your company were trying to raise investment money in this economic environment and your multiplier was a now-reduced 8x multiplier, you may have just walked away from an extra $16m in valuation ($167,000x12x8). Still want the techies setting the testing priority agenda?
Which leads to me to my point: “It’s hard work doing hard work!”
On the surface that may sound obvious, but somehow human nature makes this easy to forget. Or to under-estimate the amount of work between “thinking it up” versus “getting it done”. Or to hand off to an assistant. But the problem hasn’t gone away; you’ve just transferred it to someone else’s gym membership. And soon it will be February and either your jeans will be fitting looser or you’ll have a couple of completely legit-sounding reasons for being a Conversion Potato.
Back to our actual client: the trick seems to be how to effectively tickle clients into realizing the cost of not testing continuously and to constantly re-state that information in bottom-line financial terms that shows the impact of such testing. It doesn’t sound like much to push things off a week, or two weeks, or a month, up until you finally get there there and realize “oh my goodness, do you realize how much more money we’d have if we’d improved this back when we planned to?”
So … what is your testing plan for 2009?