South African bank Absa just rolled out their new online banking portal. There are two things about this launch that raise red flags for me. First, from ABSA rolls out new Internet Banking revision:

This launch follows a successful trial with the bank’s 36 000 employees over the past few months. The trial allowed the project team to identify and solve any defects and gauge the response from users, via over 1 300 feedback emails received from employees.

It’s shocking that we still have to talk about this, but let’s just state it again, as clearly as possible: You are not the user. You cannot test a system on employees — who know all the intricacies inside and out — and think that you’ve done appropriate user testing. There are plenty of solid arguments and evidence for this, but for now I’ll just quote Jakob Nielsen:

One of usability’s most hard-earned lessons is that “you are not the user.” If you work on a development project, you’re atypical by definition. Design to optimize the user experience for outsiders, not insiders. The antidote to bubble vapor is user testing: find out what representative users need. It’s tempting to work on what’s hot, but to make money, focus on the basics that customers value.

Also see Myth #14: You are like your users and You are not your user, which both have a lot of great points and research around this.

Second, there’s this quote from the Head of Retail Markets at ABSA:

The development of Absa Online saw up to 140 individuals, working across three continents, putting in an astonishing 450 000 man-hours of development work. Four million lines of code were written in this, a “first-of-its-kind” technology deployment in South Africa.

Well, that is just silliness. As one of the commenters on the article points out, it’s Bill Gates of all people who said, “Measuring software productivity by lines of code is like measuring progress on an airplane by how much it weighs.”

The size of the project, how long it took, and how many people worked on it is completely immaterial. What matters is if the thing works well to help real users accomplish their goals. I really hope it does, because this looks like an outrageously expensive project. Hopefully it doesn’t become South Africa’s version of the Four Seasons $18m redesign.