There must be millions of forms on millions of web pages which ask the user to enter a date.  They are not hard to create and they are even easier to copy.  That’s good because we, as users, know pretty much what to expect.  So confronted with a page on the Microsoft web site, I was pretty sure I knew what to do

MS date boxEasy.  Drop from January down to June.  Drop from 1 down to 3.  Drop from 2007 down to 2008.  Done.  Except that the 3 has mysteriously become 1 again.  That’s odd, try again.  And it turns out that if you change the year, the day resets to 1.  So you have to set the year before the day if you want to get the right answer.

Utterly trivial, of course, and easy enough to get round.  But it’s quite impossible not to spot the problem the moment you try to fill the form in, which can only mean that nobody who has anything to do with the page has ever bothered to use it.

Seth Godin reports on  a more complicated version of this, trying to navigate telephone options:

Let’s say the person in charge of your retail operations does the following every single day:

  • Puts up a sign indicating which of five doors customers should use.
  • Locks that door.
  • Randomly unlocks another door.
  • When someone figures out which door to use, he runs out and kicks them in the groin, then locks the door.

Maybe, just maybe, after a day or two of this, and a few warnings, you’d realize that this person was doing serious damage to your organization, no?

As Seth observes, outside the physical channels, the problems, and the frustrations of the users are simply much less visible, so problems can just sit for long periods without being obvious.

If the person in charge were stealing laptops or peeing in the soup, it’s unlikely he’d still be around, no?

It’s pretty obvious: the CEO would notice the angry crowds in front of the store, she’d notice the police being called and the riot out front if the person in charge of the front doors was such a jester. But voice mail trees are invisible and the CEO doesn’t notice them. She should.

Exactly the same phenomenon is apparent in other areas of customer experience.  A few years ago I was in a room full of senior managers who were suddenly asked to fill in a form which thousands of their customers had to grapple with.  Nobody found it easy and there were some pretty shamefaced looks being exchanged:  in a couple of places in the form, it just wasn’t clear enough from the context exactly what was being asked for.

The requirements are pretty clear: an obsessive attention to the detail of how your service works, and the ability to fix glitches quickly and easily.  It really shouldn’t be hard, but far too often it clearly is.


  1. One potential implication of this post is “Go out and hire obsessives!”. This may well be good advice, and this sort of personality type is probably one whose chances of being hired are adversely impacted by prejudice even where their skills are objectively required. However, what about the role of metrics and technology? Surely it you monitor: abandoned calls, numbers of times people have to retrack their steps on the voice mail tree, number of customers whose speech patterns indicate anger when they eventually arrive at their human destination etc, you should be able to see quite quickly whether your soup is permanently awful or exactly when someone started peeing in it? The same would obviously apply to online forms and, although it is harder with written forms, there are probably still some metrics-based approaches that enable you to monitor when things are not going well – numbers of non-returned forms, numbers of empty boxes on forms, numbers of crossings out on forms, number of times customers ask staff for help in filling in forms etc. So in addition to having an obsessive form checker I would suggest: robust testing of forms with customers; ample scope for customer feedback; and some metrics to confirm that customers are not having problems with the form.

  2. You are right, of course, that lonely obsessives can’t be anything like the full answer, and all your suggestions are good ones.
    The gap, though, it seems to me, is more about ownership than obsession. All the metrics in the world about the way a web page is used won’t make a difference unless there is somebody who is passionate about wanting it to be the best possible web page. So yes, it might work better for them to be obsessive about watching people drink (or reject) the soup, rather than to obsessively drink soup all day – but it is the combination of ownership and commitment which is often the gap.

Comments are closed.