Procedures: when do users stop using them?

[Geoff Hart <Geoff-H -at- MTL -dot- FERIC -dot- CA>]

Susan Harkus reported an interesting test result:

<<I once worked with a software development/writing team that faced the
failure of a field test for their product because the users had used the
installation procedure in the same way as many (and I would suggest a
majority of) users do. The procedure was long. As soon as the users felt
they knew what was going on... they stopped reading the procedure and
"proceeded" using the famous "winging it" strategy.>>

Actually, that's not a failed test, but rather a remarkable success: you
just discovered a major aspect of how your audience behaves, and can take
appropriate measures in writing your procedures. For one thing, you need a
clear introduction to the procedure ("this procedure is different from what
you've done in the past in the following ways [list of ways]; don't try to
wing it until you've gone through the whole set of instructions at least
once"). For another, you need "advance organizers" at all the points where
readers are most likely to jump away from the procedure and strike out on
their own; an example might be "unlike in [software name etc.], the next
step is _not_ to refurb the bafflegab; in this procedure, you will have to
glarb the bafflegab instead".

<<I read an article some years ago now that ... indicated that men resented
procedures because they felt a sense of disempowerment and that women were
more accepting.>>

Like you, I'd be very cautious basing any usability decisions on such
simplistic overgeneralisations about male and female roles; typically, there
is more variation within a male audience in behavior than there is between
male and female audiences. To use a really extreme example to dramatize the
point (reductio ad absurdum), would this generalisation apply to a female
dominatrix and her male submissive? <g> More realistically, would a female
engineer and a male bureaucratic drone meet this generalisation?

<<The more user interfaces offer opportunities for learning by exploring,
the greater the tendency users will have to "jump" out of a procedure as
soon as they feel in control.>>

Agreed, and this will only get more true as our audiences grow increasingly
familiar with computers. Currently, we still have a large number of
computer-illiterate people in our audience, but given how aggressively
schoolchildren today are beginning to use computers, that situation will
begin vanish within our professional lifetimes. And that being the case, we
need to understand this change and monitor our audiences more carefully; at
some point, we'll be providing more contextual "here's what you need to know
to explore on your own" information, and far less "follow the steps and
don't stray from the beaten path" information.

<<So what does this aspect of user behaviour say about numbers and procedure
steps?>>

All excellent points, but I'd add one more: find out where and why your
users are likely to strike off on their own, and provide a map and safety
instructions so they can do so with confidence. Less metaphorically: Give
them the information they need to know so they can stop using the docs. I'm
a bit heretical in this, perhaps, but I feel that the best documentation in
the world would teach users enough about the product that they could then
throw away the documentation, or keep it on the shelf for only occasional
use.

<<The user interface might suggest the work flow and you only need to
explain the "why", "when", "why not", "what also", "what next" and "what if"
issues around the user task.>>

That's the holy grail of usability: matching the interface so closely to the
way people think and work that they no longer need documentation. Within our
professional lifetimes, I predict that we'll see a strong move away from
"external" documentation (on paper or in online help) and an equally strong
move towards incorporating our knowledge of user behavior directly in the
interface; context-sensitive help is a pale, feeble shadow of this, and
affordances are a step in the right direction, but it's our ability to
understand how users think and work that will keep us all employed 20 years
down the road.

--Geoff Hart, Pointe-Claire, Quebec
geoff-h -at- mtl -dot- feric -dot- ca 
"The paperless office will arrive when the paperless toilet
arrives."--Matthew Stevens