I love Jared Spool, but I think he is the Upton Sinclair of UX lecturers. Sinclair wanted to criticize capitalism but all he did was reform the meat packing industry. I’ve seen Spool deliver this speech twice. If his message was not to be consistent then that is lost is his wholesale ridicule of people who try new things and fail.

As Spool says here, UX testing isn’t cheep. You know what is though? Well…Cheep shots. Maybe next time I see him he can show some examples of successful UX design that broke with convention. I don’t know if it would get as many laughs, but it would probably be less discouraging.

Note that he says “foolish consistency,” which would fall into Jared’s concept of local consistency. I would argue that consistency without taking users’ mental models (or crrebt knowledge) into account is foolish, indeed.

Consistency is not necessary for good design. Meeting the users’ expectations is. They are not the same. The users’ expectations come from many different places — their previous experience with your design, their previous experience with other peopl…

So the fact is, the writer appears to actual be saying ‘Don’t just be consistent in your own world, be consistent with the rest of the world’ (or alike for non web related topics).

Nice one!

I’m a frequent reader, and respect your work greatly. First time poster.

I had the same impression as Partrick Moss (#3) about the disjunct messages between title and body. (Maybe for the sake of a concise title?) I came here ready to argue there are times when consistency is a good thing, but after reading the article, I find we agree when those times are (e.g., industry conventions). I would also say there are times when consistency for designs sake is useful too, but I don’t really mean “internal” design as you seem to be.

I think if you had titled this “Consistency in Internal Design is the Wrong Approach,” and define up front what you mean by “internal design,” then it would all jive better in my mind. For what it’s worth.

@wion

Really, you think! Consistent with what, may I ask? Consistent for the sack of consistency? Consistent with sounding proper at a meeting when you really have no idea what your talking about?

Whenever someone says “Make it consistent” I usually break out in allergies, because in my experience consistency has been the bail of anything creative and the force that misdirects a design team.

Obviously certain things have to have a uniformity of structure, and a repeating theme in the design, but what they really are saying is, “at all cost, make it an easy project that wont be a problem later.”

It’s a cursed word because it usually has the opposite effect. It’s a word that save you from looking radical in a corporate meeting, (and we all know where that can lead to). It’s a word that fails to address the very fact that “consistency” often times can be equally apply to doing something consistently wrong.

Every company has a nebulous definition of consistency, even when we think we are saying the same thing. Perhaps the word should be banned!

The AVIS folks shot for internal consistency, however they didn’t take into account that many users would come with an already existing perception of what the asterisk meant — the opposite of the designer’s choice. The result: confusion *because* of the consistency.

Instead of focusing on consistency, the right place to start is “What do the users know?” If users are familiar with design elements, and you design for that familiarity, you’ll end up with a consistent interface. But, that’s really an accidental side effect of designing for current knowledge.

Secondly, does this tend to exaccerabte the problem of constantly changing to meet the whims of customers, reinforcing an eclectic approach to life rather than attempting to build foundations? I’m not advocating a “stone age” approach to life (i.e. everything is in concrete and can never change). No, consistency provides a framework, an identity even a sanctuary. Some consistency, I think, it vital. Addressing customer needs and wants to important too and I support researching customer knowledge as a key means of delivering the product. But I also think some dynamic balance (i.e. a consistent concept applied upon customer knowledge) would be good, too. Example: pants; still fit over two legs but look at the pockets!