Enterprise Bearware

The price customers pay for software and the level of usability they get with that software is inversely proportional.  As long as it solves a big business problem or two, it doesn't matter how easy it is to actually use the stuff.

Consider the obvious extremes when calculating this philosophical average:  What version of Oracle has anyone ever installed pleasurably?  It's not cheap, either.  Is there any doubting how easy and free it is to find obscure data from Google?

Selling Bearware
There's more to it than that, sure.  But it does make sense.  Enterprise software salesfolk are bear hunters not fisherman.  It's all about the million or multi-million dollar deal. 

If we assume that what the bear pays for actually solves a million or more dollars of problems, it's easy to see how insignificant a glitchy installation or configuration experience becomes.  Hire a few more consultants if you need to.  No biggie.  In that scheme of things, the unusability cost becomes irrelevant.

This is the sorry quanundrum the lone usability engineer in an enterprise software company faces.  He's rightfully concerned when 12 million rows of data are presented in a double-nested, double-scrolling, multitabbed, 1...100000000000000000-paged web application (Seibel anyone?).  But do his cries get heard? 

No.  He may even get fired for consistently not understanding and adjusting to business priorities.

But the business folks have a point, too.  How much will it cost to rewrite the page to make navigation easier?  Will this become a differentiator that justifies the cost?  Probably not.

The competition - what if a competing vendor's product is twice as usable, you ask?  This would raise an eyebrow in a proof of concept demonstration, but not much more.  No profit-oriented company sends an incompetant, weak-hearted person to prove their software to a big bear customer prospect.  They know how to work around the usability problems.

When Usability Matters
Usability improves as the price decreases.  It's not always a smooth sloping curve, though.  I say this because even in enterprise software, there are the 'cheaper' ones.  In frantic deals, some are even free. 

When software is affordable for the individual consumer, everything changes.  Here on the graph, if users get frustrated or unhappy, they bad mouth or jettison the piece-of-@#$% in favor of an equally suitable, inexpensive alternative.  Not so with Enterprise Bearware.

For consumer software, word-of-mouth is often the only marketing vendors have.  Bad word-of-mouth is really bad for business.  For them, usability is worth every nanovolt of thought burned on it.  Since great usability is hard to do, getting it right sets you apart, leading to great word-of-mouth marketing.  All the right usability incentives are in place.

So, how does an enterprise software company write usable software?  Two answers: luck and altruism.

Comments

[Dylan Greene]

Ey! More monies for bosses! Charge 'em for the training, charge 'em for support, charge 'em for the consultants, charge 'em if they want to get the fixes!

[Nathan Lee]

Never ceases to amaze me how much god-ugly enterprise stuff there is out there.. I'm not talking in house produced stuff (that's a whole different type of ugly), I'm talking off the shelf packaged enterprise products.
Sure they're monolythic, monstorous apps, but surely someone could take the time to sit back and go: "hey, these days user interfaces are a bit different from this 20year old hunk of crap we're peddling for a million bucks a hit, maybe we should spend one or two licence fees worth of developer time on the UI".

On the cost front: I think a lot of the time it's all in the mind, big companies want to have what they think is the best. And I often do the same thing, if I'm looking through the wine list and I don't know any of them, I look at the prices, pick the one a couple from the top that's within the realm of reasonable. Likewise with other products, some you might know about and can be a bit more descerning, others you tend to trust that price = quality. I think the people involved in big big sales are quite often applying this principle (which is fair enough, as market forces tend to mean prices reflect what something is worth). It just happens that software is a bit of an intangible beast, so you can't really go up and kick the tyres, nor can you really go and "play with it" because to set the thing up you need X machines, Y gigs of ram, Z terrabytes of disk space and the multiple of all that in consultantcy fees to install it etc etc..

The other thing I've seen in action is business people who are pushing the project but not actually end users (99% of the time) having a very different view of how something should work than the end users would like. Because they aren't the ones to be stuck using it, the impact in terms of convenience, eyestrain, RSI from having to click 200 times to do something isn't a big deal to them for the 10 minutes they spend poking around the system during live demos..

Nath
PS like the bear hunter analogy.. I picture some of the vendors walking around with a laser scoped tranquiliser cos they're hunting elephants.. That and they have a hunting party of big consultancy firms just waiting for the thing to get taken down so they can find the most expensive team to carry it away to lock it into a zoo for the next 20 years.. that and it'll take twice as long as they say it will, and will invariably end up just building a new zoo around the thing rather than take it to where the original plan was. The similarities never end do they ;)