Most business software stinks. That isn't coincidence, it's the result of the same "coercive command economy" dynamics that fail economically. The people who make this software don't sell it to the people who will use it, and the people they do sell it to can usually force people to use whatever they pick, no matter how bad it is.
This is slowly changing, in part because of experiences employees are having on the web. Can you imagine if people had to take training classes to use Amazon.com? Of course not. Those kinds of experiences have an effect - once people see that there is a proven alternative to unsatisfying arrangements, they tend to begin asking for something better. Software vendors like SuccessFactors, whose HR software is being brought into companies because managers are asking for it, or QlikView's approach to "how is my business doing", are leading the way.
Within most organizations, however, it's still an uphill battle in most places to sell the idea of doing up-front work to make software easier to use. Jakob Nielsen has some ideas that sort of speak to IT executives, but they don't really speak to business executives. Based on my experiences, here's a pitch that ought to do better:
Like every other project you fund, our software project has a business case. But software business cases depend on usage. Even if you can force our people to use it, software that makes life difficult will get minimum use, not maximum use. The catch is that people always assume maximum use when they give you business cases.
Now... I don't have to tell you the difference between employees you've managed before who just get by, and those who give you the maximum. You know how big that is. Well, that's the difference between what engineer-designed software that gets by gives you, versus well designed, usable software. And the difference to your business cases is comparable.
Which is part of the reason why most business cases for software fall short of projections, in practice, and why executives (like yourselves) mentally apply a discount to those projections that you don't apply elsewhere.
Like anything else, it's going to take a bit of time and practice for our organization to get better at this. But if we start with this project, and continue to develop the idea of fully used software as a corporate competence, we can get closer to realizing the projections in our business cases. And that means dollars on the bottom line.
While the above pitch is designed for internal audience, it can be modified for commercial software development as well. After all, people don't buy your product - they hire it to do a job. There are a number of ways to make your product better, but encouraging people to get more out of the capabilities you've already invested in is usually an excellent start.
I simply can't stand the way MS Office is always trying to "help" me.
So, ya wanna stir things around, big time?
Come up with a suite of robust, simple, effective software that runs on Linux:
a) Word Processing
b) Spreadsheet
c) Image management and manipulation
d) Accounting
Right now Linux is limited in its use because there aren't many good programs for it ... and nobody writes programs because the distribution is limited.
There are tens of thousands of small businesses who remain with the MS OS and MS Office just because shifting to Linux still cuts them off from a lot of basic programs.
Bart,
You should check out Open Office.
Note that Microsoft recently moved the goalposts with its so-called OOXML default save format (.DOCX, etc.), which Open Office doesn't read yet, though it reads the older .DOC files.
Turkeys (--meaning Microsoft, not the Open Office folks).
I don't offer training on new sales of my software and explain to clients that if they need training they either:
A: Have the wrong employees who don't understand their business.
or
B: I've created some lousy software they shouldn't have bought in the first place.
Go ahead and plug the software with a name and description, Davebo - if that's your policy, you've earned it.
Based on his description, i'm gonna take a wild guess and say he's Mac based. PC people always need training just to make the damned things work in windows.
What about the romanticism of spending hours sending search commands directly to the databases in order to obtain the urgent information you need and the costly program is unable to get?
grep, anyone?
Nortius Maximus-
Open Office supports the regular .doc, .xls and other extensions. We use Office 2007 because we are part of the 'just-outta-beta' group BUT I always end up saving things in the older formats because most folks cannot read the new ones. [sigh]
I am investigating other avenues. Open Office. Thunderbird email client from Mozilla. Linux OS to run some of this. BUT you can have machines under Linux set up as dual boot and rapidly change to Windoze for things like AutoCAD, etc. There are other ways to do things besides the Windoze routes. I do have to say that XP Pro is pretty stable though when treated properly. My boss is always blowing it up and I cannot understand how he does it. But he does.
Have faith, the computing world has come a long way in just a few years.
Dual boot is so 2001. Welcome to Virtualization. VMWare's Fusion is a client-side example of a growing trend on the server side. The effects are different, though, and amount to "Why dual boot when you can just run 'em both?"
Am about to test to proposition on my new Macbook Pro, when it arrives, because there will still be a number of Windows programs I'll be using regularly. Could load and run Linux, too, if I wanted, but OSX already comes with built-in UNIX derived from BSD so I'll probably just train on that... as Evariste has been urging me to do for some time now.
Longer term, the interesting question is what happens as operating systems with different interface conventions are increasingly running on the same machine. Cross-pollination, to be sure, but also some confusion and clash, I suspect. Life is about to get mnore interesting for usability engineers...
Joe, you might want to check out VirtualBox, as another option.
Joe, I agree entirely - but there are a few issues that you ignore, why usability is actually not the largest priority when purchasing software at a larger business.
a. What's the old adage? - you never lose your job if you buy from IBM? Something like that, it is still at work in the corporate world. Buy from some fly by night web 2.0 company? Why risk your job?
b. The enterprise feature set - no data loss, separate storage arrays, "proven" uptime (always a joke, in terms of proving it, of course, but still a cya requirement), failover, point in time recovery, etc - once you start including those features, you are halfway down the road, of not being nimble.
c. Customizability - always a big issue, companies need to customize, either a feature set, or, especially the reporting modules - and successfully built simple software really, doesn't customize, as you need to keep innovating your product, and it isn't easy to manage different implementations (though it has been getting easier, admittedly). So that's why you end up with great solutions like SAP that have the ability to track whatever you need, but take an army to implement, in a detailed, comprehensive way. Not to mention, any company considering moving to new software, usually has to painfully prune/adapt to a necessary existing customized process, that really is horrible software, but nevertheless works for, say - that particular big ordering system.
Now, CAN the issues above be overcome? Yes, of course. Mainly, configurability is being built more and more right into software products, so you get customization without modification, by changing configurability settings. An example would be something like My Google, where there isn't a "customized" software - only saved configurations for each ID. And this is well within the parameters of the software.
But then, you immediately get more complicated as well, from the interface.
The trend now, tends to be towards simpler usability for the users of software, while keeping a more complicated interface for the administrators, who need a different type of access to the different User, Product, and Process fields, to be able to run different reports and institute the hiring processes, ordering and tracking, etc, that are needed.
Hot thing today - Google App Engine that hopes to solve a lot of the backend for you.
Interesting stuff, thanks.