Thriving Through Dis-Automation, Part II

By CIOinsight  |  Posted 06-14-2005

Thriving Through Dis-Automation, Part II

In my last column, I was telling you about Savage Beastand the Westergren/Glaser model.

The model approaches automation issues by establishing what mix of automation and manual work will product the greatest efficiency not on automating everything one might be capable of automating.

It parses lists of functions the way you'd do it with a group of human employees - doling out tasks based on competencies and effectiveness.

Those tasks IT does best should be assigned to it, and only those tasks.

When you get down to the processes that either technology or humans could handle equally well, the idea is to balance quality against cost.

As I've explained before, most projects over-automate because the people in charge of the projects tend to be technologists.

As such, they view automation as a virtue in itself, and tend to prefer the work (and sometimes the company) of machines to that of humans.

A secondary but frequent reason is that the finance types who tend to run American organizations are still in the thrall of the de-staffing fad, the mediocre manager's simplistic choice to increase returns.

Competitive Advantage

You should play around with the Savage Beast model if you're in an organization where mediocrity isn't good enough.

There are giant potential advances in effectiveness to reap by tuning your mix of technology and human work.

And because the unchecked self-preservative tendencies of most people within organizations results in a remarkably consistent pursuit of mediocrity - it's an advantage your competitors are unlikely to match any time soon.

How do you transfer their model to your own line of work? How do you figure out a better way to design processes to balance most efficiently between human and machine?

If automation is a faith in pure technology without people and Luddism is a faith in pure human effort without technology, what I'm urging you to aim for is something I call rationalization.

The Savage Beast model is an analogue to how you should think about basic design of your own staffing and projects.

Rationalization means serving customer needs or wants through human/technology partnerships that are designed to customer and business goals instead of just eructating the same technology everyone else is.

There are several paths to breaking through "the way things are done" to balance your processes more effectively, and the best will differ according to your organizational strengths and weaknesses.

I'll tell you one that works well in many cases, especially if finance has much of a rôle: Zero-Based Automation (ZBA).

In a ZBA exercise, you don't start by taking into account where you are.

You ask the naïve question: "What processes would we automate if we were starting from scratch?"

You walk through each, keeping track of where you probably automated something more effectively done by humans.

(It's possible you've got humans doing work you should have automated for greater effectiveness, and you should note these, too, though I've found in my practice that they're fairly rare).

As with the familiar tool zero-based budgeting (ZBB), the analysis doesn't require an examination of everything in place that doesn't match best practices or current needs you'll dump.

It's an exercise in perspective, because even if you don't do anything about the shortfall, you've recognized it publicly and increased the gravitational field to tweak or replace all or part of the process.

Next Page: ZBA exercises.

ZBA Exercises

ZBA Exercises

Assemble a ZBA team from multiple disciplines that go beyond the disciplines that designed your current systems.

Only a minority of the group should be professional technologists, and you should have at least two members who are cooperative end users who do line work and who aren't good with technology (people who normally need a lot of help, but not professional complainers, either).

Start your ZBA exercise on underperforming pieces of your existing portfolio.

Do any of them fall short because of imperfect design, delivering the wrong information, or the right information in forms that create too much friction to use?

Do any lack of end-user buy-in? Are any of them using technology to do things people have done better? What would you have to do to dis-automate pieces of them or the entire project?

You can add texture to the answers if you interview end users; the cooperative-but-low-tech members can be superb interviewers for this purpose.

Your interviewers, however, have to make sure they don't intimidate informants or make them feel like they have to display understanding when they don't.

Move on to projects that have not yet been set in stone.

Here you have the opportunity to do rationalization in its most purely profitable way especially if your goal is a business-oriented and not just gadget-oriented one.

How do human skills and technology best complement each other to achieve your goal?

Play with the concept of zero-based automation, that is, if you had no existing automation to get you to the defined goal, which pieces are humans intrinsically poor at to a degree that it would be worth inventing technology from scratch to assist or replace the humans?

What pieces do people never improve by their touch but only create the opportunity for degradation? What technology exists that might work as a replacement?

Don't Be Make Yourself Roadkill

Even if you're too timid to be a leader in this unstoppable trend, you can prepare yourself to hop aboard by running these kinds of thought exercises now.

The farther you strip away the concept of technology-as-human-replacement and build instead on the idea of open-eyed realism in creating partnerships between technology and human aptitudes, the higher the potential returns and competitive advantage.

The end of the intensification cycle has a few bits of margin to squeeze out, but they are diminishing returns on diminished principal.

Much of the gain to be harvested is in zones where social pushback will be high enough to undercut the calculated benefits.

A prime example (of both gains and whirlwinds to be reaped) is the British initiative to use prisoner-tracking technology to make sure workers don't take unauthorized breaks and to help them navigate the stock in large, dynamic warehouses.

Those intensifications of de-staffing projects will still exist.

There are still mediocre gains to be had in them for organizations with a strong commitment to mediocrity.

The big rewards will be in rationalizing your technology, and in getting the most out of people, technology and the patterns that bring out the complementary value of the two.

Jeff Angus is a management consultant and has been working with IT since 1974. He has held IT management positions in user interface design, marketing, operations and testing/analysis. Look for his book, "Management by Baseball: A Pocket Reader." Jeff's columns have appeared in The New York Times, The Washington Post, the St. Louis Post-Dispatch and the Baltimore Sun. He can be reached at

Check out's for the latest news, reviews and analysis on IT management from CIO Insight.