Imagine we're talking with a visitor from outer space, one sent to help us assess our situation and what's required for the future.

Our guest asks first about the big picture: Over time, what has most changed the human condition?

In exploring this, we soon agree that the big changes started with new knowledge. We learned about -- and then took advantage of -- bronze, coal, steel, antibiotics, hybrid corn, atomic energy, the net (TCP/IP) and the web (html).

Early learning was about how the world works: Will the wheel make it easier to move things? Will the bridge hold the weight? Will the engine pull the train? Will the code compile and compute?

We've long needed people good at such problems: puzzles with clear right or wrong answers. We usually recognize good puzzle solvers as "smart" when they are quite young. People smart in this way have tended to become mathematicians, scientists, engineers, computer programmers.

I see, says our guest. You need to be smart about things you can use -- tools -- as it is tools that let you expand your capabilities.

But is that the same kind of "smart" needed to apply those tools?

Well… No, not exactly. The problems and smarts to apply tools are not the same as those to build them.

In general, problems of application are social more than physical. New applications require people to change their behavior and division of labor. To fully apply plants, plows, and other tools of agriculture, we move from hunter-gatherers to farmers. To apply steel and steam engines, we produce at greater specialization and scale, organizing at the level of nation and even continent. Powerful new tools require people to change what they do and who they do it with.

These problems are not puzzles with provably right answers. They require deep involvement and judgment via interpersonal communication and negotiation. They require street smarts about people and politics -- a different kind of intelligence.

I see, says our guest. You need first to be smart about things -- to invent tools -- and then about people -- to apply tools successfully.

Well then, given where you are now, what is the right balance of smarts for the road ahead? Do you have what you need, or will you need a new balance?

Important questions, eh?

In a networked world, as we move beyond the delivery of online services to redesign and transform our globally interconnected value chains, what do we need in the way of smarts? Are we smart enough to know what we need and how to get it?

Let me know your answers.

We'll be exploring this and other questions at our Leadership and Strategic Management for Chief Information Officers Program (June 17-18 at Harvard). Here's where to learn more and apply. Hope to see you there.

Regards,

Jerry

Among my first "real world" assignments after graduate school was trying to improve productivity in the Sanitation Department of Mayor John V. Lindsay of New York City.

At that time, the typical garbage collection crew loaded 6 tons into their collection truck, then did something else as the truck traveled to the dump and back. They then continued loading for an average of one additional ton per crew.

But wait. Trucks were available that held 8 tons. We tested them to see if they were too big to maneuver in city traffic… and in almost all cases they worked.

So, how would you estimate the expected productivity improvement? And how would you get it?

We thought we could turn most of the extra truck availability into collection work. The Budget Bureau announced an expected productivity improvement of roughly 15% (from 7 tons to 8 tons per crew per day).

But the next year, as we prepared the budget, we had actually gotten no productivity growth.

Zero. Nothing.

Over the next several years, however, we boosted productivity measured by tons per truck shift by more than 20%.

How? Many things were brought to bear: legalizing garbage bags rather than cans (they weighed less); bringing in special trucks with automated container loading for large locations; improving truck maintenance; etc.

But the biggest factor by far was Walter Scharaga, the Cleaning and Collection Bureau Superintendant from the Bronx. Walter was tough. More than that, he was respected as fair by the Sanitation Union, whose prime complaint was inconsistency and unfairness that could creep in as the routes were extended.

We put together a Task Force with Walter in charge, moving it from district to district, laying out explicit new collection routes throughout the city. This changed the assignments and accountabilities for over 1400 peak-day (Monday) collection crews. District by district we laid out new routes, measured the results, reported publicly what we found, and moved on.

It wasn't rocket science, by any means. But it was very effective.

Our lesson was that technology typically enables productivity improvements, but harvesting requires the workflow to change, and this requires leadership to move the group from one pattern of assignments and behavior to another.

As a simple formula, this can be expressed as:

ΔT + ΔW + ΔL = ΔV (where T is technology, W is work, L is leadership, and V is value as productivity)

Years later, the key technology is digital data, computing, and networks, not bigger garbage trucks. But it's still true that we can't harvest results without changing the workflow. And we can't change the workflow without effective leadership.

This may seem obvious, and is. But it's all too often ignored.

If we need all three factors working together -- ΔT + ΔW + ΔL -- then what do we really need CIOs to be good at today? And how can we help them become more effective?

A short version answer to this question appears in today's Governing Magazine HERE.

We'll explore these issues in more depth in a June workshop at Harvard described HERE.

Let me know your answers to the "How do we harvest productivity?" question. I need new answers (and not all based on garbage).

We hope to see you in Cambridge in June.

Best regards,

Jerry