In our Big Shift world, we confront the imperative of institutional innovation – shifting from institutional models built on scalable efficiency to institutional models built on scalable learning. I’ve written and spoken about this a lot over the years and one of the most common pushbacks I get is – “so, are you against efficiency?” This post seeks to answer that question.
Let me lead with the answer – no, I’m not against efficiency. I’m instead suggesting two things. First, we need to broaden our horizons to look beyond efficiency. Second, while efficiency is still important, our existing ways of achieving efficiency are becoming increasingly inefficient
Broadening our horizons
I’ve written about the paradox of the Big Shift. On the one hand, it’s creating exponentially expanding opportunity. On the other hand, it’s generating mounting performance pressure. Our imperative is to find ways to harness the expanding opportunities while responding to the mounting performance pressure.
While some of the opportunities created by the Big Shift involve ways to become more efficient, those are actually a very modest segment of the overall opportunity. The opportunity that’s exponentially expanding involves the opportunity to create and deliver far more value to markets and society. As one small example, just think for a minute about the potential that biosynthesis has to extend lives and dramatically increase the quality of our lives.
We need to become much more aware of the new capabilities that are becoming available to deliver far more value and find ways to harness those capabilities so that we can have a much more positive impact on those around us and, in the process, create more value for the institutions that are delivering that impact.
If all we do is focus narrowly on efficiency, we’re going to miss most of the opportunity that’s emerging. Here’s the thing – while efficiency is important, it’s ultimately a diminishing returns proposition. The more efficient we become, the longer and harder we’re going to have to work to get that next increment of efficiency. That’s not an exciting future.
On the other hand, if we focus on addressing exponentially expanding opportunities to create more value, the sky’s the limit, especially if we find ways to pursue leveraged growth. For the first time, we have the ability to tap into an increasing returns curve where the value increases exponentially over time. Who wouldn’t want that?
But to tap into that opportunity, we need to find ways to learn faster – at scale. That’s why I put so much emphasis on the imperative to move from scalable efficiency to scalable learning institutional models. These emerging opportunities are new opportunities, never encountered before. We need to rapidly learn what these opportunities are and what approaches will be most effective in addressing the opportunities. And the opportunities will be rapidly evolving, so we can’t just learn once and be done with it – the challenge is to learn more and more rapidly at scale, forever.
And, just in case it’s not abundantly clear, the learning I’m talking about here is a very different form of learning relative to the learning that we do in most institutions today. All large institutions have learning programs, but these programs focus on transmitting existing knowledge. We go into a training room and listen to an “expert” who shares what they know.
The learning that’s going to be necessary to harness these opportunities is learning in the form of creating new knowledge. That kind of learning isn’t done in a training room – it’s done in the work environment where workers are confronting new opportunities to create value on a daily basis. As I’ve explored in a recent research report – Redefine Work – this involves redefining work for everyone – rather than executing tightly specified routine tasks, we need all workers to address unseen problems and opportunities to create more value. If we take that seriously, it involves cultivating a whole new set of practices and redesigning work environments from the ground up.
Achieving efficiency differently
So, we need to expand our horizons to target opportunities to create more value. Does that mean than efficiency no longer matters? Of course it matters. In a world of mounting performance pressure, we need to strive to become more and more efficient in everything we do.
But, how do we become more efficient? The traditional approach to scalable efficiency was very clear – we tightly specify every task and standardize it so that it’s done in the same efficient way throughout the institution. We tightly integrate those tasks, removing all the inefficient buffers between them, by crafting highly efficient end to end business processes.
This approach worked very well throughout the past century, providing a foundation for large, global institutions that generated enormous wealth. But, now the world is changing, and that traditional approach is paradoxically becoming less and less efficient.
The traditional approach to scalable efficiency worked very well in more stable environments. That’s what made it possible to specify in advance, standardize and tightly integrate. But, how well does that approach work in a world that’s more rapidly changing, with more and more uncertainty and unexpected events that no one anticipated?
Not very well at all. In fact, based on surveys we’ve done, we found that 60 – 70% of worker time in large companies is spent on “exception handling” – dealing with unexpected situations that can’t be handled by the existing processes and that require workers to scramble around and improvise to find ways to address something that’s never been seen before. And they do that very inefficiently, precisely because the organization hasn’t been designed to address these exceptions. In fact, there’s a strong tendency to try to deny that these exceptions even exist, because they call into question the scalable efficiency model.
If we want more evidence that the scalable efficiency model is fundamentally broken, perhaps our analysis of the performance of all public companies in the United States from 1965 until today provides an important indicator. We used return on assets as our measure of performance and it turns out that ROA has basically collapsed over this time period – declining by 75%. There’s no sign of it leveling off, much less turning around. So much for scalable efficiency.
So, efficiency is still very important, but the approach to become more efficient is fundamentally changing. Rather than tightly specifying and standardizing all tasks in advance, we need to create environments and practices that help workers to address unseen problems and opportunities whenever and wherever they emerge. And we need to find ways to “loosely couple” all those tasks so that workers have more room and ability to improvise and test new approaches without creating unintended ripple effects that cascade into some major disaster further down the process chain.
In fact, it turns out that the scalable learning model that I’ve advocated is far more efficient in a rapidly changing world than the scalable efficiency model. By creating environments that help workers to see, address and learn from unexpected situations, we will build much more efficient institutions. Efficiency is still essential – it’s just that we have to pursue very different approaches to achieve it.
Models don’t mix
I hope I’ve persuaded you that I am very committed to efficiency as well as learning – in fact, I see them as increasingly tightly integrated. Scalable learning will produce more and more efficient institutions.
On the other hand, I hasten to add that I'm not a proponent of trying to embed both scalable efficiency and scalable learning models within the same institution. I know there's a lot of interest in the “ambidextrous” organization, where one part of the organization executes routine tasks and another part pursues innovative and creative initiatives, but count me as a deep skeptic of this idea for two reasons.
First, as I’ve suggested, in a rapidly changing world filled with uncertainty, I question whether routine tasks are even feasible, much less helpful. To the extent that routine tasks are necessary, I've made the case that they will be quickly taken by robots and AI – they shouldn’t be done by humans. My belief is that all workers should be focused on addressing unseen problems and opportunities to create more value.
Second, my experience is that these two institutional models are fundamentally incompatible and that scalable efficiency quickly and “efficiently” crushes any attempt to foster innovation and creativity. The scalable learning model that I’ve proposed as a replacement for scalable efficiency requires a fundamentally different culture and way of organizing and operating that is viewed as deeply threatening, or at the very least as a distraction, by those who hold on to the scalable efficiency model. I have yet to find an institution where these two models co-exist at scale.
A choice needs to be made, but it’s a choice in institutional models, not a choice between efficiency and more value creation. The scalable learning model delivers both greater efficiency and more value creation.
The bottom line
We don’t need to choose between greater efficiency and more value creation, but we do need to choose between institutional models. We also need to choose to expand our horizons and move beyond a narrow focus on greater efficiency.
This is not just an opportunity – it’s an imperative. In a world of mounting performance pressure, if we just focus on efficiency, we’ll ultimately be squeezed out of existence. Those who will survive and thrive in the Big Shift world are those who aggressively pursue both value creation and efficiency. And the key to both, in a rapidly changing world, is to learn faster at scale.
Reminds me of Stafford Beer.
Especially this part:
"In fact, there’s a strong tendency to try to deny that these exceptions even exist, because they call into question the scalable efficiency model."
Like when he talks about the "Shoot the cat" instinct in Designing Freedom.
http://ada.evergreen.edu/~arunc/texts/cybernetics/beer/book.pdf (like page 5. With nice hand-drawn diagrams!)
Posted by: Matthew Alhonte | March 31, 2019 at 09:11 PM