Digital transformation doesn’t work. This is the lapidary statement that we often hear and which is not lacking in foundation.
It doesn’t work because like any change, any transformation, it is accompanied by fears and faces cultural brakes.
It doesn’t work because digital has often been put before the transformation.
It doesn’t work because where it worked it had a different name and digital was only part of an overall strategy.
It does not work because where it is implemented its results struggle to convince. Not because there aren’t any, but because they disappoint.
The results of change disappoint. Always.
Here we are facing a constant. The subject of this post is digital transformation since the subject is (still?) topical but it is a phenomenon as old as the world, recurrent in any process of change, transformation and a fortiori when it involves technology.
Businesses have invested and are still investing a lot, for example in collaborative solutions, which are more than ever a hot topic. Is the ROI promised by vendors being achieved? Rarely. Does information flow better? Not always. Is knowledge better shared? No. Have employees become more productive? No. Have we avoided spending hours in unnecessary meetings? Even less so.
And yet in theory it works and the benefits are obvious. The technology is beautiful, the use cases are well defined, the tests have validated them. So where does the problem come from?
What exactly is the purpose of technology?
Before going any further, we should ask ourselves what exactly is the purpose of the technology. What is its promise? To put it simply, I often say “go faster and operate on a larger scale“.
Choose the technology you want, it fits into this framework.
But the technology is not self-supporting. Its use is contextual. It accelerates work only if we do not deliberately decide to slow it down. It only allows you to operate on a larger scale if you don’t build walls around the operators.
It is often said that deploying technology requires rethinking the organization and this is often true. But when we touch the organization we are always afraid to touch a load-bearing wall, to break things, so we are often very (too?) careful. Moreover, we often mistakenly believe that everything is load-bearing.
But very often we are scared for nothing. There is no question of touching a load-bearing wall, just a decorative plasterboard partition. The kind of thing that was useful at one time but is now useless and that would give much more volume to a room if we got rid of it, without endangering the structure of the building!
To put it another way, what is blocking in many cases are not structural elements of the organization but band-aids that were put in place at one time and are no longer needed today. But they are so much a part of the landscape that we hesitate to touch them.
1 constraint = 1 patch
Let’s keep in mind the idea of operating faster at a larger scale, which is nothing more than what companies have always wanted to accomplish from the outset. They have always faced constraints preventing them from achieving it.
For example :
- Distance or lack of common availability slows down exchanges.
- Lack of skills which slows down processes or reduces quality.
- The impossibility to reach all its customers directly which lengthens the distribution chains.
- Lack of time, a slowing factor.
So companies have found palliatives, patches to cope with these constraints. To get around them.
Distance? They preferred colocated work, which slowed down multi-site cross-functional projects.
Lack of skills? They added control, reporting or recruited more qualified people.
The impossibility to reach customers? Well, they went through distributors. And as it has been going on since the dawn of time, we even forgot that we could do otherwise.
Lack of time? They hired as much as possible.
Almost 20 years ago I remember one of my first missions in a large family SMB that wanted to go further in its “computerization”. Understand: not to be satisfied with office automation tools but to tackle business and management activities.
They had set up a CRM, invested time and money in setting it up, took the time to train everyone. The result: no efficiency gains. I’m not necessarily talking about the turnover, which depends on many other factors, but really on the execution of the process. The sales proposals did not come out any faster, so, with constant resources, they were not processing any more.
It didn’t take long to find out why (and if you’re wondering why sometimes one pays expensive consulting firms to find answers that are right in front of one’s eyes it’s precisely because their new eye allows them to see aberrations that we find so normal that we don’t notice them anymore) and thinking about it again is almost laughable.
15 years ago this company had experienced a great international expansion. Problem: the management positions were held by the founder and some of his sons, who all had one thing in common, which was their inability to express themselves in a fairly good English. Including commercial management. And most of the sales people in this regional company were no better. That was the constraint.
The palliative was simple to implement. Recruitment of a bilingual sales assistant who reviewed all the commercial proposals that were sent. Logical.
Over time things changed. The founder retired, the grandchildren took on many responsibilities. And the company had grown and gradually increased its recruitment requirements. All this to tell you that English was no longer a problem at all.
But the sales process still included the control of the sales proposals by the assistant who, with the growth of the activity, had become the new constraint of the system (needless to say that it was no longer the same assistant as 15 years earlier). No matter the technology, as long as it was at the end of the process, it alone cancelled all the investment made.
But the story is not over.
We presented our conclusions to the company fairly quickly (be assured that we had found other areas for improvement). And guess what the management’s reaction was: “we’re going to have to recruit more assistants and that’s going to be expensive”. Our answer: “No, you don’t need any more assistants to do that anyway. Remove this role and move the person to a sales position”.
Their reaction was somehow logical: with time they had lost sight of the reason for the presence of the assistant at the end of the chain, forgetting that she was only a “patch”, a response to a constraint that had since disappeared. She was part of the landscape and was taken for a load-bearing wall because she had always been there.
The lesson to be learned is that it is a mechanism that can be found absolutely everywhere and that continues to produce its effects:
- A company identifies a constraint
- It sets up a palliative to cope with the constraint.
- The constraint disappears but the palliative remains
- The palliative becomes the new constraint
Between the beginning and the end of the story it can happen 5, 10, 20, 50 years. In other words, at some point those who must lead the change, if they don’t ask themselves the question, don’t know why the palliative is in place and therefore don’t ask themselves the question of its usefulness or not.
Businesses are cemeteries of vanished constraints
If one tried to list everything that prevents a change supported by a technology from delivering its full potential a life would not be enough. But here are two that are often find.
There is no denying that in spite of the wide diffusion of information and communication technologies, not everything is going as fast as one might expect and there is no impression that people have saved time from tasks with little value to devote themselves to tasks that have more.
The cause is easily pointed out: the complication of organizations which means that even if information goes faster in theory, it cannot go faster than the capacity of humans to make decisions, elaborate and validate information.
But where does this abundance come from, this constant inflation of coordination and control structures (partial explanation of course, there are many reasons, such as, among others, the poor implementation of delegation mechanisms).
As far as coordination structures are concerned, operating procedures were put in place at a time when instant group communication did not exist. Today we have Slack, which despite its shortcomings allows us to function differently (and there were other things before Slack). But we have kept most of the structures.
Regarding the obsession with control, we have to go back even further, especially to Taylor. At the time, we had a workforce with little or no qualifications and therefore the need to supervise it and to think about work in its place. The result was tasks that were oversimplified and a strict separation between design and execution activities, with control activities in between. Logical and indispensable.
Who can tell me that today, and even more so in the so-called ” knowledge activities “, we are supervising a workforce with little education and incapable of thinking about what it does and how it does it? Basically, we recruit people with five years of higher education even for the simplest tasks and we impose on them a model initially designed for almost illiterate people.
The constraint related to competence has fallen but we have kept everything we did to accommodate it.
Two more things before I finish:
- Even if things have improved (thanks to the pandemic?) the friction between traditional retail and digital commerce within the same company comes from the fact that at the very beginning of e-commerce it seemed “abnormal” because distributors were part of the model by essence. In fact the very existence of distribution networks as we know them comes from the fact that “Direct to Consumer” was impossible when these companies were born. And we took a palliative for a natural law.
- Less politically correct to write, but just as real: when a lack of time or production capacity (which amounts to the same thing) has been compensated for by recruitment, we will only obtain 100% of the benefits of technology (whether we are talking about machines or software) when the corresponding jobs have been eliminated (or transformed into jobs corresponding to a real need).
Today’s problems are the solutions to yesterday’s problems.
When I try to introduce a new way of doing things and the technology that goes with it, it is frequent and normal that I come up against a reflex to protect the operating procedures in place.
When I ask “why do we do it this way” the most frequent answer is “we’ve always done it this way“, as a way of saying “we don’t know”.
Most of the time when we dig we find that there is a very good reason for the way it is done. Except that this reason no longer exists and that we have to carry as a burden the answer to a problem that has disappeared. Do you want to continue to wear a cast once your fracture is resolved? I think you get it now.
Most of the obstacles that prevent changing operating methods today or that prevent deployed technologies from fully producing their effect are not “native”, indispensable things inherent to the nature of the organization or its activity. They are just bandages that had their raison d’être at one time but have become useless and, worse, have become sources of infection.
Today’s problems were yesterday’s solutions. There is no longer any reason to sacralize them when their raison d’être has disappeared.