An Agile Playbook: Play 4: Agile Management

Book extract from "An Agile Playbook", available on Amazon as Print or eBook .

?<<< Previous Chapter ??---------------------------------- Next Chapter >>>

Build projects around motivated individuals.? Give them the environment and support they need, and trust them to get the job done. - Principles behind the Agile Manifesto

“Scientific Management” was a term invented in the early 20th century.? We imagine that the modern technology revolution is unique.? Yet nineteenth century automation had similarly transformed UK business.? The Industrial Revolution lay a century earlier.? By the mid nineteenth century the UK manufactured half of all the world’s cotton cloth, despite not growing a single cotton plant in the UK.?

The transformation came from the efficiency of production.? An 18th century worker would require around 500 hours to spin a pound of cotton.? At the start of the twentieth century a single worker could process two pounds of cotton in an hour.? That represents roughly a thousand fold improvement.? This is perhaps the equivalent of microprocessor speed increases from the 1980s to today.

Those efficiency gains had come from technology improvements.? Automation, steam power and processing many threads in parallel with one worker all contributed.? Large, factory-based organizations were appearing and needed new approaches. The management of business, it was argued, could be treated scientifically no less than any other area.?

Frederick Winslow Taylor was a foreman in the Midvale Steel Company.? Taylor felt that detailed study of how workers performed their tasks could provide the basic information for scientific analysis.

Taylor’s theory was that every task would have one identifiable best method for performing it.? Rational and scientific analysis could identify this method.? Most critically he believed in the creation of a class of managers.? These would define and document the correct method for the work.? Managers would be completely separated from workers.? The workers would then exactly apply the techniques the managers had defined.

Each man receives … complete written instructions, describing in detail the task which he is to accomplish, as well as the means to be used in doing the work. - “The Principles of Scientific Management” F.W. Taylor 1911

This is a reductionist viewpoint.? Taylorist theory is built on the idea that work is defined and repeatable.? If the work is repeatable, then we can build a set of rules for completing it in the best way.? If it is complicated, we can break it down into simple steps.? For each step we can write detailed instructions.

At the time, Taylor’s theories were a significant step forward.? He focused on removing “common sense” in favour of studying the work to be done.? He argued against assigning workers randomly but instead matching the work to their skills.? And he wanted to ensure that the requirements for the work were made clear to the workers.

But for some work, scientific management is flawed.? And regrettably these flaws persist to the present day.? Taylor assumed that work is repeatable.? Therefore there is a “best” way of performing the work.? And that best way is found by looking at identical work and analysing it.? We have seen that technology work is instead complex.? In complex environments, new work is not exactly like the old work.? Some of the knowledge is emergent and comes from doing the work. More deconstruction and analysis will not make the problems solvable.

Inherent in the concepts of Scientific Management is the separation of managers and workers.? This is an idea which has persisted from Taylor’s time to the present day.? Many organizations still focus on maximising this separation.

In most cases one type of man is needed to plan ahead and an entirely different type to execute the work. - “The Principles of Scientific Management” F.W. Taylor 1911

The reality of the technology world tends to be different.? Successful technology companies have found that managers and individual contributors do specialise in different skills.? And both skillsets are valued.? But the managers cannot plan and write instructions in isolation.? Critically, the successful technology companies have understood the emergent nature of complex work.?

People are highly variable and non-linear, with unique success and failure modes. Failure of process and methodology designers to account for them contributes to the sorts of unplanned project trajectories we so often see. - Alistair Cockburn

Many organizations have realised that the key mind shift which is needed is to put the individuals, not the managers, at the heart of the organization.? Individual contributors generate value for the business.? Managers enable and multiply that value generation.

Managers serve the team. The default leadership style at Google is one where the manager focuses not on punishments or rewards but on clearing roadblocks and inspiring her team - Laszlo Bock

An hierarchical approach causes major problems when we look at decision making.? In the “Scientific Management” model, decisions are made exclusively by managers.? It is assumed that they are the best decision makers. Perhaps they are more experienced or more highly trained.

Let’s imagine there is a new feature request from a customer.? A decision needs to be made about how to proceed.? In a Taylorist model, the team will escalate this to their manager.? They may need to discuss this with other managers, or with their own manager.? This will need meetings to be set up.? These typically occur on a weekly cadence, so it is not unusual for this to take several weeks for a decision to be made.

On average, respondents spend 37 percent of their time making decisions, and more than half of this time was thought to be spent ineffectively. - “Decision making in the age of urgency” McKinsey

In the modern technology world, we rely on rapid feedback.? Feedback needs to be faster than change.? If we are riding a bicycle, you must respond to leaning by turning the handlebars.? If the response time is too long, you will fall. In a complex environment you cannot wholly predict how your actions will affect the outcome.? You will need to respond and then adjust your response according to the effect.? Feedback becomes even more tightly coupled.

Escalating responses up a hierarchy is too slow an approach.? Worse still, managers are not typically the people with accurate, current data.? There is not only a delay in the response, but there is loss of detail in the data which is being used for the decision.? The local teams understand the local situation while abstracted management are unlikely to do so.

We need a decision making process which is tailored to this fast moving environment.? Decisions need to be focussed and rapid.? We need to minimise the time and effort lost from the decision process.

I only wear gray or blue suits. ?I don’t want to make decisions about what I’m eating or wearing. Because I have too many other decisions to make. - President Obama

A simple hierarchy is not going to work well for our technology company.? Successful technology organizations are moving to more distributed management systems.? We aim to make decisions quickly and efficiently.? This is done by taking the decisions at the point as close as is possible to the situation being decided.

Taylorist approaches aim to escalate decisions and typically tend to make them at the highest possible level.? Technology companies tend to make decisions at the lowest viable level.? Every hierarchy loop we avoid saves time and gives more rapid feedback.? It means that the people with the latest information are at the heart of the decision.

Pushing more decisions to the team requires us to rethink some areas of traditional management.

• Managers need to step back from decision making.? They must avoid over-asserting their views.? Otherwise teams will tend to continue to ask for approval.? The manager will need to encourage independent decision-making among team members
• Teams need to be empowered to make decisions.? They need to be taught which decisions are appropriate and which should be escalated.? This requires coaching and mentoring from the managers.
• In order to make decisions quickly at the team level, the team need to be kept informed.? This means that managers will need to communicate well and give more contextual information to the team.
• If decisions are not being taken in one place, team decisions will need to be broadcast to managers and to other teams.
• With more decision makers, we cannot rely on “innate knowledge” of past decisions.? This means that teams and managers must keep better records such as decision logs to record the key decisions being made.

Managers need to make sure that the teams and individuals are supported.? They do not have to perform the services themselves.? Indeed it is unlikely they will be the ideal person to perform every service.

Management acts as a venture capitalist … we open up our purse but keep our mouth closed - Takeuchi and Nonaka

About Jay Alphey

I have built and led technology groups in multiple high growth scaling organisations, working across the organization to coach teams and develop and scale the structures and practices which allow the teams to thrive and achieve repeatable delivery of business value.          

I am available for both consulting and full-time VP Engineering/Operations roles in Cambridge and London.