How Do You Build the Plane While Flying It?

Scaling a business has one rather inconvenient feature.

You have to scale it while the business is still running.

Revenue is growing.

You are winning new clients and customers.

Demand is increasing.

Headcount is growing.

Perhaps you are entering new markets or geographies, launching new products or services, or simply doing considerably more of what made the business successful in the first place.

Everything says growth.

To all intents and purposes, everything looks tickety-boo.

The organisation may be telling you something else.

Hairline fractures may already be starting to appear underneath that growth.

A client complains about something that never used to go wrong.

A decision that once took an afternoon now takes a week.

People who once solved problems themselves start escalating them.

Different parts of the business start giving different answers to the same question.

Errors start appearing in work that used to run smoothly.

Decision-making becomes crowded. More people expect to be consulted. More people believe they have a say. Nobody is entirely clear who actually gets to decide.

Work takes longer to move through the organisation.

Cash is tighter than the revenue figures suggest it should be.

Leaders and founders find themselves pulled back into decisions they thought they had delegated.

A few people leave. More concerningly, some of them are people the business has quietly become rather dependent upon.

Your reputation as an employer may start telling a rather less flattering story than the growth numbers.

None of these things, in isolation, necessarily looks particularly alarming.

Together, they may be telling you something rather more important.

The organisation that got you here is beginning to strain.

The business, meanwhile, is still growing, and the plane has already taken off.

You cannot put the business on hold for six months while you redesign the organisation underneath it.

Clients still expect delivery. Work still needs to get done. People still need to be paid. If you have investors, they still expect to see the growth they invested in.

You have to keep flying.

While you do, you have to work out what needs to be built, rebuilt or strengthened.

This is where scaling becomes an organisational design problem.

It probably worked before

A business does not necessarily reach the point where cracks are beginning to appear because it was badly designed or badly run.

Quite the opposite, in fact.

Some of the ways of working that are now beginning to strain may be the very things that helped the business succeed in the first place.

As the business grows, however, the conditions change.

More people become involved. Work becomes more specialised. The business may operate across more markets, geographies, products or services. There are more decisions to make, more dependencies to manage and more work to coordinate.

What once happened naturally between a relatively small number of people now has to work across an increasingly complex organisation.

In a smaller organisation, an extraordinary amount happens informally.

Someone has a question. They know who to ask.

Two people disagree about what a client needs. They talk.

Something goes wrong. Somebody knows who can fix it — and can probably find them fairly quickly.

A decision needs making. The person who can make it may be sitting ten feet away.

Roles overlap, but people understand enough about one another’s work to navigate the gaps.

Information travels through conversations, relationships and shared context rather than formal channels.

People step outside their roles because something needs doing. They can see how their contribution connects to the success of the business.

Leadership can still see much of what is happening across the organisation and intervene when something gets stuck.

A surprising amount of complexity is being managed without anybody necessarily thinking of it as organisational design.

Proximity helps. So do relationships, shared context and visibility. Together, they allow the organisation to coordinate itself remarkably well.

Until they do not.

As the business grows, work crosses more boundaries. Knowledge becomes more distributed. More handovers and dependencies appear. Add new markets or geographies and the network becomes more complex again.

Technology becomes more embedded in how work gets done. It may make some things considerably easier while also creating new systems, workflows and dependencies that have to work together.

People join. People leave — sometimes taking with them crucial tacit knowledge about how the business actually works, including the workarounds nobody ever thought to document.

What once relied on somebody simply knowing who to ask now has to work across teams, systems, locations and increasingly complex networks of work.

The problem is not complexity itself.

It is when complexity grows faster than the organisation’s ability to handle it.

The sequence matters

Recognising that growth is placing different demands on the organisation does not mean everything needs to change.

Nor does it mean replacing everything that worked before with something that seemingly looks more sophisticated, structured or scalable.

Some of the things that enabled the business to grow to this point may be exactly the things it needs to preserve.

Equally, waiting until the hairline fractures have become structural problems can leave the organisation trying to redesign itself from behind the growth.

The challenge is knowing what needs to change, what needs to stay, and when.

That means understanding what made the organisation successful in the first place, what still creates value, and what may only have worked because the business was smaller, less complex or more closely connected.

Imagine a business planning to grow from £20 million to £40 million.

The first question is not how many more people it will need.

It is:

What will this organisation need to be capable of at £40 million that it does not need to be capable of today?

Perhaps it will need to deliver significantly more work without compromising quality or consistency.

Perhaps it will need different technology, stronger financial infrastructure, clearer decision rights or management capability that does not currently exist.

Perhaps the existing operating model can carry much of that growth perfectly well.

Perhaps parts of it are already approaching their limits.

The point is not that all of these things need to be built now. The point is that they need to be sequenced.

There is evidence to support this approach. OECD research following firms in Finland, Italy, Portugal, Slovakia and Spain found that preparation for high growth often began before the growth itself became visible, including investment and changes in how firms operated.[1]

In other words, some of what an organisation will need for its next stage of growth may need to be built before that growth arrives.

That raises a different set of questions.

What needs to be true before the next stage of growth begins?

What needs to be built now because it will take time to develop?

What capability do we need now in order to build what comes next?

What can continue to evolve as the business grows?

What can wait?

What should we deliberately not build yet?

Building too early has a cost too.

Introduce process where human judgement was already sufficient and you can create bureaucracy rather than capability.

Build systems for complexity the organisation does not yet have and you can burden it with infrastructure designed for a scale it has not yet reached.

Wait too long, however, and a different problem emerges.

Growth begins to depend on workarounds, heroic effort and a handful of people who know how everything really works.

At that point, you are no longer building ahead of the growth. You are trying to catch up with it.

Sequence is not about following a predetermined order.

It is about understanding dependencies.

What needs to exist before something else can work?

What has a long lead time?

What can safely evolve later?

What would create unnecessary weight if you built it now?

The question is not what a scaled organisation needs.

It is what this organisation needs next.

Best fit, not best practice

There is no universal design for a scaled organisation.

There cannot be.

Nor is scaling confined to one particular type of business. OECD research using firm-level data from 17 OECD and accession countries found scalers across different sizes, ages, sectors and locations. More than half of the scalers studied had invested before scaling, while another substantial group invested and transformed during the growth period itself.[2]

Even the route to growth differs.

Two businesses of the same size, with similar growth ambitions, may need entirely different things next.

One may have an operating model that can comfortably support considerably more growth but lack the technology to make that growth economically viable, whilst the other may have excellent technology sitting on top of work that was never properly designed in the first place.

One may genuinely need more people. The other may already have enough, but too much of their capacity is being absorbed by duplication, poor handovers, unnecessary approvals or decisions that take too long to make.

One may need more process. The other may need considerably less of it.

This is a question of best fit, not best practice.

As a business grows, there will be no shortage of advice about what it should look like next.

The structures it should have. The processes it should introduce. The technology it should use. The management layers or capabilities it should build.

Some of that may be considered best practice. Some of it may be entirely appropriate for your business. Some of it may not.

Being considered best practice does not necessarily make something right for your business.

Best fit means choosing what is appropriate for the business you actually have and the one you are trying to build.

That depends on where the business is now, where it is trying to go, how it creates value, the clients or customers it serves, its people and capabilities, its markets and geographies, and the expectations of those invested in its success.

What works exceptionally well elsewhere may be unnecessary, unaffordable or simply inappropriate here. In some cases, adopting it may make the organisation less effective rather than more.

The aim is not to make the organisation look like somebody else’s idea of a well-designed business. It is to work out how your organisation needs to operate for the business you are trying to build.

To do that, you need to understand what you are asking the organisation to support.

What will be different as the business grows?

More clients or customers? Different ones? New products or services? New markets or geographies? Greater volume? A different delivery model? An acquisition? Technology changing how the work itself gets done?

Look at the organisation you have today

Once you understand what the growth will require, you need to understand what the organisation can already support.

What should you preserve because it still creates value?

What is already beginning to strain?

Where are people compensating for weaknesses in the operating model?

Where is work taking longer than it should?

Where are decisions getting stuck?

Where are clients beginning to experience inconsistency?

Where are leaders becoming the workaround?

Where is capacity being absorbed without creating corresponding value?

Where are cash flow, margin, rework or write-offs telling you something about how the organisation is operating?

Where are you adding people, process or technology without being entirely clear what problem you are trying to solve?

These are diagnostic questions. The answers tell you what is happening in the organisation today.

They establish your baseline: what is working, what is beginning to strain and what the organisation can already support.

Only then can you look forward and ask what the organisation will need for the next stage of growth.

Look at what the organisation will need next

The questions now change from what the organisation can support today to what it will need to support tomorrow.

What capability will be required?

How will work need to flow?

Where will decisions need to sit?

What will managers need to be capable of?

What technology and infrastructure will the business depend upon?

What should disappear because it no longer serves the organisation you are becoming?

This is where the distinction between design and addition matters.

A new system is not necessarily better design.

Another management layer is not necessarily greater capability.

A new process is not necessarily greater control.

Another twenty people are not necessarily twenty people’s worth of additional capacity.

The system is being built either way.

Every hire, process, system, workaround, reporting line and decision changes it.

The choice is whether those things accumulate around the organisation as it grows, or whether they are deliberately shaped around what the business is trying to become.

That requires judgement.

Not a maturity model telling you what an organisation of your size ought to look like.

Not somebody else’s scaling playbook.

Certainly not a benchmark that says you have reached 100 people and are therefore due another layer of management.

The right operating model is not the most mature one.

It is the one that best fits what the organisation needs to do next.

You cannot design the operating model from the boardroom

Working out what the organisation needs next requires information that no single person has.

Founders and leaders know where the business is trying to go, the commercial choices being made, the capital available, the risks the organisation is prepared to take and the outcomes it needs to create.

People doing the work see something different.

They see where work is harder than it needs to be, where clients repeatedly ask for something the organisation struggles to deliver, where information gets lost, where work is duplicated and where people have created workarounds simply to get things done.

They also know where the organisation depends on particular individuals because important knowledge, relationships or ways of working have never travelled beyond them.

Information from outside the organisation matters too.

Clients change. Markets move. Competitors behave differently. Technology changes what is possible. Regulation changes what is permissible.

What worked yesterday may cease to be the right answer because the world around the organisation has changed too.

None of those perspectives, on its own, tells you how the operating model should work.

The direction of the business has to meet the reality of how work gets done and the conditions in which the organisation is operating.

Otherwise, you risk designing an operating model around an incomplete picture of the business.

Adding things is not the same as designing

As organisations grow, there is a natural tendency to add things.

More people. More managers. More systems. More processes. More meetings. More controls.

More things invariably mean more complexity.

Sometimes that complexity is necessary. Sometimes it is not.

The question is whether what you are adding solves the problem you actually have.

If accountability is unclear, adding another management layer may simply give the confusion somewhere else to go.

If work is poorly designed, adding more people may increase capacity without improving productivity.

If information does not move properly through the organisation, another system may simply digitise the problem.

Adding something can make the symptom disappear without fixing what caused it.

Worse still, the addition itself becomes part of the organisation — another role, process, system, meeting or control that creates its own dependencies and complexity.

Scaling requires the right infrastructure. It does not require accumulation.

The question therefore becomes whether you are building capability or simply adding complexity.

Is what we have built still right for what comes next?

Even a well-designed operating model has a shelf life.

The business changes.

Clients change.

Products change.

Technology changes.

People and capability change.

The economics of the business may change.

The assumptions on which earlier decisions were made may no longer be true.

That does not mean the organisation needs to be continually redesigned.

However, it does mean it needs to be regularly reassessed.

Is the operating model still supporting the way the business needs to work?

Are decisions being made at the right level?

Are people spending their time where it creates the most value?

Are systems making work easier or creating work of their own?

Are processes still solving the problem they were designed to solve?

Has something that was deliberately kept informal now reached the point where it needs more structure?

Has something introduced to create control now become an unnecessary constraint?

Where is the organisation beginning to strain?

Where are people compensating for something the system should be doing?

Where has the business changed without the organisation changing with it?

Some of those questions will produce no action at all.

Good.

The purpose is not to keep redesigning the organisation.

It is to keep seeing it clearly enough to notice when the organisation you built for one stage of growth has become the constraint on the next.

Eventually, the question changes.

It is no longer:

What do we need to build?

It becomes:

Is what we have built still right for what comes next?

Work out the sequence

Scaling requires a series of decisions about what the organisation will need, what it already has and what needs to change.

Those decisions need an order.

There is no universal sequence. What needs to happen first will depend on the business you have, the growth you are trying to create and what the organisation needs next.

A suggested approach might look like this:

DIRECTION

↓

OPERATING MODEL

↓

CAPABILITY

↓

SEQUENCE

↓

SEE WHAT HAPPENS

↓

ADAPT

Start with direction

The starting point is the direction of the business. Growth is not simply a question of doing more of what the organisation already does. It may change what the business delivers, who it serves, how it operates and the capabilities it will depend upon.

Understanding those ambitions is essential because the organisation needs to be designed around the business it is becoming, not simply the one it is today.

What growth are we trying to create?

What will be different about the business if we succeed?

More clients or customers? Different markets? New products or services? Greater volume? Different economics? A different way of delivering the work?

Look at the operating model

How will the organisation need to work to support that growth?

What should remain exactly as it is because it creates value?

What needs to work differently?

How will work flow?

Where should accountability sit?

Where does judgement need to sit?

How will decisions be made?

What will need to work across the organisation that currently depends upon proximity, individual relationships or somebody knowing who to ask?

Identify the capability required

An operating model is only as effective as the capability available to support it. The challenge is to establish what the organisation already has, where the gaps are and what it will need to build or acquire as it grows.

What will the organisation need to be capable of?

What capability already exists?

What is missing?

What needs to be created, bought, developed or redeployed?

Where are we underpowered?

Where are we already carrying capability we may no longer need?

What capability do we need now in order to build what comes next?

Sequence it

Identifying what needs to change is only part of the challenge. The order in which those changes happen matters just as much. Some decisions create the conditions for others to succeed, while getting the sequence wrong can introduce complexity before the organisation is ready to manage it.

What needs to happen when?

What needs to exist before something else can work?

What has a long lead time?

What needs investment now?

What can evolve as the business grows?

What should deliberately wait?

Sometimes that will mean adding people first.

Sometimes technology.

Sometimes management capability.

Sometimes financial or operational infrastructure.

Sometimes redesigning the work before adding anything at all.

There is no prize for building everything early.

There is also no prize for discovering six months too late that something you now urgently need takes twelve months to build.

See what happens

Making changes is one thing. Understanding whether those changes are strengthening the organisation is another. Implementation tells you what has changed. Performance tells you whether those changes have made a difference.

Is the organisation actually becoming more capable?

That question matters because organisational capability is not separate from performance.

International research across firms and countries has consistently found an association between management and organisational practices and outcomes including productivity and profitability.[3]

That does not tell you what this organisation should build next.

It does reinforce the importance of looking at whether the way the organisation is being managed and organised is actually producing better outcomes.

Can it handle more work without simply creating more activity?

Are decisions becoming easier to make or harder?

Are people becoming productive more quickly?

Is performance improving?

Is the client experience holding?

Are the economics working?

Are leaders gaining capacity or being dragged further into the machinery?

Growth itself provides information about the design.

Use it.

Adapt

Scaling is not a one-off design exercise. As the organisation grows, its needs, pressures and capabilities change. Decisions that were right at one stage may need to be reconsidered at another. Adaptation means recognising those changes without automatically dismantling what is already working.

Is what we have built still right for what comes next?

Keep what works.

Change what does not.

Build what is becoming necessary.

Remove what has outlived its usefulness.

Sequence the next stage around the organisation you now have — not the one you had when you made the original plan.

The sequence matters.

It is not always the same.

Keep flying

There is no point at which the organisation politely stops growing while you redesign it.

The plane keeps flying.

The job is not to build the perfect organisation before the next stage of growth arrives.

It is to understand what that growth will require, recognise what the organisation can already carry and build what it will need next — early enough to matter, but not simply because somebody else’s scaling playbook says it should be there.

Some things will need protecting.

Some will need changing.

Some will need building before you need them.

Others should be left alone.

Occasionally, something you carefully built for the last stage will need dismantling because it has become a constraint on the next.

That is an uncomfortable reality of scaling.

You are designing an organisation that is already moving.

You will never have perfect information.

You will not get every decision right.

By the time you have built some of it, the conditions around it may already have changed.

So perhaps the skill is not learning how to build the plane while flying it.

It is knowing what part of the plane you need to build next.