One of the easiest ways to make transformation too large is to begin with a true observation: organizational performance is systemic. Growth, productivity, speed, and customer experience rarely depend on one process, technology, team, or behavior. They emerge from the way decisions, workflows, roles, technology, information, capabilities, and leadership work together.

The mistake comes next. If many things influence the outcome, leaders can conclude that many things need to change. They usually don't.

The purpose of understanding the whole system is not to put the whole system into transformation. It is to know where change will matter most. That distinction separates integrated transformation from comprehensive transformation.

Look broadly before deciding what to change

Consider a marketing organization trying to materially reduce time to market. A broad assessment might uncover dozens of legitimate opportunities. Planning could improve. Technology could be better configured. Roles could be clearer. Data could be cleaner. Skills could be stronger. Governance could be simpler. All of those things may be true.

But suppose most of the delay comes from a smaller combination of conditions: work begins before priorities are resolved, decision authority is unclear at two critical handoffs, and late approvals create repeated rework. Those problems are connected. Unclear priorities create exceptions. Exceptions require additional decisions. Slow decisions push reviews later. Late reviews create more rework.

In that situation, a broad technology overhaul or ten-workstream transformation could improve plenty of things without solving the problem that matters. The better transformation would address the few constraints actually holding the outcome in place—and address them together.

That requires looking at more of the organization before deciding to change less of it.

Integration is about fit, not volume

Research on organization design has long treated organizations as systems of interdependent choices rather than independent levers. Changes in structure, processes, technology, coordination, management practices, and other elements can reinforce one another, and the value of one change can depend on conditions around it.

That does not imply that every interconnected element needs to change. It implies that leaders need enough understanding of the surrounding system to know which changes cannot succeed alone. A new technology may require a different workflow to create value. A new workflow may require different decision rights. Those decision rights may need changes in governance or leadership behavior to hold.

The transformation does not become stronger because another workstream was added. It becomes stronger because the necessary changes fit together around the outcome.

This is why integration should create focus. As understanding improves, leaders should become better able to distinguish between what could be improved and what must change.

Every additional change has a cost

This discipline matters because change is not free. Every additional initiative consumes leadership attention and organizational capacity. It introduces new decisions, dependencies, communication, and demands on people who still have a business to run.

A large transformation may sometimes be necessary. But size is not evidence of seriousness, and comprehensiveness is not evidence of rigor.

For every proposed change, leaders should be able to answer a simple question: why must this change for the outcome to move? If there is no strong answer, the improvement may still be worthwhile. It simply may not belong in this transformation.

AI makes that discipline even more important. As organizations gain the ability to automate more tasks, redesign more work, and pursue more opportunities, the number of things they could change will grow much faster than their capacity to change them well. Transformation therefore requires choice, not accumulation.

Find the few changes that matter together

A more focused approach begins with the outcome: what needs to become materially different? Then understand the system producing it. Where is performance actually constrained? Which constraints reinforce one another? What new capability does the organization need in order to produce a different result?

Only then decide what to change. The objective is not the smallest possible program. It is the smallest coherent set of changes sufficient to create the capability the outcome requires.

For one organization, that might include technology, two decision rights, an approval structure, and a new way of prioritizing work. For another, the technology may already be adequate and the real constraints may sit in governance, incentives, or leadership behavior. Different outcome. Different system. Different transformation.

What matters is that each change earns its place by contributing to the result—and that the changes reinforce rather than compete with one another.

This is one of the central disciplines of transformation: understand broadly enough to focus precisely. Change is easy to accumulate. Transformation requires something harder: knowing what matters, what does not, and what has to change together.

More change is not more transformation.

Research foundation

Research on organization design and organizational complementarities provides the foundation for this perspective, including work examining how structures, processes, technologies, coordination mechanisms, and management practices interact to shape organizational performance.