The Hidden Cost of Overdesign

Nobody ever got criticised for adding a bit more capacity.

A larger air handling unit. A bigger Air Source Heat Pump. Additional electrical allowance. A more conservative cooling load. A little extra pipework capacity. A slightly larger riser. A plantroom that quietly grows because every discipline, every stakeholder throughout every stage of the process has added their own margin “just in case”.

Individually, these decisions can look sensible. Collectively, they can become a serious problem.

Certainty Tax: When caution becomes accumulation

Overdesign is one of the most persistent cultural issues in building services engineering. It is rarely malicious. It usually comes from understandable places: incomplete information, compressed programmes, unclear briefs, fragmented responsibility, future flexibility, risk transfer, or the fear of being the person who cut things too fine. But good intentions do not automatically produce good buildings.

The result is often a hidden “certainty tax” that the client never knowingly agreed to pay. It appears in capital cost, plant space, ceiling voids, riser sizes, electrical infrastructure, embodied carbon, controls complexity, maintenance burden and long-term operational inefficiency. It also shows up in buildings that are harder to commission, harder to operate and, in some cases, less comfortable for the people using them.

This is not an argument for underdesign

To be clear, this is not an argument for cutting corners. It is not an argument against resilience, safety factors, future allowances or recognised industry guidance. Quite the opposite. Good engineering requires due allowances for peak conditions, maintenance, tenant flexibility, future change and operational risk.

The issue is whether those allowances are visible, justified and proportionate; or whether they are buried inside layer after layer of conservative assumptions. That distinction matters.

A design margin that is consciously agreed is engineering judgement. A design margin that appears because nobody has challenged a default input, a duplicated allowance, or an outdated rule of thumb is something else entirely. It is not robustness. It is uncertainty expressed as extra kit.

The problem with layered margins

Take a typical commercial HVAC strategy. We should absolutely design for agreed peak conditions. We should consider credible climate scenarios, realistic occupancy profiles, internal gains, resilience requirements and the client’s appetite for future change. That is our job. The problem starts when those already cautious assumptions are padded again. A conservative occupancy density is combined with conservative equipment gains. A peak weather file is combined with additional safety margins. The result is rounded up to the next available plant size. Then, during procurement or coordination, further allowances are added because nobody wants to be exposed.

By the time the system is installed, it may be sized not for the reasonable worst case, but for something beyond it.

Bigger systems don’t necessarily perform better

That does not always make the building perform better. In many cases, oversizing can create its own operational problems: poor turndown, inefficient part-load performance, short-cycling, weaker humidity control, increased maintenance stress and more complicated commissioning. The plant may be capable of meeting an extreme peak condition, but for the vast majority of its operating life it is running well below that condition, often inefficiently.

This is where overdesign becomes self-defeating. We tell ourselves we are protecting the client from risk, but we may be giving them higher capital costs, higher running costs, larger plant areas and a building that is more difficult to operate well.

The same applies to electrical infrastructure, pipework, ductwork and controls. Each individual oversizing decision may look minor. But buildings are systems. A larger load allowance can affect the incoming supply, transformer capacity, switchgear, cable sizes, containment routes, risers, plantroom space, heat rejection, standby provision and cost. The consequences are cumulative.

The carbon cost of excess 

There is also a carbon cost. 

The industry has rightly focused heavily on operational carbon. But as buildings become more efficient and the grid continues to decarbonise, embodied carbon becomes increasingly important. Every unnecessary metre of ductwork, every oversized cable, every avoidable section of pipework, every piece of plant selected two sizes too large carries a carbon consequence before the building even opens.

Precision is sustainability

A lean design is not simply a cheaper design. It is often a lower-carbon, more buildable and more operable design. The best engineering is not about adding more until everyone feels safe. It is about understanding the problem deeply enough to know what is genuinely required. 

That requires confidence, but not recklessness. It requires first-principles thinking, but not a disregard for standards. It requires engineers who can interpret guidance with judgement, context and confidence. CIBSE, BSRIA and other industry guidance is essential. The problem is not the guidance. The problem is when guidance is treated as a substitute for project-specific judgement.

Better tools, better judgement 

Modern tools should help us challenge this. Dynamic simulation, detailed load analysis, metered data, computational fluid dynamics, post-occupancy feedback and better commissioning records can all give us a clearer picture of how buildings actually behave. But these tools should not be used simply to validate a design that has already been decided. Used properly, they allow us to test assumptions, identify real risks and remove unnecessary capacity with confidence.

That is where engineering adds value.

Not by blindly reducing everything. Not by pretending that buildings are predictable or that clients never change their minds. But by making risk explicit. By showing where margin has been included and why. By distinguishing between genuine resilience and accumulated conservatism. By being able to say: “This allowance is necessary,” and just as importantly: “This one is not.”

Designing for the worst case, not beyond it 

Clients should expect it. Contractors should welcome it. Designers should lead it. 

Because overdesign is not a harmless habit. It consumes space, money, energy, materials and carbon. It can compromise architecture, reduce net lettable area, increase plant replacement costs and leave facilities teams managing systems that were never properly matched to the building’s real needs. 

In one sense, true engineering excellence can be thought of as the art of subtraction: knowing what can be taken away without compromising the integrity, resilience or function of the whole. It is also the art of proportion: knowing when to add capacity, when to hold the line, and when to remove what is not needed. It means designing for the worst case, absolutely, but not for worse than the worst case. 

That is the line we need to defend.

If we cannot make that distinction, we reduce engineering to box-ticking and risk avoidance. The better outcome is not the biggest system. It is the right system: properly justified, clearly coordinated and capable of performing as intended. 

Defaulting to more capacity may feel safer in the moment. But over time, better engineering comes from precision: understanding the risk, justifying the allowance, and designing systems that are no larger or more complex than they need to be.