Factory expansion: densify, take more of the building, or open a second site

Find the binding constraint before choosing a route

Sites almost never run out of floor area first. They run out of incoming electrical capacity, effluent headroom, compressed air, roof height over a new machine, floor capacity under one, dock positions at the despatch peak, yard space for the vehicles that arrive, or permitted site coverage. Each of those has a different cost to relieve and a different lead time, and some cannot be relieved at all in the current building. Establishing which one binds, with evidence rather than assumption, changes the question from how much space we need into what it would take to make this site work harder.

Densifying is the least costly route and the one that removes slack

Recovering space inside an existing building — narrowing aisles, compressing storage, adding a mezzanine, tightening the layout — is fast and comparatively inexpensive, and it consumes the buffer that let the plant absorb disruption. The plant that runs at high density copes badly with a delivery arriving early, a machine needing a strip-down, or a quality hold that has to sit somewhere. Density also interacts with the building's fire assumptions, its floor loading and maintenance access, none of which is an operational judgement. It works well where variability is genuinely low and poorly where it is not.

Taking adjacent space is limited by what is adjacent

Extending or occupying a neighbouring unit keeps one management team, one quality system and one set of shared services, which is why it usually beats a second site on cost. The constraints are practical: whether the adjacent space is available at all and on what notice, whether the two can be physically connected without compromising fire separation or structure, whether services can be extended, and whether site coverage and permitting allow a new building. Where a landlord holds the neighbouring unit, this becomes a negotiation to start long before you need the space, not when the machines are on order.

A second site duplicates more than the building

The costs that make second sites disappointing are rarely the property ones. Two locations mean two management teams, two sets of certifications and audits, duplicate tooling and equipment for anything either site might need, inventory held twice, transfers between them, and a quality system that must be demonstrably applied in both. Against that, a second site buys resilience against losing one plant, access to a different labour market, and proximity to a customer or supplier base. That is a genuine strategic case, but it should be made on those grounds rather than presented as a way to add capacity cheaply.

Testing whether the growth is durable enough to build for

Expansion decisions are frequently taken against a forecast that would not survive a contract review. Before committing, separate contracted volume from expected volume, and identify which customers or products would have to disappear for the expansion to become surplus. Then consider the reversibility of each route: densifying can be undone, additional leased space can sometimes be given back, and a second site is hard to unwind. Any structural, electrical or drainage change involved in any route needs designing and approving by qualified engineers, which also sets the real schedule. A phased plan that relieves one constraint at a time preserves the option to stop partway, which a single large commitment quietly removes.

Frequently asked questions

How do we establish whether our electrical supply can support more equipment?

Start with actual measured demand across a representative period rather than the sum of nameplate ratings, which overstates it considerably, then ask the distribution operator what the connection and the local network will support. Reinforcement, where it is needed, is often the single longest lead item in an expansion and is entirely outside your control. Getting that answer in writing before committing to equipment is the difference between a phased plan and machines waiting for power.

Is it worth expanding a building we lease rather than own?

It can be, but the terms decide it. Establish who owns the improvement at the end of the term, whether you must reinstate the building to its original condition when you leave, whether the landlord will extend or vary the term to match the investment's life, and what consent process applies to the works. Investing significantly in a building you occupy on a short remaining term, with a full reinstatement obligation attached, is how companies end up paying twice.

When does adding a shift beat adding space?

Whenever the constraint is machine hours rather than floor area, and the labour market supports it. Running an additional shift uses assets you have already bought, avoids capital entirely and can be reversed if demand softens, which makes it the natural first test. It fails where the constraint is physically spatial, where the process needs the building for cleaning or maintenance during the unused hours, or where recruiting for unsocial shifts is unrealistic in your location.