Is it cheaper to own exam centres or book pop-up space?

It depends on utilisation, not on the rate. Owning converts a variable cost into a fixed one, so it wins when a site is used heavily and steadily and loses when it sits idle. Exam and training demand is spiky and seasonal, so owned capacity in most cities is paid for on days nothing runs. Booked space costs more per day and nothing on the days you do not use it.

Why are pop-up classrooms better for expansion?

Because expansion is usually about coverage — being present in more cities — and coverage multiplies the cost of owning while leaving the cost of booking roughly linear. Thirty cities means thirty leases, thirty fit-outs and thirty utilisation problems if you own, against thirty bookings if you do not. Pop-ups also make a wrong city reversible: you simply do not book it again.

When does owning your own centre make sense?

When a single location has high, predictable, year-round utilisation; when the environment must be controlled to a standard no third party will meet; when security or accreditation requires custody of the premises; or when the space is a product in itself rather than a means of delivery. These are density arguments, and they apply to a small number of locations, not to a footprint.

What is the anchor and pop-up model?

Own or hold long-term space only where volume is dense and continuous — typically a handful of metros — and cover everywhere else on demand. It keeps control where control pays for itself and keeps the long tail variable, which is where the utilisation risk of owning actually bites.

← Knowledge

Analysis10 min read28 September 2026

Own infrastructure or pop-up classrooms? A comparison for expansion

Owning centres and standing them up on demand fail in opposite directions. Which one is right depends on a single question — whether you are expanding for coverage or for density.

ExpansionPop-up classroomsCapacity planningExam centresStrategy

In short: owning capacity converts a variable cost into a fixed one. That is a good trade when a site is used heavily and steadily, and a bad one when it is not — and exam, training and drive demand is neither heavy nor steady in most cities. The question is not which model is cheaper. It is whether you are expanding for coverage or for density.

Every organisation that runs exams, assessments or training across a country eventually asks whether it should stop renting and start building. The instinct is reasonable: rates look like money leaking out, and an owned lab feels like an asset rather than an expense.

The instinct is also where the arithmetic usually goes wrong, because it compares a day rate to a day rate. That is not the comparison. The comparison is between a cost that stops when you stop, and a cost that does not.

The two models

Own infrastructure. You take space — leased or bought — fit it out with workstations, power backup, connectivity and furniture, get it audited, and it is yours. It runs when you want it to. It also exists on the 300 days a year you do not use it.

Pop-up classrooms and booked centres. You stand up capacity in a space that already exists, for the days you need it, and hand it back. Someone else owns the utilisation problem, the maintenance and the lease.

Side by side

Dimension Own infrastructure Pop-up / booked

Cost shapeFixed. Capex plus rent, salaries and maintenance whether or not anything runs.Variable. Zero on days you do not book.

Cost per day usedLow at high utilisation, punishing at low.Higher per day, flat regardless of how often.

Time to openMonths — site, fit-out, power, connectivity, audit.Days.

ReversibilityA wrong city is a lease and a written-down fit-out.A wrong city is a booking you do not repeat.

Coverage costMultiplies — every city is another full build.Roughly linear — every city is another booking.

Peak capacityCapped at what you built.Elastic — book more rooms for the peak.

Control of environmentTotal.Contracted and audited, not owned.

Who carries idle riskYou.The venue owner.

Scales withCapital.Demand.

The arithmetic that actually decides it

Strip it to one line. An owned site is worth it when

days used per year × (booked day rate − marginal cost of running your own) > annual fixed cost of holding the site.

Everything follows from the first term. At 200 days a year the fixed cost spreads thin and owning usually wins. At 20 days a year it does not, and no negotiation on the day rate closes a gap that size — because the thing you are paying for is the other 345 days.

So the honest first question is not "what does a centre cost?" but "how many days a year will this specific location actually run?" For a head-office training facility in a metro, the answer can genuinely be most of the year. For an exam centre in a tier-2 city serving two hiring cycles, it is a fortnight. Those two answers point in opposite directions, and they are often asked about the same organisation in the same meeting.

Why expansion changes the answer

Expansion is normally about coverage — being present where candidates are, so they do not travel and attendance holds up (see how many seats to book). Coverage is exactly the case where owning compounds against you.

Going from 3 cities to 30 does not multiply your demand by ten in each place. It spreads roughly the same volume thinner. Owned capacity in that world means thirty leases, thirty fit-outs, thirty audits and thirty separate utilisation problems — each one worse than the last, because the thirtieth city has the least demand. The booked model in that world means thirty bookings.

Expanding for… What it looks like Which model fits

CoverageMore cities, similar volume per city, seasonal peaksPop-up / booked

DensitySame cities, much more volume, year-roundOwn, in those cities only

This is why the two models are not really competitors. They answer different questions, and most organisations expanding a national footprint are asking the coverage one while evaluating the density answer.

The cost that does not appear in the model

Three things reliably get left out of the build-versus-book comparison, and all three favour not building:

Refresh. A computer lab is not a one-off cost. Workstations age out, and the fit-out you are amortising over ten years contains equipment with a much shorter life.
Asset specificity. A 200-seat exam lab is a poor office, a poor warehouse and a poor anything else. If demand moves, the asset does not move with it, and it does not resell into a broad market.
Management attention. Every owned site is a landlord relationship, a maintenance schedule, a security question and a person whose job it is. Thirty of them is a property operation attached to an organisation that did not set out to run one.

Where owning genuinely wins

A comparison that only argues one way is not worth reading. Owning is the right call when:

Utilisation is high, steady and in one place. A permanent assessment or training centre in a metro running most weeks of the year is exactly the case the fixed-cost trade was designed for.
The environment has to meet a standard no third party will. Specialised equipment, a fixed physical configuration, or an accreditation that inspects the premises rather than the session.
Custody is the requirement. Where security obligations extend to the building itself and not just to the exam, a contract with a venue may not be enough.
The space is the product. If people are buying the room — a flagship, a showcase, a brand experience — then it is not delivery infrastructure and this comparison does not apply to it.

Notice that all four are arguments about particular locations. None of them is an argument for a footprint.

What most organisations land on: anchor plus pop-up

Own or hold long-term space only where volume is dense and continuous — in practice a handful of metros. Cover everything else on demand.

This keeps control where control pays for itself, and keeps the long tail variable, which is precisely where the idle-capacity risk of owning bites hardest. It also means a new city is a test rather than a commitment: run it booked for two cycles, look at the volume, and only then ask whether it has earned a permanent site. Very few do.

Question If yes

Will this location run more than ~150 days a year?Consider owning it

Is demand seasonal or tied to hiring cycles?Book it

Are you adding cities rather than volume?Book it

Could the right city change within three years?Book it

Does accreditation inspect the premises, not the session?Consider owning it

Do you need 10× capacity for two weeks a year?Book the peak, whatever you do with the base

The practical point

A network already spans the footprint that owning would take years and a great deal of capital to assemble. Ours runs to 730 centres across 225 cities in 35 states. Reproducing that by building is not a budget decision; it is a different company.

The useful way to frame it internally is not "build or rent". It is: which of our locations have earned a fixed cost, and why are we carrying one anywhere else?

Expanding into new cities? The drive planner takes cities, shifts and tentative numbers and comes back on availability — which is usually the cheapest way to find out whether a city deserves a permanent site.

Common questions

Is it cheaper to own exam centres or book pop-up space?: It depends on utilisation, not on the rate. Owning converts a variable cost into a fixed one, so it wins when a site is used heavily and steadily and loses when it sits idle. Exam and training demand is spiky and seasonal, so owned capacity in most cities is paid for on days nothing runs. Booked space costs more per day and nothing on the days you do not use it.
Why are pop-up classrooms better for expansion?: Because expansion is usually about coverage — being present in more cities — and coverage multiplies the cost of owning while leaving the cost of booking roughly linear. Thirty cities means thirty leases, thirty fit-outs and thirty utilisation problems if you own, against thirty bookings if you do not. Pop-ups also make a wrong city reversible: you simply do not book it again.
When does owning your own centre make sense?: When a single location has high, predictable, year-round utilisation; when the environment must be controlled to a standard no third party will meet; when security or accreditation requires custody of the premises; or when the space is a product in itself rather than a means of delivery. These are density arguments, and they apply to a small number of locations, not to a footprint.
What is the anchor and pop-up model?: Own or hold long-term space only where volume is dense and continuous — typically a handful of metros — and cover everywhere else on demand. It keeps control where control pays for itself and keeps the long tail variable, which is where the utilisation risk of owning actually bites.

Next steps