The cheapest option is the one that cannot be delivered.
Your lease is ending and four proposals are on the table. This model tests every one of them against what you actually need — desks, area, time, capital — before any of them is priced. Gates first, cost second.
Why rent per square foot is the wrong instrument
When your lease approaches its end, your agent will present a comparison ranked by net effective rent per square foot. That metric was built for negotiating a single deal, and it goes wrong in three specific ways when you use it to choose between options.
It divides by floor area — the one variable you are actually deciding. A larger option looks better per square foot precisely because it is larger.
It sees the lease document rather than your occupancy cost. In the UK that means it misses business rates entirely, commonly a fifth to a quarter of the total and appearing in no lease. In Hong Kong it misses out-of-hours air conditioning, which is billed by the hour and is a recurring cost for anyone working across time zones.
It ranks every option, including the one that cannot seat your headcount and the one that cannot be built before your deadline.
The requirement comes first
Before a single proposal is entered. Headcount year by year, desk-sharing ratio, space standard, the date by which you must have decided, and the capital you could fund in one year.
The sheet returns how many desks you need, how much area, how long you actually have once approval time and contingency are taken out — and whether the review should already have started. That last line has a habit of arriving unwelcome.
Then four gates
Each gate returns Pass, Conditional or Fail against a threshold you set, with the margin stated in words: provides 74 desks, 82 required, short 8. Options that fail are not ranked on cost — a building that cannot seat your people does not become viable by being cheap.
Then cost, for the survivors
On total occupancy cost per head. When options differ in area, term and density, people are the only stable denominator — and it is the number a finance director already thinks in.
| Reported | Answers | Why both |
|---|---|---|
| Stabilised run-rateCost per head, per desk, per unit area | What will this cost us? | The cost once incentives have run off. The right basis for comparing options and for calculating payback. |
| Year 1 cash requirementCapital plus first-year running | What do we have to fund? | A long rent-free period can make an expensive option the easiest one to start. Presenting only the run-rate hides the funding requirement. |
| NPV over the planning horizon | Which is better over the period we can see? | Discounted on a mid-year convention, because UK rent is payable quarterly in advance. |
| Payback on the extra capital | How long until the cheaper option repays its capital? | Against the lowest-capital option that passed, not against staying put. |
Four things it does that a lease analysis tool does not
A stress test that reverses the answer. One switch re-runs every gate and every cost at minus twenty per cent headcount. In the worked example the renewal that failed on capacity becomes the cheapest option by a wide margin — about £722,000 better over three years, on the same set of proposals. Taking too little space is a problem you can solve. Taking too much is a problem you pay for every month until the break.
Gross-to-net conversion, on the record. Enter the area as the lease states it and the building's efficiency ratio beside it. Hong Kong offices are quoted on gross floor area; your space standard is a net area. The model derives the net figure for every comparison and carries the ratio through to the summary — rather than leaving it in somebody's calculator, where nobody can question it.
Blank cost lines are estimated, not zeroed. An unanswered cost is scaled from your current building rather than treated as nil, so the option nobody has fully quoted looks dearer rather than cheaper. Every option carries a Complete, Estimated or Incomplete label through to the page a board reads.
The dilapidations switch. One control decides whether leaving crystallises the liability on your current building. Renewing defers it; relocating pays it. That single difference is the mechanical distinction between staying and going, and it is the one a lease-document comparison has nowhere to put.
What is in the cost pool
Ten recurring lines and eleven one-off lines. Two of them are worth naming, because they are the ones usually missing.
Office Lease Options Appraisal Model
Up to six options side by side, on one cost pool, with the requirement fixed before any of them is entered.
- Four gates — capacity, lead time, capital, commitment — with the margin stated and what would make a failed option pass
- Full occupancy cost pool including out-of-hours air conditioning, dilapidations, dual running and technical migration
- Serviced and traditional space levelled onto the same comparison, with the implied density of each
- Gross-to-net area conversion using the building's efficiency ratio, carried through to the summary
- Stabilised run-rate and Year 1 cash requirement reported separately
- Minus twenty per cent headcount stress test on one switch
- One page formatted for an approval paper, with every assumption recorded beside it
It supplies no market rents, no construction rates and no benchmark costs. Every figure is your own lease terms and your own quotes.
The demo is a five-page worked case showing everything the model produces. Buying gets you the workbook, a fully populated worked example and the guide in English and Traditional Chinese.
Questions before you buy? info@chargeablearea.com
Where this model is weakest
Stated here rather than discovered after you have paid.
| Limit | What it means, and what to do |
|---|---|
| North American leasesBase year escalation | Operating expenses are commonly passed through as escalations above a base year, compounding from nil. The workbook carries a recurring cost and inflates it, which is a different shape — not a different number. Calculate the escalation year by year and enter the result; the guide explains how. Every other supported market is handled natively. |
| No lease accounting | No right-of-use asset, no lease liability, no straight-lined charge. It answers what an option costs and whether it can be delivered, not how it will be reported. |
| Partial surrenders | The dilapidations trigger carries one figure. Where you hand back part of a floor and keep the rest, you split the liability yourself and record how. |
| Not a negotiation tool | It does not reverse-engineer what a landlord would have to concede to match a rival offer. Once you have chosen a direction, that is your agent's job. |
The guide is not a manual
Twenty-five pages in English, and the same guide in Traditional Chinese — localised rather than translated, with a short VAT chapter and a considerably longer Hong Kong section.
Four chapters cover the mistakes that cost the most and are hardest to see once the numbers are in a spreadsheet. VAT basis mismatches, where a traditional lease quoted net sits next to a serviced office quoted gross and the flexible option is a fifth dearer than it really is. Transaction tax, which on a UK lease is charged on the net present value of rent over the whole term rather than as a percentage — and which a long rent-free period materially reduces. Run-rate against first-year cash. And what "all-inclusive" excludes in a flexible office, with a checklist to work through before you treat the price as comparable.
Each of the four markets has its own conventions set out: measurement basis, how service charge and property tax are structured, what falls due on exit, and the quality classification in use.
Already decided to stay?
The Office Occupancy Cost Allocation Model splits occupancy cost across departments, cost codes and legal entities, with common area grossed up at the level it is genuinely shared at — on a basis that survives a challenge from a department head.