Office towers seen from below against a white sky.
enzwa lab  ·  concept

Prism

A counterfactual engine for a property portfolio. It costs every option over ten years against the one that never goes on the slide: changing nothing.

A tool for comparing property options against the cost of leaving things exactly as they are.

The problem

A business case usually compares the two or three options somebody has already written down. The option nobody writes down is carrying on unchanged, and that has a price too. Without it there is nothing to measure the others against.

What it does

Prism costs every route over ten years, including doing nothing, and shows what each one is worth against that baseline. Change what matters and the ranking moves.

the option that is never written down

Three options go to the board. The fourth one wins.

A portfolio review arrives as a shortlist. Consolidate. Relocate. Re-gear the lease. Each one is costed, each one has a capex number and a disruption estimate, and each one is defensible.

Then the meeting reaches the thing that actually happens most of the time. Nothing. It arrives as a decision to revisit after the next planning cycle, it is not on any slide, and it is the only option in the room with no number against it.

Across global enterprises, office utilisation sits between thirty and fifty per cent while the fixed lease cost underneath it has not moved in proportion. The gap is what deferral costs, and it compounds.

the prototype portfolio
assets
12
cities
9
area
506k sq ft
seats
4,590
annual cost
US$48m
owned
1 campus
live break option
1 HQ
what the baseline actually does
Zero is not flat.

Every business case is written against a baseline, and the baseline is nearly always drawn as a straight line. It is not one. Rent reviews, indexation, dilapidations and the cost of holding space nobody uses all sit inside the do-nothing case, and they climb.

600450 300150 0 cumulative occupancy cost, US$m yr 0yr 2 yr 4yr 6 yr 8yr 10 change nothing · 550 re-gear · 512 consolidate · 489 anchor + flex · 471 79 the gap
swipe the chart →
change nothingre-gear the anchor core consolidate to eight citiesanchor core plus a flexible edge

Prototype portfolio, ten-year cumulative occupancy cost in US dollars, indexed. Every figure on this page is illustrative and comes from Prism’s demonstration data rather than a real portfolio.

A business district seen from directly above at midday.
twelve assets  ·  nine cities  ·  one live break option
what goes in

A register, and three questions.

The asset list is the easy half. Drop a register in any shape and the data agent maps the columns, validates what is there and fills the gaps from benchmark data, so the model starts from the portfolio rather than from a template.

The three questions are the half that decides the answer. They are not preferences. Each one sets a weight the engine carries all the way through, which is why two companies with identical portfolios come out of Prism with different winners.

the three drivers
forcing function
why now
workforce mix
attendance
capital posture
capex : opex
core envelope
US$20m
discount rate
7%
horizon
10 years
01

Why now

The forcing function behind the review: a lease event, a cost target, a merger, a growth plan. It sets how every trade-off downstream is weighed.

02

How people work

The workforce mix, by work style. Each style carries its own attendance pattern and space signature, so demand is modelled from people rather than from the desks that happen to exist.

03

Where capital works

The portfolio splits into a stable anchor core and a variable flexible edge. The edge is funded on consumption by flex operators, so the question narrows to how the core is paid for.

04

What comes back

Three costed scenarios, each one measured against the baseline over ten years, ranked by the weights those three answers set.

how the answer is built

Eight dimensions, scored per asset.

Specialist agents score every asset, model demand from the persona mix, generate candidate actions and validate them against lease dates, capex envelopes and adjacency rules before anything reaches a scenario.

Each figure that ends up in the report can be opened back to the asset and the assumption that produced it. That traceability is the difference between a model somebody will sign and a number somebody has to trust.

capacityfinancialefficiencyflexibility employeequality & ESGlocationfit
one run
assets scored
12
dimensions
8
demand modelled
4,590
candidate actions
36
validation checks
112
scenarios built
3
run time
~20 sec
One office facade, flat and frontal, an unbroken grid of windows.
the baseline is a decision, and it has a price
what comes out

One portfolio, four cases, one table.

The baseline is the first row rather than an assumption in a footnote. Everything else is read against it.

case10-yr costNPV vs baselinecapex citiesseatsdisruptionscore
Change nothing Renew in place, absorb indexation, hold the space nobody uses. 550—2 94,590none41
Re-gear the core Extend the anchor leases early against a softer market, no moves. 512246 94,590low58
Consolidate Nine cities to eight, two buildings into the owned campus. 4894020 84,310high67
Anchor plus flex Hold a smaller owned core, move the variable edge onto consumption terms. 4715214 94,590medium74

US$m unless stated. NPV of savings against the baseline, ten years at seven per cent. Score is a composite of the eight dimensions weighted by the three driver answers. Illustrative throughout.

the part a recommendation hides

Change why you are asking, and the winner changes.

The same four cases, the same portfolio, the same ten years. The only thing that moves is the answer to the first question, and the ranking turns over. A recommendation buries that weighting in a paragraph. A model puts it on a dial and lets the room argue about the dial instead of the answer.

forcing function

A cost target from the board

  1. Anchor plus flex74
  2. Consolidate67
  3. Re-gear the core58
  4. Change nothing41
forcing function

A merger and a hiring plan

  1. Consolidate71
  2. Anchor plus flex66
  3. Change nothing52
  4. Re-gear the core49

Under a hiring plan the do-nothing case rises from fourth to third, because the thing it protects, headroom already paid for, is suddenly worth something. That movement is the output.

the model is live

Put your own portfolio through it.

Drop an asset register or start from the prototype portfolio, answer the three questions, and watch the agents score, model and validate before the scenarios assemble. Every figure opens back to the asset it came from.

Open Prism →