How it works
The inputs are the same as in manual planning: headcount by team and forecast growth, measured attendance and peak occupancy, which teams need to sit together, floor plans with desk counts, and constraints such as closed floors or fixed rooms. The difference is that the AI holds all of them at once.
An optimizer or planning agent proposes a stack plan (which team goes on which floor) and then a seating plan (which desks within each floor), scoring options against goals such as keeping teams together, fitting the busiest day or minimizing moves. Planners ask questions in plain language, lock decisions they want kept and let the AI rearrange around them.
Scenarios are the main output. Leaders compare, for example, releasing one floor against two, or growing a team in place against moving it, with the desks, teams and moves for each. A person approves the chosen plan before anything changes in the booking system.
Results depend on the data. Plans built on headcount alone assume everyone comes in; plans built on measured occupancy size each team to how it actually uses the office.
How to measure it
- Check that the plan uses measured occupancy, not headcount alone.
- Ask for the reasoning behind each placement and the trade-offs between scenarios.
- Lock the constraints that matter, such as adjacencies and fixed rooms, before optimizing.
- Compare the approved plan with actual occupancy after the move.
Examples
- A planner asks, 'Can Engineering take ten contractors next quarter without splitting across floors?' and gets a revised stack showing the desks and teams affected.
- A company models three consolidation options, from one floor released to three, and approves the one that keeps two closely linked teams together.
Benchmarks
In CBRE's Global Workplace & Occupancy Insights, data quality issues and lack of expertise were named as the biggest challenges, cited by 55% of respondents.
CBRE, 2026 Global Workplace & Occupancy Insights
Common mistakes
- Planning from headcount instead of measured attendance.
- Accepting an AI plan without seeing its reasoning or constraints.
- Stopping at the plan: without scheduling and execution, the approved layout never reaches the booking system.
- Letting the AI assume adjacencies nobody has confirmed with the teams.
How Kadence helps
In Kadence SpaceOps, the AI planning agent builds or changes stack and space plans from plain-language requests. Stack planning with auto-stack places teams on floors, scenario planning compares options side by side, space planning goes down to seat level on real floor plans, and guided workflows cover restacks, floor consolidation and finding space for a team. Plans use WorkOps attendance, occupancy and visitor data, export as PDFs and, once approved, publish to live neighborhoods and assignments. The SpaceOps page describes the aim as to "model, plan, stack, and execute change in hours, not weeks."
Space planning software · SpaceOps · Scenario planning · Office space planning guide
Frequently asked questions
What is AI space planning software?
It is software that uses optimization or AI agents to place teams on floors and desks from headcount, occupancy data, floor plans and adjacency rules. Planners ask for changes in plain language, compare scenarios and approve a plan instead of redrawing it by hand.
Can AI replace a space planner?
No. AI speeds up drafting, checking constraints and comparing scenarios. Planners still set the goals, confirm adjacencies with the teams involved, judge the trade-offs, explain the plan to stakeholders and own the recommendation to leadership.
What data does AI space planning need?
Headcount by team with forecasts, measured attendance and peak occupancy, current floor plans and desk inventories, adjacency requirements, and constraints such as fixed rooms, closed floors and lease dates. Measured occupancy stops the plan from assuming everyone comes in every day.
Which companies offer AI-powered space planning tools?
Workplace management platforms, integrated workplace management systems and specialist planning vendors all offer some AI features. Compare them on whether they plan from measured occupancy, show their reasoning, handle both stack and seat-level plans, and carry an approved plan through to live bookings.
Related terms
- Stack planning A floor-by-floor plan of which teams occupy which floors and buildings, used to test moves and consolidations first.
- Scenario planning Modeling several possible futures for space, cost and teams, and comparing them side by side before committing to one.
- Agentic workplace A workplace where AI agents carry out tasks, such as booking space and drafting plans, with people approving decisions.
- Space planning Deciding how office floor area is divided and allocated, from which teams go on which floors to where each desk and room sits.
- Occupancy planning Deciding where people and teams will work across floors and buildings, and how much space each needs, as the organization changes.
Sources
Kadence editorial team · Last reviewed