So first off, we treat the workplace as a strategic business tool, not a thing of beauty or an engineering feat: something acquired and designed to produce a business result. If it doesn’t add value, why do you have an office at all?

A lease event is a moment in time at which you can work out what kind of tool the business needs next, but you can only do it by working out the scenarios your business might face and with a strong appreciation of how your current business works.

This guide sets out the sequence that starts with working out what the business is trying to achieve, how it works today, what is changing around it, how much space it needs, and how, eventually, it should be used and managed to help achieve business goals.

Decide what you are trying to achieve through the new workplace, before anyone views a building. The first strategic question we ask of your ‘C’ suite leaders is ‘what are you seeking to achieve through the new workplace?’ It’s a simple question and one that requires some deep thought about the organisation’s goals, its external environment, and the way it works.

We’ve found a good way to answer the key question is to issue our Organisational Priorities app to each of the leaders and invite them to allocate 100 points per person to a series of 10 organisational capabilities in relation to their significance to the future success of the business.

The points are allocated in relation to the importance of the specific capability to business success. We play them back in a Vision workshop and reveal all the results to stimulate a discussion that leads to a clear prioritised list of capabilities that the team is committed to.

This priority list can then be used to drive the design, manage ‘value engineering’ tradeoffs, and measure operational performance through time.

Most leaders have a strong opinion on how big their office should be, and almost no data on how it’s actually used. So, to have a meaningful dialogue, our project team captures an evidence-backed picture of the ‘current’ state.

This may include an understanding of existing data surrounding headcount (actuals and forecasts) and attendance, a workplace utilisation study that assesses the use of every space at agreed intervals during the day over a two-week period, along with a thorough review of the space provision from drawings and benchmarks.

Focus groups, interviews, infrastructure reviews, and surveys can help to get a richer picture of the experience employees are sensing on a day-to-day basis. Increasingly an assessment of the cognitive performance of the workforce and the impact of AI are needed to shape the understanding of the here, now and future demands.

The gap between belief and evidence is usually wide. Our Hybrid Working Index 4 (November 2024, 55 offices across 19 countries, 68,000 employees) found average office attendance of 34% against average desk utilisation of 55%.

In terms of location, it’s useful to undertake a ‘Commute Study’. Using specialist software, we’re able to assess the current domestic locations of your workforce and the impact of different locational options. We can quantify the relative change in commuting difficulty along with the sensitivity of a relocation on the likely retention of key skills.

Sizing a 10-year lease on current headcount or next year’s hiring plan is the most common error we see. But we start in a different place. Our question is ‘do you need an office at all?’ and if you do, what do you need it for and what will it be used for.

Advisers associated with organisations that deliver transactions, design services and project management will be reluctant to start in that position. But all these questions need to be considered.

Then there are business changes. What happens when you hire faster than planned and run out of desks? What happens if AI means you need less space? Or if you hire slower and have to pay for empty space?

Scenario modelling is vital. Modelling different business, occupancy and headcount scenarios at twelve months, two, three, four and five years is critical, along with the consideration of strategies that might enable growth without taking on more space.

In today’s world, office attendance is focused on Tuesday, Wednesday and Thursday, with limited attendance on a Monday and barely any on a Friday. The problem is that you may need to size your future office on the peak demand on a Wednesday if you are unable to smooth the demand for space across the week. Call it the peak-day tax: you build for one day and pay for it on all of them.

Also worth remembering that the building will be yours 24 hours a day, 7 days a week, a total of 168 hours. In our experience most organisations are only occupying them for a fraction of that, with genuinely intense use across a narrower window still.

Scenario modelling gives you an understanding of space quantum needed under different circumstances. What happens if the headcount grows, or shrinks? What happens if AI is implemented? What happens if we shift our model to 7 days a week working? What happens if we do more hybrid work? Can we use space at night to serve other geographies?

The next question is what kind of work those people will actually be doing.

AI changes the mix of work inside roles before it changes the number of people, and both change the building. We know from our research that AI will automate mechanistic tasks within roles and amplify the cerebral tasks, allowing people to model, write, create, and develop in ways that are more efficient and economically viable than ever before.

Organisations under pressure for margin in steady state business situations will likely use AI to reduce heads and costs. Organisations growing are likely to avoid taking on more people by increasing the productivity of existing workers.

Our approach is to review jobs using our AI role assessment tools, breaking them down into the mechanical, repetitive aspects versus the cerebral tasks. AI, on one hand, is good at repetition and pattern matching, making it easier to automate repetitive tasks.

For instance, a research scientist might be 70% cerebral and 30% mechanistic. A call centre agent might be the other way round. And over time, the cerebral-heavy roles increase while the mechanistic ones may reduce as automation kicks in.

The scale is large enough to change a lease decision. Our AI Impact Report 2026 analysed 412 UK occupations across 18 task categories and models a net cut in UK office demand of around 28 million sq ft at three years, 54 million at five and 90 million at ten.

We expect organisations to shrink in size, but grow in cerebral activity, meaning organisations need to cater for a highly intelligent, knowledge-intensive organisation. AWA’s work on Knowledge Worker Productivity, launched in 2016, now becomes an important asset in defining the future workplace.