Space management insights #4: Smarter space decisions with minimal spend

Michael Grant, GM EMEA, Lambent

Editor’s Note: This article is part of an exclusive series featuring insights from our ebook, The smarter office: Your guide to space management in the hybrid era. Read the full guide for more insights on actionable space intelligence, making the office a destination, AI’s role in space management, and more.

Most facilities teams are still measuring the wrong things.

They may know how many people enter a building. But what often gets missed is what happens after that. Where are they going? Where do they spend their time? Which areas of the office are booked the most, or the least? Which spaces look busy on paper but sit empty in practice?

Once you can answer questions like these, then you can start making the right adjustments to your space. 

That is the shift I think space management teams need to make. The goal is to collect the right data, connect it to the right decisions, and use it while you still have time to act.

Put the decision before the sensor

Before you choose a data collection approach, you need to check a few basic boxes:

• What do we need the data for? 
• What decisions will be made based on the data? 
• What level of detail do we need? 
• How accurate does the data need to be?

Once you’ve answered those, you’re better equipped to look at what data is already available to you and what you may need to build out.

Too often, teams start with the technology. They ask whether they need sensors, badge data, Wi-Fi data, booking data, or a new platform. But the better starting point is the decision you’re trying to support.

If you’re making a lease decision, floor-level data may be enough. If you’re redesigning a neighborhood or changing meeting room layouts, you may need a more detailed view. If you’re trying to validate ghost meetings, you may need to compare booking data against actual occupancy.

Don’t gather everything just because you can. Gather what helps you decide.

Your existing infrastructure is more useful than you think

A lot of organizations already have useful occupancy signals moving through their existing systems.

Wi-Fi data, badge readers, desk booking data, and room booking data can all help you understand how spaces are being used. When you overlay those sources, you can uncover ghost meetings, validate utilization assumptions, and start building a more accurate picture of demand. That combination is often a practical place to start.

Start at floor-level data. This level of detail is enough to build reporting and dashboards that can be shared at the executive level. It can also support major decisions and validate assumptions, not to mention a faster path to value.

Hardware and sensor solutions can be expensive to buy, install, and manage over time. When spaces are reconfigured, sensors often need to be physically moved and reconfigured. The total cost of ownership of hardware and sensor solutions can be up to 90% more expensive than software-based solutions.

On the other hand, software takes less time to deploy, is easier to scale, and can often deliver a richer dataset.

That matters for large portfolios. If you can connect to a Wi-Fi occupancy data platform at a rate of hundreds of buildings a day, you can move faster than a hardware-first approach would usually allow.

Scaling up also scales your data problems

Your challenge gets more complex when you’re managing a large campus or multi-building footprint.

Many buildings, network domains, and space types all need to be stitched into a single utilization view. Each building may have different Wi-Fi configurations, legacy systems, and naming conventions for rooms and zones. 

That variation creates a real governance problem. Procurement costs and operational overhead can escalate quickly too. At scale, a hardware-first approach can become a significant ongoing burden. I’ve seen companies dedicate workers just to changing batteries and reconfiguring sensors after replacement. The time and money wasted there is significant.

Geographic complexity adds another layer. IoT occupancy sensor companies tend to be relatively small, and their devices may not be certified in every country or region. Some customers have had to procure locally made devices that don’t integrate with the devices they have in other regions. That creates management and support headaches that grow as the portfolio grows.

This is why I recommend resisting isolated point solutions. For significant operational efficiencies, consider investing in a portfolio-wide occupancy analytics platform. Assume more teams will want access to the data, and make sure the approach can scale across buildings, campuses, and stakeholders.

Time to data should be a core selection criterion. Insights need to arrive in time to influence live projects, not just describe what happened in the past.

Better data changes the leadership conversation

Leadership teams don’t need another report that describes what happened months ago. They need something they can act on.

The best way to predict what will happen is to understand what patterns of utilization have happened in the past. The more high-quality data you feed into AI models, the more powerful they become for forecasting and scenario planning. That forecasting power changes how you make the case for investment.

Start with quick, tangible wins. For example, if you can use occupancy data to adjust HVAC settings and save 30% on energy costs, you have a compelling ROI statement.

Then add other use cases. Occupancy forecasts can help right-size collaboration zones, reduce real estate spend, and increase overall productivity. Present those in a simple dashboard or one-page executive summary showing ROI metrics alongside strategic outcomes like faster lease decisions and agility for growth. This proves that your team is driving P&L impact, not just managing costs.

Energy savings are often the easiest place to start. Most buildings are run terribly. Turn on the heating, cooling, and lights at 6 in the morning, and turn them off at 9 at night. But if two of your three floors sit empty that day, that creates huge waste.

If you can point to meaningful savings, you’re more likely to get buy-in for the next layer of investment.

Collect enough data to see the patterns

You need enough data to understand how the space is being used, rather than just how many people come in and out.

You can’t do that in a month, or two months, or even three months. For fundamental space changes, I recommend collecting at least three to six months of data. The bigger the space and the bigger the decision, the more data you want behind it.

That time matters because workplace behavior is not perfectly consistent. One-off anomalies start to even out when you look at continuous occupancy data. Your curve gets less volatile, and you get a much better sense of how to project forward.

Seasonality matters too. At the beginning of the year, people may be more collaborative and need more shared spaces. At the end of the year, when everybody is trying to wrap up, they may close off and start needing pods.

If you make a major space decision from a narrow snapshot, you risk designing around a temporary pattern.

Compare booking data to occupancy data

As you redesign spaces, it’s best to go beyond static floor assignments.

Compare booking data against occupancy data. If you overlay these two data sets, it helps you understand what areas are booked the most, what areas are booked the least, and where people go. That combination reveals whether the spaces you’ve designed are actually serving the people in the building. Just looking at one of those datasets alone doesn’t give you the full picture.

A meeting room may look unavailable all day in the booking system, but in reality, it sits empty for half of that time. A neighborhood may look lightly used from badge data, but show heavy dwell patterns during certain parts of the week. A desk area may technically have enough seats, but fail because people don’t want to work there. The point is to make the space fit the way its users work.

Right-sizing needs both data and people

Data is important, but it’s only a starting point. You need other touchpoints and human interactions to understand why a space is or isn’t occupied. 

A dashboard can tell you that a room is underused. It usually can’t tell you that the room is cold, awkwardly located, too loud, or missing the right furniture.

That’s why I recommend supplementing your data with direct conversations. But be deliberate about who you ask. You’ve got to find the right people for their opinion. Not the people who are emotional and just say, “that’s my space and you can’t get rid of it,” but the people who say, “I use these types of spaces for these reasons.”

Treat that feedback as a third dataset, alongside occupancy and booking analytics.

The way you reinvest right-sizing savings matters too. Follow your data to see what you need more or less of. Spend your initial savings on improving the quality of the space, such as with better furniture or new facilities.

This kind of approach helps right-sizing feel less like a loss and more like a smarter investment in the places people use.

Space data should shape what happens next

Anticipating trends in office usage turns space management from a reactive cost center into a strategic advantage. When teams can see what is likely to happen next, they can design spaces, services, and investments that meet demand before friction shows up.

That is where AI becomes interesting. Building automation is one area where AI can generate measurable results. A building can start and stop heating or cooling based on occupancy and thermal thresholds. It can turn lights and power on and off based on actual usage. It can route lifts toward floors that data shows are consistently most active.

This can lead to cost savings, extend the life of equipment, and make spaces more comfortable for people to use.

AI can also uncover patterns humans would never catch in spreadsheets, like seasonal dwell shifts, cross-building flows, or micro-utilization trends in breakout rooms. Those trends can inform neighborhood design and amenity placement.

Predictive maintenance is another use case. AI can monitor trends and anomaly patterns to trigger maintenance workflows before failure occurs. That becomes predictive maintenance rather than break-fix maintenance.

The thread across all of this is the same: understand what is happening in the space, use the data to see what is likely to happen next, and make changes while they still matter.

At the end of the day, space management gets better when the data stops living in reports and starts shaping decisions.

Get more expert insights in our ebook, The smarter office: Your guide to space management in the hybrid era.