Register for NexusCon 2025 - Now Accepting Abstracts!
“Wednesday afternoons saw high demand for 2-person huddle rooms, but low engagement in larger meeting spaces. So the team rebalanced room types to match real patterns of work.” If that sounds like an obvious outcome of IoT deployments, why do so many workplace teams struggle to get there?
In the last few years, corporate occupiers invested heavily in Workplace IoT—networks of occupancy counters and indoor air quality (IAQ) monitors—to navigate post-pandemic-era challenges. We counted people to support hybrid work, and we measured CO₂ to reassure everyone the air was healthy.
A 2022 JLL study reported that 56% of companies planned to invest in environment and occupancy sensors by this year, but those investments have left workplace teams swimming in data (and sometimes old batteries) without a clear plan for what comes next. As James Wu, CEO of InnerSpace, puts it, “we’ve arrived at a ‘so what?’ moment” for workplace IoT.
The pressure is on to translate all this raw sensor data into safer, more effective, more engaging workplaces—and to prove it’s worth the ongoing cost of the IoT stack. Simply knowing how many people are in the office or seeing a dashboard of CO₂ levels isn’t enough. Up to 88% of IoT data in organizations goes unused, and workplace leaders are acutely feeling that statistic.
Michael Krall of JP Morgan Chase framed the market’s challenge well: “The real question is who can quickly tell me what to do with the data and simplify the next step to do it. If [solutions] are just dashboards or heatmaps, then they’re a luxury.” In other words, if occupancy/IAQ sensors don’t lead to action, why bother?
Yet, we’re optimistic. Through conversations with workplace pros and tech vendors at the forefront (Sauer Strategy Works, Butlr, XY Sense, R-Zero, InnerSpace, and Airthings—plus a workplace team that will remain anonymous), a picture emerges where data overload gives way to measurably better workplaces. In this next phase, the focus shifts from where people are to what they’re doing, what the environment is like, and what to do about it.
Five key trends are helping workplace teams cure the data overwhelm and connect the dots from sensors to outcomes. We explore each below, with real examples from the workplace frontlines.
Our biggest surprise? We thought we’d hear about the inroads AI is making. Get data to the cloud and let the software tell the user what they need to know… right?
Wrong.
As we’ve lamented over the years, it’s not uncommon for workplace and facilities teams to inherit sprawling dashboards filled with charts, filters, and generic reports—none of which they have time to navigate. “The industry is sick of visualizations and alerts,” one workplace leader told us. “And all the dashboards that are too complicated and crowded.”
Despite this fatigue, vendors still believe they can help users find the signal in the noise. But that doesn’t mean AI is ready to jump in and solve things. Every vendor we spoke with is experimenting with AI, but none claim it’s the silver bullet. Elizabeth Redmond of R-Zero explained that clients often have very specific questions that are customized to their needs, which makes it difficult for AI to come up with insights on its own at scale.
James Wu, CEO of InnerSpace, said a lot of vendors are exploring allowing users to ask plain-English questions like “Which spaces were most used last quarter?” But he cautioned that it may “just move the problem.” If users don’t know what to ask, a blank AI search bar doesn’t help.
Right now, anomaly detection is a promising use case for R-Zero. “You’re really looking for outliers and anomalies,” Redmond said. Instead of making users dig through trend lines, the platform flags what’s unusual—like a floor that suddenly jumps to 80% occupancy on a Tuesday or a room that goes unused for a week.
Similarly, InnerSpace is prototyping AI that spots changes in behavior—like a team suddenly using less space—and suggests explanations, such as a new work policy. While still early, that kind of contextual AI could help workplace teams move from reacting to predicting.
Former Amazon workplace strategist Kevin Sauer, who’s now an independent consultant, compared it to Google Maps: “You can plug in your route, and it predicts what traffic will look like.” In a workplace, that might mean forecasting low turnout before a holiday and reducing catering orders, or anticipating high CO₂ levels during an all-hands meeting and preemptively boosting ventilation. These kinds of proactive nudges are still early-stage—but they point toward a future where systems flag issues before they happen.
In the absence of application layer AI breakthroughs, vendors have been making up ground by improving visualizations and analytics. Shivaun Ryan of XY Sense told us about their new animated historical replay visualizations. Instead of forcing users to scroll through static heatmaps or pivot tables, platforms like XY Sense now allow teams to watch time-lapse maps of office usage. These replays show how people move through space during the day, where congestion happens, and when zones go quiet.
“You don’t need to be a data analyst to use this,” said Shivaun Ryan of XY Sense. “That’s the role of the product.” Their clients are using replays to identify patterns—like which desks never get touched near the café—and to share stories with stakeholders who might never open a dashboard. “Movement replays are universally understandable.” It’s the kind of feature that helps a workplace team say to HR or Finance: “Here’s how your teams are actually using the space”—with a GIF, not a spreadsheet.
Another workplace leader described how their team worked closely with their space utilization software vendor to enhance visualizations and add filters that matched their analytics needs. That meant being able to slice data by neighborhood or day of week, not just room, and export clean insights for design reviews and leadership presentations.
Similarly, Butlr offers clients the ability to tag and group spaces into “virtual zones”—for example, clustering all phone booths across a floor to see aggregate usage. These enhancements reduce the need for manual Excel work and help the software answer real workplace questions like “How are our quiet zones performing?”
When it comes to IAQ, simplifying the data is even more critical. A CO₂ alert in Room 501 might mean open a door, run ventilation, or nothing at all. Without context, alerts create noise. Vendors like Airthings are refining alerts to focus on duration above thresholds, not just spikes, to determine the long-term performance of rooms and spaces for data-driven decisions. They’re also introducing IEQ (Indoor Environmental Quality) grading systems—like an “A” to “F” score for each space—that summarize multiple parameters into a single signal. This echoes how WELL or LEED certifications communicate performance.
The goal is simple: help a facilities manager or HR partner quickly assess, “Is this space healthy or not?” without needing a crash course in VOCs and PM2.5.
One not-so-obvious way to avoid overwhelm: we don’t always need humans in the loop to make use of sensor data. Prioritize sensor use cases that trigger actions automatically or surface information directly to the people who need it, in real time, so the workplace team isn’t stuck poring over charts.
“Technology should be passive... and make everyone’s job easier,” says Shivaun Ryan of XY Sense, which makes optical people-counters and software that communicates real-time space info to occupants who can choose where they want to work.
In practice, this means occupancy and air quality data should immediately translate into something useful—whether it’s an automatic HVAC adjustment or a live map that helps an employee find a free desk.
Real-time integrations are key here. Consider the common problem of conference room “no-shows” (ghost meetings). Traditional occupancy sensors might free up a room after a few minutes of no presence—but if the system only polls every 5 minutes, that’s a clunky experience when someone shows up a little late to find their room already released.
For example, an optical sensor can tell when a room is truly empty and trigger the reservation system to release it without a lag, or conversely detect “passive occupancy” (when someone’s stepped out to grab coffee) so it doesn’t mistakenly count the room as empty.
With air quality, automation is just as critical. Raw IAQ data by itself isn’t very actionable to a workplace strategist—what is useful is when that data drives building systems or maintenance teams’ to-do lists. Andrew Knueppel from Cushman Wakefield emphasized integrating IAQ with Fault Detection & Diagnostics (FDD) software to “actually diagnose a problem and trigger a work order,” skipping the step of a human analyzing a graph.
Landlords are on board too: some are installing IAQ monitors in their buildings specifically to catch comfort issues early. “If they can get ahead of those comfort complaints and build that into their predictive process, that’s extremely valuable,” adds Elizabeth Redmond of R-Zero, noting it helps landlords who manage multiple sites with lean staff.
Redmond emphasized the importance of putting a more comprehensive set of IEQ directly into employees’ hands to improve their day-to-day experience and support their neurodiversity. “If you can give occupants insight into what conditions are present in the office, so they can look for a dark and quiet place or a bright, vivacious environment conducive to their style and task at hand, ideally they’re going to be more productive in that hour,” Redmond says.
Some vendors now display live space availability on kiosks or apps, powered by people-counting sensors. Others send real-time air quality info to occupants’ smartphones. The key is self-service: rather than a manager constantly reallocating space or fielding comfort complaints, employees can respond to live data themselves (e.g. move to a less crowded area if noise is high or CO₂ is creeping up). This “pull” model empowers occupants to make decisions that the workplace team used to have to anticipate.
Finally, automation closes the loop on energy and sustainability goals. A great example is occupancy-driven ventilation. R-Zero recently rolled out occupancy-based demand-controlled ventilation (DCV) sequences. When sensors detect a space is empty, the HVAC automatically throttles down; when people return, setpoints and fresh air ramp up—all without facility staff intervention. They layer in IAQ monitoring to validate that air quality stays optimal when ventilation is reduced.
In the same vein, targeted pollutant sensing can enable advanced energy strategies. For instance, formaldehyde is a contaminant named in ASHRAE Standard 62.1 (Ventilation for Acceptable IAQ); if you can actively remove formaldehyde from indoor air (say with special filters) and prove it’s absence, you may be allowed to reduce the amount of outside air you bring in, significantly cutting energy. Redmond calls this “a really interesting and compelling energy efficiency story”—one that only emerges when IAQ data feeds straight into operational decisions or control sequences.
Not all sensors are created equal. The sensors you choose determine which questions you can answer—a lesson savvy teams have learned through trial and error. “Sensors aren’t a commodity yet… selection matters,” says Ryan. We’ve seen companies assume any sensor will meet their needs, only to find out (too late) that different technologies have very different capabilities and therefore enable different use cases.
The current trend is a more nuanced approach to sensor selection: start with the end in mind. What actions or insights do you eventually want? Ensure your hardware can deliver that. Take occupancy counting. You have a few broad options—optical, thermal, ultrasonic, infrared, or network-based sensors. Each comes with trade-offs in privacy, accuracy, granularity, cost, and speed.
Optical sensors are essentially smart cameras that use onboard AI to detect people. They can do fine-grained things like distinguish individuals, recognize if someone is at a desk vs. at a whiteboard, or even infer posture (sitting vs. standing). This opens up use cases beyond basic counting—one XY Sense customer is exploring a “sit/stand” wellness gamification, encouraging employees to stand 50% of the day by tracking behavior (a novel application that optical data makes possible).
The downsides? They require infrastructure (often one per ~1000 sqft mounted on ceilings) and can be pricey; and even though the data is anonymized, some organizations are wary of anything involving cameras.
By contrast, thermal sensors detect body heat and movement, but nothing identifiable. They are very privacy-friendly—essentially just heat signatures—and tend to be lower cost and easier to install (often battery powered, wireless). Historically, thermal sensors could count people in a zone, but the new generation is finding creative ways to glean insights from motion patterns. “We can use moving speed, gait, [and] how long people stay... Those data can be useful for different verticals,” Butlr co-founder Jiani Zeng told us.
For instance, in a workplace you might detect how long someone remains seated in a focus area vs. how often they get up—a measure of utilization and even engagement. In healthcare, thermal sensors can flag if a patient has been in bed too long or hasn’t visited the bathroom. Butlr’s vision is to be “the foundation… to understand people movement data” across contexts. The trade-off: thermal lacks the detail of optical (you won’t know what people are doing, just that they are there and moving).
A third approach gaining popularity is leveraging existing Wi-Fi networks to sense occupancy. Companies like InnerSpace take a “sensor-free” approach, listening to Wi-Fi signals from smartphones and laptops to infer where people are. The obvious advantage is you don’t have to deploy tons of new hardware—your existing Wi-Fi access points become the sensors.
Wi-Fi analytics can cover entire buildings (signals go through walls) and identify repeat presence by the same device, enabling individual-level patterns without knowing anyone’s name. InnerSpace’s platform, for example, can tell how often Employee A (anonymized) comes in, how long they dwell in various neighborhoods, and even automatically cluster people into cohorts based on their behavior (e.g. a group that always collaborates in the cafe vs. a group that stays at their desks.
This goes far beyond the “average occupancy” stats of old. With it, you can uncover that the sales team consistently uses the social lounge on Fridays while the engineering team prefers quiet zones—insights that inform space planning for each group. The downside to Wi-Fi-based occupancy is often a hit to fidelity: it may not be as accurate or immediate as dedicated sensors and will miss nuances like passive occupancy that an optical sensor might track.
There’s also the caveat that not everyone carries a Wi-Fi device or has it on, so you might miss visitors or edge cases. But Innerspace’s Wu is finding that, “businesses now want more nuanced behavior data that Wi-Fi can provide.” The good news is that these technologies are not mutually exclusive; some firms use a mix (e.g. optical in key areas and Wi-Fi analytics as a net across the whole building).
Your choice of sensor tech should map to the use cases and level of detail you need. “You need to think about what you might want to do. If you want to get into behaviors and you haven’t picked the right sensor, that doesn’t help,” advises Ryan.
In practice, this means engaging stakeholders upfront and learning from early adopters. If occupancy data will feed into real-time occupant-facing apps, favor sensors with instant communication. If your leadership cares about wellness and how active employees are, consider sensors that can measure something like sitting time or posture. If space planning for different departments is a goal, maybe a Wi-Fi approach to identify team patterns makes sense.
The same logic applies to IAQ sensors. These devices come in all flavors—some just measure CO₂ and temperature, others also measure VOCs, particulate matter (PM2.5), humidity, noise, light, etc. A few are “all-in-one” IEQ nodes that even include occupancy or people-counting.
First ask: what decisions or actions will this enable? Redmond noted that some R-Zero customers find more value in “picking apart the parameters and really applying each for specific use cases” rather than chasing a generic IAQ index. Conversely, if your main concern is occupant comfort/health, you’ll want a broader spectrum of IAQ metrics (CO₂ for freshness, PM for pollution, humidity and temp for thermal comfort, maybe even noise for acoustics).
“On the integration side of things, sensor selection matters. If you go with [sensors from proprietary BAS systems] and you switch control systems, you might lose the ability to pull those readings into a different BAS,” cautions Kyle Megna of Airthings. In other words, an IAQ device that only talks to one type of supervisory controller (let alone cloud environment) might leave your data in a silo (we’ll get to silos next) unless you plan ahead.
IoT sensors are not yet one-size-fits-all commodities. A thoughtful selection upfront prevents regret later. As Ryan sums up, choosing correctly means unlocking the use cases you care about; choosing poorly could mean hitting a dead end. We’re even seeing vendors partner up to cover each other’s gaps (XY Sense, for example, partners with Airthings and Kaiterra to overlay IAQ data on occupancy maps; Butlr and InnerSpace partner to combine thermal and wifi-based occupancy analytics). The message: match the tool to the job.
[join Nexus Pro to keep reading the final 2 trends, plus unlock all past articles and event recordings!]
Even the best sensors, taken in isolation, only tell part of the story. The next trend is combining data streams and adding context—essentially, breaking the silos between occupancy, environmental data, base building systems, and other workplace data.
Beyond manual observations, the technology architecture trend is toward integrated data layers that pull in multiple streams into one data model, rather than jumping between an IAQ dashboard, an occupancy heatmap, the badge swipe logs, and a survey tool. Eliminating silos also means connecting occupancy/IAQ systems with the base building systems (BMS, HVAC controls, room booking, etc.) as part of a unified tech stack.
Imagine a 3D model of your floor plan where every sensor reading, work order, and piece of metadata (room capacity, equipment info) is pinned to its location in real time. Sauer mentioned that younger facility professionals (raised on video games and 3D tech) are accelerating a shift to more visual, integrated ops tools.
This is happening gradually. Sauer paints a forward-looking picture: “Say I’ve just scheduled a 100-person all-hands event. Based on our integrated data, we know what usually happens—we need to bring the temperature down by 10 degrees beforehand because otherwise it’ll spike, and we know CO₂ will rise, so we increase air changes. We can much more efficiently operate the building and ensure a healthy environment for that event.” It’s a far cry from the siloed approach of the past where HVAC ran on fixed schedules oblivious to actual room usage.
Another area where integration is showing dividends is in holistic environmental analytics. Some workplace IoT providers now combine all IEQ factors—including air, temperature, light, sound— alongside occupancy. The rationale is that productivity and experience are multifaceted.
Redmond described how R-Zero’s customers have been experimenting with space effectiveness metrics: for example, identifying the mix of conditions in a space (noise level, brightness, air quality, occupancy density) that correlates with high utilization and positive user feedback—essentially, what makes a space productive.
“We can at least begin to report on how productive an environment is with the combination of these data sets,” notes Redmond. A concrete case: her team found a meeting room that was almost never used after 1pm. Utilization data alone flagged the underuse, but by adding context —the room’s temperature and ambient light data—they discovered a big afternoon spike due to solar gain. In that case, the fix was adjusting HVAC settings (and planning for better shading in a renovation)—an actionable outcome only revealed by overlaying occupancy with environmental context.
As you can tell by now, technology hasn’t yet progressed to solve this problem, my friends. The reality is people are more important than ever in making sense of workplace IoT data. The final trend is an acknowledgement that expert interpretation and change management are the linchpins of turning data into meaningful action.
One practical example is the marriage of sensor data with old-fashioned workplace research. One workplace team I studied doesn’t rely on sensors alone to evaluate space usage. The sensor data is just one piece of the puzzle, combined with observational studies, employee surveys, focus groups, and before/after studies around any big workplace change.
This team might deploy people-counting sensors in a new collaboration space for a few months, but they’ll also interview users and observe behaviors. This helps them discern if the design intent is being met. Are people actually collaborating there, or just taking private calls? The blend of quantitative and qualitative data provides far richer insight than either alone.
As Kevin Sauer noted, many big companies that rolled out sensors everywhere are now circling back to do exactly this —“going back and saying, okay, let’s do these observation studies, because the data is good… but we don’t know what’s happening in the space. We actually want to know that human activity.”
In other words, occupancy counts told them where and when space was used, but not how or to what end. By correlating sensor metrics with direct observations, they can finally answer the “so what”—e.g. a meeting room averages 2 people not because of lack of interest, but because people use it as a quiet drop-in spot (not its intended purpose!). Armed with that knowledge, the team can adjust the space or educate users.
One hindrance in expecting humans to glean insights is that few humans have the data science background needed to do it. Kevin Sauer points out that workplace teams span a wide range of technical expertise and organizational homes: “Most traditional companies [workplace] would lead up to the CFO, so it’s always finance focused. But now some report into HR… interested in employee experience, but they’re probably even less technically adept… then some report into IT. So you’ve got a whole spectrum: soft skills, hard skills, fully tech-ready, completely inept at technology.”
It’s a blunt observation, but it rings true. Some teams are very data-savvy, others are not. This is why we see a rise in either hiring or outsourcing for a “data translator” role. “A lot of companies right now got sensors, got the data. It’s like, what’s it mean? They don’t have a translator,” says Sauer. “There’s no one there going, ‘Okay, this is what we see… and here’s what we suggest you do about it.’” Without that translation step, even the best data can languish.
What does a “translator” do? In our conversations, this ranged from internal analysts on the corporate real estate team to external consultants or services from the sensor vendor. The translator figures out the so what: they turn those utilization and IEQ reports into a narrative and recommendations that the business can act on.
For example, an analyst like Sauer might reveal, “I’ve evaluated your portfolio, and you still build all these giant meeting rooms. You never occupy them. So let’s change that in your design guidelines.” He’d also note, “we see a lot of one-person-in-a-meeting-room habits—let’s do some change management and etiquette around that, help them understand they’re wasting a very expensive resource.” These are human-centric solutions (policy, design standard changes) derived from data, and it takes an experienced eye to draw them out.
There’s also the task of “humanizing” the data—cleaning it, contextualizing it, and avoiding misinterpretation. Sauer gave a striking example from his tenure at Amazon: they pulled badge swipe data for 380,000 employees to gauge office attendance, and the raw data made it look like only a small fraction of employees were coming in. But the savvy analyst realized many people in that timeframe were new hires, recent exits, or on leave – i.e. not eligible to be in-office. After filtering those out, the attendance rate was much higher among eligible employees. Had they taken the data at face value, leadership might have overreacted.
Because many organizations lack this kind of expertise in-house, a burgeoning service industry is stepping in, of which Sauer is a part of. Major real estate firms (JLL, CBRE, Cushman & Wakefield, etc.) have workplace analytics teams that will take a client’s sensor data and produce a report with findings and recommendations. Even furniture dealers have started using utilization sensors to advise clients on how their products (like expensive collaboration furniture or phone booths) are actually being used.
At peak IoT sensor hype, some thought software could replace consultants, but instead the leading vendors are partnering with them or emulating them. R-Zero’s team acknowledged that a “better business path” is often to partner with a workplace consultant who brings the context of the client’s goals and can drive the changes coming out of the data.
Sensor vendors themselves are also adding data insight services on top of their software. Butlr is one example: Co-founder Jiani Zeng told us about their “data insights report” offering. “We have our own data science team,” she says, “we take the data from our product and do a very scientific analysis with them to surface the insights.”
Butlr’s team will analyze trends for the customer—e.g. identifying which days or months have high vs. low occupancy, which zones are most popular, etc. Crucially, they tie it back to decisions. Zeng described a global client considering a new type of adjustable desk. “If you want to implement a certain new device or furniture for your company, how do you know it’s actually being used by people?” she poses.
By deploying sensors in a pilot area, they gathered data on how many hours those desks were used vs. others and helped the client decide whether to invest company-wide. Zeng summed up the impact: The client may not have had the time or skills to derive that on their own, but with a packaged report, the value of the data is realized.
It’s worth noting that “keeping humans in the loop” doesn’t just mean data scientists —it also means engaging the people actually using the spaces. Workplace strategists often act as the bridge between data and human experience. They validate sensor findings with employee feedback (e.g. “Yes, that room is unpopular because it gets too hot in the afternoon”) and then drive the change management to address it (install shades, communicate the fix to employees, etc.).
An era of more actionable insights from IoT deployments is beginning to take shape. It’s clear that technology alone isn’t a silver bullet, but it drives value when combined with the right approach.
Instead of sensors as gimmicks or dashboards as “luxury”—we’re seeing tangible outcomes: underused spaces repurposed, overcrowded areas relieved, energy wasted on empty rooms recaptured, air quality issues preempted, and workplaces tuned to actual human behavior. The result is work environments that aren’t just instrumented for data’s sake, but are actively improving day by day based on what the data tells us.
In the end, curing IoT data overwhelm is about closing the loop: sensors -> insights -> actions -> better workplaces. It’s a journey many of us are still navigating.
Even the best sensors, taken in isolation, only tell part of the story. The next trend is combining data streams and adding context—essentially, breaking the silos between occupancy, environmental data, base building systems, and other workplace data.
Beyond manual observations, the technology architecture trend is toward integrated data layers that pull in multiple streams into one data model, rather than jumping between an IAQ dashboard, an occupancy heatmap, the badge swipe logs, and a survey tool. Eliminating silos also means connecting occupancy/IAQ systems with the base building systems (BMS, HVAC controls, room booking, etc.) as part of a unified tech stack.
Imagine a 3D model of your floor plan where every sensor reading, work order, and piece of metadata (room capacity, equipment info) is pinned to its location in real time. Sauer mentioned that younger facility professionals (raised on video games and 3D tech) are accelerating a shift to more visual, integrated ops tools.
This is happening gradually. Sauer paints a forward-looking picture: “Say I’ve just scheduled a 100-person all-hands event. Based on our integrated data, we know what usually happens—we need to bring the temperature down by 10 degrees beforehand because otherwise it’ll spike, and we know CO₂ will rise, so we increase air changes. We can much more efficiently operate the building and ensure a healthy environment for that event.” It’s a far cry from the siloed approach of the past where HVAC ran on fixed schedules oblivious to actual room usage.
Another area where integration is showing dividends is in holistic environmental analytics. Some workplace IoT providers now combine all IEQ factors—including air, temperature, light, sound— alongside occupancy. The rationale is that productivity and experience are multifaceted.
Redmond described how R-Zero’s customers have been experimenting with space effectiveness metrics: for example, identifying the mix of conditions in a space (noise level, brightness, air quality, occupancy density) that correlates with high utilization and positive user feedback—essentially, what makes a space productive.
“We can at least begin to report on how productive an environment is with the combination of these data sets,” notes Redmond. A concrete case: her team found a meeting room that was almost never used after 1pm. Utilization data alone flagged the underuse, but by adding context —the room’s temperature and ambient light data—they discovered a big afternoon spike due to solar gain. In that case, the fix was adjusting HVAC settings (and planning for better shading in a renovation)—an actionable outcome only revealed by overlaying occupancy with environmental context.
As you can tell by now, technology hasn’t yet progressed to solve this problem, my friends. The reality is people are more important than ever in making sense of workplace IoT data. The final trend is an acknowledgement that expert interpretation and change management are the linchpins of turning data into meaningful action.
One practical example is the marriage of sensor data with old-fashioned workplace research. One workplace team I studied doesn’t rely on sensors alone to evaluate space usage. The sensor data is just one piece of the puzzle, combined with observational studies, employee surveys, focus groups, and before/after studies around any big workplace change.
This team might deploy people-counting sensors in a new collaboration space for a few months, but they’ll also interview users and observe behaviors. This helps them discern if the design intent is being met. Are people actually collaborating there, or just taking private calls? The blend of quantitative and qualitative data provides far richer insight than either alone.
As Kevin Sauer noted, many big companies that rolled out sensors everywhere are now circling back to do exactly this —“going back and saying, okay, let’s do these observation studies, because the data is good… but we don’t know what’s happening in the space. We actually want to know that human activity.”
In other words, occupancy counts told them where and when space was used, but not how or to what end. By correlating sensor metrics with direct observations, they can finally answer the “so what”—e.g. a meeting room averages 2 people not because of lack of interest, but because people use it as a quiet drop-in spot (not its intended purpose!). Armed with that knowledge, the team can adjust the space or educate users.
One hindrance in expecting humans to glean insights is that few humans have the data science background needed to do it. Kevin Sauer points out that workplace teams span a wide range of technical expertise and organizational homes: “Most traditional companies [workplace] would lead up to the CFO, so it’s always finance focused. But now some report into HR… interested in employee experience, but they’re probably even less technically adept… then some report into IT. So you’ve got a whole spectrum: soft skills, hard skills, fully tech-ready, completely inept at technology.”
It’s a blunt observation, but it rings true. Some teams are very data-savvy, others are not. This is why we see a rise in either hiring or outsourcing for a “data translator” role. “A lot of companies right now got sensors, got the data. It’s like, what’s it mean? They don’t have a translator,” says Sauer. “There’s no one there going, ‘Okay, this is what we see… and here’s what we suggest you do about it.’” Without that translation step, even the best data can languish.
What does a “translator” do? In our conversations, this ranged from internal analysts on the corporate real estate team to external consultants or services from the sensor vendor. The translator figures out the so what: they turn those utilization and IEQ reports into a narrative and recommendations that the business can act on.
For example, an analyst like Sauer might reveal, “I’ve evaluated your portfolio, and you still build all these giant meeting rooms. You never occupy them. So let’s change that in your design guidelines.” He’d also note, “we see a lot of one-person-in-a-meeting-room habits—let’s do some change management and etiquette around that, help them understand they’re wasting a very expensive resource.” These are human-centric solutions (policy, design standard changes) derived from data, and it takes an experienced eye to draw them out.
There’s also the task of “humanizing” the data—cleaning it, contextualizing it, and avoiding misinterpretation. Sauer gave a striking example from his tenure at Amazon: they pulled badge swipe data for 380,000 employees to gauge office attendance, and the raw data made it look like only a small fraction of employees were coming in. But the savvy analyst realized many people in that timeframe were new hires, recent exits, or on leave – i.e. not eligible to be in-office. After filtering those out, the attendance rate was much higher among eligible employees. Had they taken the data at face value, leadership might have overreacted.
Because many organizations lack this kind of expertise in-house, a burgeoning service industry is stepping in, of which Sauer is a part of. Major real estate firms (JLL, CBRE, Cushman & Wakefield, etc.) have workplace analytics teams that will take a client’s sensor data and produce a report with findings and recommendations. Even furniture dealers have started using utilization sensors to advise clients on how their products (like expensive collaboration furniture or phone booths) are actually being used.
At peak IoT sensor hype, some thought software could replace consultants, but instead the leading vendors are partnering with them or emulating them. R-Zero’s team acknowledged that a “better business path” is often to partner with a workplace consultant who brings the context of the client’s goals and can drive the changes coming out of the data.
Sensor vendors themselves are also adding data insight services on top of their software. Butlr is one example: Co-founder Jiani Zeng told us about their “data insights report” offering. “We have our own data science team,” she says, “we take the data from our product and do a very scientific analysis with them to surface the insights.”
Butlr’s team will analyze trends for the customer—e.g. identifying which days or months have high vs. low occupancy, which zones are most popular, etc. Crucially, they tie it back to decisions. Zeng described a global client considering a new type of adjustable desk. “If you want to implement a certain new device or furniture for your company, how do you know it’s actually being used by people?” she poses.
By deploying sensors in a pilot area, they gathered data on how many hours those desks were used vs. others and helped the client decide whether to invest company-wide. Zeng summed up the impact: The client may not have had the time or skills to derive that on their own, but with a packaged report, the value of the data is realized.
It’s worth noting that “keeping humans in the loop” doesn’t just mean data scientists —it also means engaging the people actually using the spaces. Workplace strategists often act as the bridge between data and human experience. They validate sensor findings with employee feedback (e.g. “Yes, that room is unpopular because it gets too hot in the afternoon”) and then drive the change management to address it (install shades, communicate the fix to employees, etc.).
An era of more actionable insights from IoT deployments is beginning to take shape. It’s clear that technology alone isn’t a silver bullet, but it drives value when combined with the right approach.
Instead of sensors as gimmicks or dashboards as “luxury”—we’re seeing tangible outcomes: underused spaces repurposed, overcrowded areas relieved, energy wasted on empty rooms recaptured, air quality issues preempted, and workplaces tuned to actual human behavior. The result is work environments that aren’t just instrumented for data’s sake, but are actively improving day by day based on what the data tells us.
In the end, curing IoT data overwhelm is about closing the loop: sensors -> insights -> actions -> better workplaces. It’s a journey many of us are still navigating.
Even the best sensors, taken in isolation, only tell part of the story. The next trend is combining data streams and adding context—essentially, breaking the silos between occupancy, environmental data, base building systems, and other workplace data.
Beyond manual observations, the technology architecture trend is toward integrated data layers that pull in multiple streams into one data model, rather than jumping between an IAQ dashboard, an occupancy heatmap, the badge swipe logs, and a survey tool. Eliminating silos also means connecting occupancy/IAQ systems with the base building systems (BMS, HVAC controls, room booking, etc.) as part of a unified tech stack.
Imagine a 3D model of your floor plan where every sensor reading, work order, and piece of metadata (room capacity, equipment info) is pinned to its location in real time. Sauer mentioned that younger facility professionals (raised on video games and 3D tech) are accelerating a shift to more visual, integrated ops tools.
This is happening gradually. Sauer paints a forward-looking picture: “Say I’ve just scheduled a 100-person all-hands event. Based on our integrated data, we know what usually happens—we need to bring the temperature down by 10 degrees beforehand because otherwise it’ll spike, and we know CO₂ will rise, so we increase air changes. We can much more efficiently operate the building and ensure a healthy environment for that event.” It’s a far cry from the siloed approach of the past where HVAC ran on fixed schedules oblivious to actual room usage.
Another area where integration is showing dividends is in holistic environmental analytics. Some workplace IoT providers now combine all IEQ factors—including air, temperature, light, sound— alongside occupancy. The rationale is that productivity and experience are multifaceted.
Redmond described how R-Zero’s customers have been experimenting with space effectiveness metrics: for example, identifying the mix of conditions in a space (noise level, brightness, air quality, occupancy density) that correlates with high utilization and positive user feedback—essentially, what makes a space productive.
“We can at least begin to report on how productive an environment is with the combination of these data sets,” notes Redmond. A concrete case: her team found a meeting room that was almost never used after 1pm. Utilization data alone flagged the underuse, but by adding context —the room’s temperature and ambient light data—they discovered a big afternoon spike due to solar gain. In that case, the fix was adjusting HVAC settings (and planning for better shading in a renovation)—an actionable outcome only revealed by overlaying occupancy with environmental context.
As you can tell by now, technology hasn’t yet progressed to solve this problem, my friends. The reality is people are more important than ever in making sense of workplace IoT data. The final trend is an acknowledgement that expert interpretation and change management are the linchpins of turning data into meaningful action.
One practical example is the marriage of sensor data with old-fashioned workplace research. One workplace team I studied doesn’t rely on sensors alone to evaluate space usage. The sensor data is just one piece of the puzzle, combined with observational studies, employee surveys, focus groups, and before/after studies around any big workplace change.
This team might deploy people-counting sensors in a new collaboration space for a few months, but they’ll also interview users and observe behaviors. This helps them discern if the design intent is being met. Are people actually collaborating there, or just taking private calls? The blend of quantitative and qualitative data provides far richer insight than either alone.
As Kevin Sauer noted, many big companies that rolled out sensors everywhere are now circling back to do exactly this —“going back and saying, okay, let’s do these observation studies, because the data is good… but we don’t know what’s happening in the space. We actually want to know that human activity.”
In other words, occupancy counts told them where and when space was used, but not how or to what end. By correlating sensor metrics with direct observations, they can finally answer the “so what”—e.g. a meeting room averages 2 people not because of lack of interest, but because people use it as a quiet drop-in spot (not its intended purpose!). Armed with that knowledge, the team can adjust the space or educate users.
One hindrance in expecting humans to glean insights is that few humans have the data science background needed to do it. Kevin Sauer points out that workplace teams span a wide range of technical expertise and organizational homes: “Most traditional companies [workplace] would lead up to the CFO, so it’s always finance focused. But now some report into HR… interested in employee experience, but they’re probably even less technically adept… then some report into IT. So you’ve got a whole spectrum: soft skills, hard skills, fully tech-ready, completely inept at technology.”
It’s a blunt observation, but it rings true. Some teams are very data-savvy, others are not. This is why we see a rise in either hiring or outsourcing for a “data translator” role. “A lot of companies right now got sensors, got the data. It’s like, what’s it mean? They don’t have a translator,” says Sauer. “There’s no one there going, ‘Okay, this is what we see… and here’s what we suggest you do about it.’” Without that translation step, even the best data can languish.
What does a “translator” do? In our conversations, this ranged from internal analysts on the corporate real estate team to external consultants or services from the sensor vendor. The translator figures out the so what: they turn those utilization and IEQ reports into a narrative and recommendations that the business can act on.
For example, an analyst like Sauer might reveal, “I’ve evaluated your portfolio, and you still build all these giant meeting rooms. You never occupy them. So let’s change that in your design guidelines.” He’d also note, “we see a lot of one-person-in-a-meeting-room habits—let’s do some change management and etiquette around that, help them understand they’re wasting a very expensive resource.” These are human-centric solutions (policy, design standard changes) derived from data, and it takes an experienced eye to draw them out.
There’s also the task of “humanizing” the data—cleaning it, contextualizing it, and avoiding misinterpretation. Sauer gave a striking example from his tenure at Amazon: they pulled badge swipe data for 380,000 employees to gauge office attendance, and the raw data made it look like only a small fraction of employees were coming in. But the savvy analyst realized many people in that timeframe were new hires, recent exits, or on leave – i.e. not eligible to be in-office. After filtering those out, the attendance rate was much higher among eligible employees. Had they taken the data at face value, leadership might have overreacted.
Because many organizations lack this kind of expertise in-house, a burgeoning service industry is stepping in, of which Sauer is a part of. Major real estate firms (JLL, CBRE, Cushman & Wakefield, etc.) have workplace analytics teams that will take a client’s sensor data and produce a report with findings and recommendations. Even furniture dealers have started using utilization sensors to advise clients on how their products (like expensive collaboration furniture or phone booths) are actually being used.
At peak IoT sensor hype, some thought software could replace consultants, but instead the leading vendors are partnering with them or emulating them. R-Zero’s team acknowledged that a “better business path” is often to partner with a workplace consultant who brings the context of the client’s goals and can drive the changes coming out of the data.
Sensor vendors themselves are also adding data insight services on top of their software. Butlr is one example: Co-founder Jiani Zeng told us about their “data insights report” offering. “We have our own data science team,” she says, “we take the data from our product and do a very scientific analysis with them to surface the insights.”
Butlr’s team will analyze trends for the customer—e.g. identifying which days or months have high vs. low occupancy, which zones are most popular, etc. Crucially, they tie it back to decisions. Zeng described a global client considering a new type of adjustable desk. “If you want to implement a certain new device or furniture for your company, how do you know it’s actually being used by people?” she poses.
By deploying sensors in a pilot area, they gathered data on how many hours those desks were used vs. others and helped the client decide whether to invest company-wide. Zeng summed up the impact: The client may not have had the time or skills to derive that on their own, but with a packaged report, the value of the data is realized.
It’s worth noting that “keeping humans in the loop” doesn’t just mean data scientists —it also means engaging the people actually using the spaces. Workplace strategists often act as the bridge between data and human experience. They validate sensor findings with employee feedback (e.g. “Yes, that room is unpopular because it gets too hot in the afternoon”) and then drive the change management to address it (install shades, communicate the fix to employees, etc.).
An era of more actionable insights from IoT deployments is beginning to take shape. It’s clear that technology alone isn’t a silver bullet, but it drives value when combined with the right approach.
Instead of sensors as gimmicks or dashboards as “luxury”—we’re seeing tangible outcomes: underused spaces repurposed, overcrowded areas relieved, energy wasted on empty rooms recaptured, air quality issues preempted, and workplaces tuned to actual human behavior. The result is work environments that aren’t just instrumented for data’s sake, but are actively improving day by day based on what the data tells us.
In the end, curing IoT data overwhelm is about closing the loop: sensors -> insights -> actions -> better workplaces. It’s a journey many of us are still navigating.
“Wednesday afternoons saw high demand for 2-person huddle rooms, but low engagement in larger meeting spaces. So the team rebalanced room types to match real patterns of work.” If that sounds like an obvious outcome of IoT deployments, why do so many workplace teams struggle to get there?
In the last few years, corporate occupiers invested heavily in Workplace IoT—networks of occupancy counters and indoor air quality (IAQ) monitors—to navigate post-pandemic-era challenges. We counted people to support hybrid work, and we measured CO₂ to reassure everyone the air was healthy.
A 2022 JLL study reported that 56% of companies planned to invest in environment and occupancy sensors by this year, but those investments have left workplace teams swimming in data (and sometimes old batteries) without a clear plan for what comes next. As James Wu, CEO of InnerSpace, puts it, “we’ve arrived at a ‘so what?’ moment” for workplace IoT.
The pressure is on to translate all this raw sensor data into safer, more effective, more engaging workplaces—and to prove it’s worth the ongoing cost of the IoT stack. Simply knowing how many people are in the office or seeing a dashboard of CO₂ levels isn’t enough. Up to 88% of IoT data in organizations goes unused, and workplace leaders are acutely feeling that statistic.
Michael Krall of JP Morgan Chase framed the market’s challenge well: “The real question is who can quickly tell me what to do with the data and simplify the next step to do it. If [solutions] are just dashboards or heatmaps, then they’re a luxury.” In other words, if occupancy/IAQ sensors don’t lead to action, why bother?
Yet, we’re optimistic. Through conversations with workplace pros and tech vendors at the forefront (Sauer Strategy Works, Butlr, XY Sense, R-Zero, InnerSpace, and Airthings—plus a workplace team that will remain anonymous), a picture emerges where data overload gives way to measurably better workplaces. In this next phase, the focus shifts from where people are to what they’re doing, what the environment is like, and what to do about it.
Five key trends are helping workplace teams cure the data overwhelm and connect the dots from sensors to outcomes. We explore each below, with real examples from the workplace frontlines.
Our biggest surprise? We thought we’d hear about the inroads AI is making. Get data to the cloud and let the software tell the user what they need to know… right?
Wrong.
As we’ve lamented over the years, it’s not uncommon for workplace and facilities teams to inherit sprawling dashboards filled with charts, filters, and generic reports—none of which they have time to navigate. “The industry is sick of visualizations and alerts,” one workplace leader told us. “And all the dashboards that are too complicated and crowded.”
Despite this fatigue, vendors still believe they can help users find the signal in the noise. But that doesn’t mean AI is ready to jump in and solve things. Every vendor we spoke with is experimenting with AI, but none claim it’s the silver bullet. Elizabeth Redmond of R-Zero explained that clients often have very specific questions that are customized to their needs, which makes it difficult for AI to come up with insights on its own at scale.
James Wu, CEO of InnerSpace, said a lot of vendors are exploring allowing users to ask plain-English questions like “Which spaces were most used last quarter?” But he cautioned that it may “just move the problem.” If users don’t know what to ask, a blank AI search bar doesn’t help.
Right now, anomaly detection is a promising use case for R-Zero. “You’re really looking for outliers and anomalies,” Redmond said. Instead of making users dig through trend lines, the platform flags what’s unusual—like a floor that suddenly jumps to 80% occupancy on a Tuesday or a room that goes unused for a week.
Similarly, InnerSpace is prototyping AI that spots changes in behavior—like a team suddenly using less space—and suggests explanations, such as a new work policy. While still early, that kind of contextual AI could help workplace teams move from reacting to predicting.
Former Amazon workplace strategist Kevin Sauer, who’s now an independent consultant, compared it to Google Maps: “You can plug in your route, and it predicts what traffic will look like.” In a workplace, that might mean forecasting low turnout before a holiday and reducing catering orders, or anticipating high CO₂ levels during an all-hands meeting and preemptively boosting ventilation. These kinds of proactive nudges are still early-stage—but they point toward a future where systems flag issues before they happen.
In the absence of application layer AI breakthroughs, vendors have been making up ground by improving visualizations and analytics. Shivaun Ryan of XY Sense told us about their new animated historical replay visualizations. Instead of forcing users to scroll through static heatmaps or pivot tables, platforms like XY Sense now allow teams to watch time-lapse maps of office usage. These replays show how people move through space during the day, where congestion happens, and when zones go quiet.
“You don’t need to be a data analyst to use this,” said Shivaun Ryan of XY Sense. “That’s the role of the product.” Their clients are using replays to identify patterns—like which desks never get touched near the café—and to share stories with stakeholders who might never open a dashboard. “Movement replays are universally understandable.” It’s the kind of feature that helps a workplace team say to HR or Finance: “Here’s how your teams are actually using the space”—with a GIF, not a spreadsheet.
Another workplace leader described how their team worked closely with their space utilization software vendor to enhance visualizations and add filters that matched their analytics needs. That meant being able to slice data by neighborhood or day of week, not just room, and export clean insights for design reviews and leadership presentations.
Similarly, Butlr offers clients the ability to tag and group spaces into “virtual zones”—for example, clustering all phone booths across a floor to see aggregate usage. These enhancements reduce the need for manual Excel work and help the software answer real workplace questions like “How are our quiet zones performing?”
When it comes to IAQ, simplifying the data is even more critical. A CO₂ alert in Room 501 might mean open a door, run ventilation, or nothing at all. Without context, alerts create noise. Vendors like Airthings are refining alerts to focus on duration above thresholds, not just spikes, to determine the long-term performance of rooms and spaces for data-driven decisions. They’re also introducing IEQ (Indoor Environmental Quality) grading systems—like an “A” to “F” score for each space—that summarize multiple parameters into a single signal. This echoes how WELL or LEED certifications communicate performance.
The goal is simple: help a facilities manager or HR partner quickly assess, “Is this space healthy or not?” without needing a crash course in VOCs and PM2.5.
One not-so-obvious way to avoid overwhelm: we don’t always need humans in the loop to make use of sensor data. Prioritize sensor use cases that trigger actions automatically or surface information directly to the people who need it, in real time, so the workplace team isn’t stuck poring over charts.
“Technology should be passive... and make everyone’s job easier,” says Shivaun Ryan of XY Sense, which makes optical people-counters and software that communicates real-time space info to occupants who can choose where they want to work.
In practice, this means occupancy and air quality data should immediately translate into something useful—whether it’s an automatic HVAC adjustment or a live map that helps an employee find a free desk.
Real-time integrations are key here. Consider the common problem of conference room “no-shows” (ghost meetings). Traditional occupancy sensors might free up a room after a few minutes of no presence—but if the system only polls every 5 minutes, that’s a clunky experience when someone shows up a little late to find their room already released.
For example, an optical sensor can tell when a room is truly empty and trigger the reservation system to release it without a lag, or conversely detect “passive occupancy” (when someone’s stepped out to grab coffee) so it doesn’t mistakenly count the room as empty.
With air quality, automation is just as critical. Raw IAQ data by itself isn’t very actionable to a workplace strategist—what is useful is when that data drives building systems or maintenance teams’ to-do lists. Andrew Knueppel from Cushman Wakefield emphasized integrating IAQ with Fault Detection & Diagnostics (FDD) software to “actually diagnose a problem and trigger a work order,” skipping the step of a human analyzing a graph.
Landlords are on board too: some are installing IAQ monitors in their buildings specifically to catch comfort issues early. “If they can get ahead of those comfort complaints and build that into their predictive process, that’s extremely valuable,” adds Elizabeth Redmond of R-Zero, noting it helps landlords who manage multiple sites with lean staff.
Redmond emphasized the importance of putting a more comprehensive set of IEQ directly into employees’ hands to improve their day-to-day experience and support their neurodiversity. “If you can give occupants insight into what conditions are present in the office, so they can look for a dark and quiet place or a bright, vivacious environment conducive to their style and task at hand, ideally they’re going to be more productive in that hour,” Redmond says.
Some vendors now display live space availability on kiosks or apps, powered by people-counting sensors. Others send real-time air quality info to occupants’ smartphones. The key is self-service: rather than a manager constantly reallocating space or fielding comfort complaints, employees can respond to live data themselves (e.g. move to a less crowded area if noise is high or CO₂ is creeping up). This “pull” model empowers occupants to make decisions that the workplace team used to have to anticipate.
Finally, automation closes the loop on energy and sustainability goals. A great example is occupancy-driven ventilation. R-Zero recently rolled out occupancy-based demand-controlled ventilation (DCV) sequences. When sensors detect a space is empty, the HVAC automatically throttles down; when people return, setpoints and fresh air ramp up—all without facility staff intervention. They layer in IAQ monitoring to validate that air quality stays optimal when ventilation is reduced.
In the same vein, targeted pollutant sensing can enable advanced energy strategies. For instance, formaldehyde is a contaminant named in ASHRAE Standard 62.1 (Ventilation for Acceptable IAQ); if you can actively remove formaldehyde from indoor air (say with special filters) and prove it’s absence, you may be allowed to reduce the amount of outside air you bring in, significantly cutting energy. Redmond calls this “a really interesting and compelling energy efficiency story”—one that only emerges when IAQ data feeds straight into operational decisions or control sequences.
Not all sensors are created equal. The sensors you choose determine which questions you can answer—a lesson savvy teams have learned through trial and error. “Sensors aren’t a commodity yet… selection matters,” says Ryan. We’ve seen companies assume any sensor will meet their needs, only to find out (too late) that different technologies have very different capabilities and therefore enable different use cases.
The current trend is a more nuanced approach to sensor selection: start with the end in mind. What actions or insights do you eventually want? Ensure your hardware can deliver that. Take occupancy counting. You have a few broad options—optical, thermal, ultrasonic, infrared, or network-based sensors. Each comes with trade-offs in privacy, accuracy, granularity, cost, and speed.
Optical sensors are essentially smart cameras that use onboard AI to detect people. They can do fine-grained things like distinguish individuals, recognize if someone is at a desk vs. at a whiteboard, or even infer posture (sitting vs. standing). This opens up use cases beyond basic counting—one XY Sense customer is exploring a “sit/stand” wellness gamification, encouraging employees to stand 50% of the day by tracking behavior (a novel application that optical data makes possible).
The downsides? They require infrastructure (often one per ~1000 sqft mounted on ceilings) and can be pricey; and even though the data is anonymized, some organizations are wary of anything involving cameras.
By contrast, thermal sensors detect body heat and movement, but nothing identifiable. They are very privacy-friendly—essentially just heat signatures—and tend to be lower cost and easier to install (often battery powered, wireless). Historically, thermal sensors could count people in a zone, but the new generation is finding creative ways to glean insights from motion patterns. “We can use moving speed, gait, [and] how long people stay... Those data can be useful for different verticals,” Butlr co-founder Jiani Zeng told us.
For instance, in a workplace you might detect how long someone remains seated in a focus area vs. how often they get up—a measure of utilization and even engagement. In healthcare, thermal sensors can flag if a patient has been in bed too long or hasn’t visited the bathroom. Butlr’s vision is to be “the foundation… to understand people movement data” across contexts. The trade-off: thermal lacks the detail of optical (you won’t know what people are doing, just that they are there and moving).
A third approach gaining popularity is leveraging existing Wi-Fi networks to sense occupancy. Companies like InnerSpace take a “sensor-free” approach, listening to Wi-Fi signals from smartphones and laptops to infer where people are. The obvious advantage is you don’t have to deploy tons of new hardware—your existing Wi-Fi access points become the sensors.
Wi-Fi analytics can cover entire buildings (signals go through walls) and identify repeat presence by the same device, enabling individual-level patterns without knowing anyone’s name. InnerSpace’s platform, for example, can tell how often Employee A (anonymized) comes in, how long they dwell in various neighborhoods, and even automatically cluster people into cohorts based on their behavior (e.g. a group that always collaborates in the cafe vs. a group that stays at their desks.
This goes far beyond the “average occupancy” stats of old. With it, you can uncover that the sales team consistently uses the social lounge on Fridays while the engineering team prefers quiet zones—insights that inform space planning for each group. The downside to Wi-Fi-based occupancy is often a hit to fidelity: it may not be as accurate or immediate as dedicated sensors and will miss nuances like passive occupancy that an optical sensor might track.
There’s also the caveat that not everyone carries a Wi-Fi device or has it on, so you might miss visitors or edge cases. But Innerspace’s Wu is finding that, “businesses now want more nuanced behavior data that Wi-Fi can provide.” The good news is that these technologies are not mutually exclusive; some firms use a mix (e.g. optical in key areas and Wi-Fi analytics as a net across the whole building).
Your choice of sensor tech should map to the use cases and level of detail you need. “You need to think about what you might want to do. If you want to get into behaviors and you haven’t picked the right sensor, that doesn’t help,” advises Ryan.
In practice, this means engaging stakeholders upfront and learning from early adopters. If occupancy data will feed into real-time occupant-facing apps, favor sensors with instant communication. If your leadership cares about wellness and how active employees are, consider sensors that can measure something like sitting time or posture. If space planning for different departments is a goal, maybe a Wi-Fi approach to identify team patterns makes sense.
The same logic applies to IAQ sensors. These devices come in all flavors—some just measure CO₂ and temperature, others also measure VOCs, particulate matter (PM2.5), humidity, noise, light, etc. A few are “all-in-one” IEQ nodes that even include occupancy or people-counting.
First ask: what decisions or actions will this enable? Redmond noted that some R-Zero customers find more value in “picking apart the parameters and really applying each for specific use cases” rather than chasing a generic IAQ index. Conversely, if your main concern is occupant comfort/health, you’ll want a broader spectrum of IAQ metrics (CO₂ for freshness, PM for pollution, humidity and temp for thermal comfort, maybe even noise for acoustics).
“On the integration side of things, sensor selection matters. If you go with [sensors from proprietary BAS systems] and you switch control systems, you might lose the ability to pull those readings into a different BAS,” cautions Kyle Megna of Airthings. In other words, an IAQ device that only talks to one type of supervisory controller (let alone cloud environment) might leave your data in a silo (we’ll get to silos next) unless you plan ahead.
IoT sensors are not yet one-size-fits-all commodities. A thoughtful selection upfront prevents regret later. As Ryan sums up, choosing correctly means unlocking the use cases you care about; choosing poorly could mean hitting a dead end. We’re even seeing vendors partner up to cover each other’s gaps (XY Sense, for example, partners with Airthings and Kaiterra to overlay IAQ data on occupancy maps; Butlr and InnerSpace partner to combine thermal and wifi-based occupancy analytics). The message: match the tool to the job.
[join Nexus Pro to keep reading the final 2 trends, plus unlock all past articles and event recordings!]
Head over to Nexus Connect and see what’s new in the community. Don’t forget to check out the latest member-only events.
Go to Nexus ConnectJoin Nexus Pro and get full access including invite-only member gatherings, access to the community chatroom Nexus Connect, networking opportunities, and deep dive essays.
Sign Up
Nexus Labs exists to move the smart buildings industry forward, without the fluff.
On Wednesday, we send out the renowned (and free) Nexus Newsletter.