Scientists and people in sensitive groups have long understood the health risks of pollution and particulates, as well as the effects of poorly ventilated spaces on cognition, but up until this year the subject of air quality and circulation has not be high on the list of concerns for the general public.

Now, as we are all cooped up trying to make life work in small spaces thanks to both Covid19 and a tragedy unfolding with wildfires in the western US, air quality has reached a level of awareness that matches its importance. It's time then for indoor air quality monitoring to move from a relatively niche product category into the mainstream.

Enter HomePod

I think Apple's HomePod is a perfect vehicle for this new capability. Launched in early 2018 as a high-end product in the smart speaker category, HomePod has seen some success, but not to the level Apple anticipated as evidenced by lukewarm interest and retail discounting. This lack of success, I believe, is due to Apple still figuring out what market HomePod should be in, specifically what bundle of features make for a compelling value proposition.

HomePod's existing value proposition of luxury smart speaker falls squarely in the nice to have category, sandwiched between more flexible, dedicated AV systems at the prosumer end of the market, and mostly indifference on the casual end. It even suffers from a strategy tax albatross of Apple's services division with Music, preventing direct use of other music apps like Spotify.

While a new, lower cost version is rumored to be in the product pipeline (and may be announced in just a few hours), unless this device does something different, I think HomePod is still a product in search of the right positioning.

Product Positioning: From Luxury to Home Health

Much like Apple Watch was initially launched with a heavy emphasis on luxury (remember all the hype around the gold Edition model?) before being repositioned as a health product, HomePod needs to shift its emphasis from luxury smart speaker toward home health. There's at least four reasons why this makes sense for Apple in particular to pursue:

1. Ability to be the best

Apple's strategy is to enter markets where they can be #1 or #2 and where they can provide a unique user experience advantage. This advantage is distilling useful technologies to their essence and making them just work, so that customers have one less thing in their lives to worry about.

Current offerings in the market tend to miss the mark in a few ways, often with inaccurate or untrustworthy sensors, superfluous features and physical interfaces that drive up cost, questionable long term software support, and designs that do not provide a straightforward or trustworthy summation of what's happening and what action a user should take in response (e.g. do I trust a random manufacturer's concocted 'air quality score'?). Many don't even monitor the full range of air quality metrics worth monitoring. None of these in my mind are Category Defining Products, because the ability to monitor air quality on its own is fundamentally a feature not a product.

Meanwhile, Apple has repeatedly demonstrated the ability to innovate on sensor technologies, providing a level of trust and consistency in a wide range of sub-optimal operating conditions that most other products cannot achieve. Coupled with their experience making complex technologies understandable to end users, Apple is ideally suited to build great indoor air quality measurement products that make meaningful improvement to the health of millions, and build awareness of these important metrics beyond the home.

2. Better alignment with brand and initiatives

Apple is taking great effort to make their brand synonymous with health and wellbeing. As a general rule, people are quite willing to spend money for any insight into and reassurance around their health. Apple Watch has been a huge success because it tapped into this basic human desire, combining well-calibrated sensors with a fantastic and straightforward Health app. It makes sense then for Apple to pursue every angle on health that its customers understand and value.

This is even more true if it concerns the health of loved ones like aging relatives, kids, or pets - things that a home health product can deliver even more for than Apple Watch. These concerns simply occupy a different and more important space in our minds than the nice to have value proposition of products like HomePod. And this psychological need will continue to grow as we enter deeply uncertain times and become more aware about how our health is affected by a worsening environment.

Meanwhile, Apple also has home-related strategic interests with HomeKit and Siri Shortcuts, which ideally unlock the promise of smart home concepts. With several new metrics around indoor air quality, customers would be one step closer to this promise because third-party home automation products could tap into automations around air purification and temperature / humidity control.

3. A bundle that makes more sense to customers

HomePod to me doesn't feel complete right now. I own two of them, and they're imperfect speakers with an imperfect assistant, and that's it. There's no real reason to buy multiple HomePods unless you really, really want music or Siri in every room.

Meanwhile, you only really want to know about air quality if it's bad; you're not going to constantly check an air quality monitor's screen or proactively engage with it, and more electronics generally mean more cables, hassle, clutter, etc.

Separately, these products are niche. Together, these can have a multiplicative effect on HomePod's value proposition, because for any room you want music or air quality monitoring a HomePod would be a good choice. One customer's desire for reassurance about home health would be the foot in the door for engaging with the Siri ecosystem and discovering the value of speakers everywhere. Similarly, another customer's desire for a smart speaker would also deliver awareness of home health, which may then also spur interest in personal health and Apple Watch, by way of familiarity with the Health app. In either case, the exposure to these secondary value propositions would build customer interest in something they otherwise might not have thought they wanted.

Finally, whether you're in a tiny apartment or huge house, indoor air quality matters just the same. Because indoor air quality varies a lot based on location, customers in larger homes would be inclined to purchase multiple HomePods to ensure good monitoring coverage - something only a true Apple die-hard would do with today's HomePod.

4. Hardware that makes sense together

Another, more straightforward reason for bundling these features is that the physical placement of a smart speaker and an air quality monitor is likely to be similar, as both should be in relatively unobstructed parts of a room. Bundling them into a single product means one less piece of technology cluttering a space, one less company to understand and trust, one less interface to learn, etc. And neither air quality monitoring nor smart speakers need visual user interfaces on device, since they can hand that job off to other devices like iPhone.

The hardware itself is also quite complementary. By virtue of how speaker drivers work, they must move air to make sound. A HomePod's driver can move up to 20mm, which is a lot of air that gets pushed and pulled through the system. A clever design would place sensors within a HomePod's driver to cause air to pass over them, likely improving accuracy with more exposure to air (and thus more data points) as well as calibration by removing any contaminants which may settle on sensors over time. We've seen this kind of dual use of speakers before on Apple Watch, which uses a speaker to eject water after being immersed (video, video).

Implementation

Building indoor air quality monitoring into HomePod is a classic hardware / software / psychology design challenge. In addition to selecting the right sensors that are reliable and sensitive enough and then smartly packaging them, Apple would need to figure out the right way to represent air quality to users to help them best understand their risk and optimize their environment. Below are some specifics on ways Apple could do that.

Hardware

To meet the core use case of peace of mind about indoor air, HomePod would need to measure a broad range of meaningful air quality measures:

CO2 - poorly ventilated spaces can affect cognition rather quickly. At even moderate concentrations of CO2 cognitive capacity is greatly diminished, resulting in lower test scores and worse performance. These issues affect any home, school, of office environment that does not ventilate properly.

Particulates - even short term exposures to particulates can exacerbate cardiopulmonary issues and trigger asthma attacks and other sensitivities, while long term exposure is linked to premature death and many other ailments. Of growing concern are particulates and exhaust from roadways, which in addition are quite carcinogenic. The smaller the particles (PM2.5 and below), the deeper they can get into our lungs and the worse the effects.

Volatile Organic Compounds (VOCs) - paints, carpets, building materials, and other household items off-gas VOCs, as do cleaning products. Indoor air tends to have much higher concentration of VOCs than air outdoors, so the problem is highly localized and not subject to general awareness the way particulates and CO2 are. In addition to just being irritating, VOCs affect the nervous system and organs.

Temp & Humidity - these have a complex relationship with each other and on quality of life, especially with certain conditions like allergies and lung function. Humidity in particular has an optimal range of around 40-60% where mold, dust mites, viruses, and bacteria are kept at bay.

Notably absent from this list is carbon monoxide, which though important is already a required and installed product in most homes and apartments, has existing consumer awareness, and uses sensors that degrade over time. Still, an argument could be made to include one if Apple were interested in taking on that additional regulatory hurdle.

Within each sensor category, there are different types of sensors (laser/IR photodiode, spectroscopy, chemical, ionization from radioactive decay, etc). Some can simply tell concentrations of particles, while others are more sensitive and can distinguish which specific particles are problematic. While I'm not qualified to have an opinion on what particular sensor to select, there should be broad coverage of the variety of things that affect indoor air quality and that can drive users to make healthy choices. It also wouldn't be out of character for Apple to use their sensor talent and operational scale to go above and beyond in one particular facet of measurement to really make the case for the product being the best consumer air quality monitor.

User Experience

Making health measures understandable and actionable to laypeople is tricky business, requiring designers to find the right path across the disciplines of research, public health, teaching, behavior change, and engineering. The process requires starting with the health effects of different measures on users, and working backwards to derive the actions that are best taken to improve those effects and the awareness that is worthwhile to deliver to users. And on the engineering side, it requires finding smart ways to implement how these metrics are captured and derived in real world conditions.

In indoor air quality monitoring, Apple really has the opportunity to deliver a significant user experience advantage over existing products in the market. They can do this by distilling air quality measures into understandable and actionable metrics that align with the science on long term exposure and that leverage Apple's ecosystem to tie health effects to rooms and users rather than simply sensors.

But first, lets explore why implementations today miss the mark.

The Problem with Metrics Today

One issue with existing indoor air quality monitors is they're both device-centric rather than user-centric, and tend to focus on present conditions rather than exposure over time. This limits their use cases primarily to awareness around temporary spikes in poor air quality. While instantaneous spikes are important to monitor, the health effects from even moderately poor air quality accrues over months and years. This represents a major blind spot in user awareness and health risk which is unaddressed today.

Additionally, many products on the market try to simplify various sensor data into a single air quality score with an optional drill-down into each sensor's reading. I'm not really a fan of this approach for several reasons:

1. A single score is too simplistic to be action-oriented, something which every health product should strive for.
2. A single score is not descriptive enough to indicate what specific effects the user may experience as a result of prolonged exposure.
3. The particular way the score is achieved is a black box to users, creating a trust gap and necessitating further investigation.

A single score approach is an artifact of air quality monitors being special-purpose tools that are simply doing their best to justify themselves to consumers through physical user interfaces and the mystique of a 'special sauce', but lack a ecosystem to leverage for better integration into the daily lives of users.

An Alternative: Safe, Fresh, and Comfortable

Here's a way I think air quality can be better distilled into understandable and actionable measures that take into account exposure over time:

• SAFE measures particulates (PM2.5, PM10), VOCs, and carbon monoxide, all of which have related health effects from both acute and long term exposure. People with chemical sensitivities and allergies would have particular interest and a low threshold for this measure. Interventions would focus on air purification and a reduction in chemical use.
• FRESH measures carbon dioxide levels, which have cognitive and psychological effects. For instance, while a room may have well-filtered air, it may still be stuffy and cause people to feel more anxious, have trouble thinking and generally feel off. Interventions would focus on opening doors and windows, and getting outside.
• COMFORTABLE measures humidity and temperature. Since these vary considerably around the world especially in areas without heating, humidity control, or air conditioning, user's agency can be rather low around this metric. Users may also acclimate to these differences in a way they cannot to the above, which is why creating a separate metric makes sense. For users with HVAC and humidity control, interventions may be to alter these at certain times of the day.

Rather than just show on a screen on a device, each of these three measures would be summarized and charted minute by minute and in daily summary on a per-user basis in the Health app, and on a per-home and per-room basis in the Home app. In addition to these daily summarized ratings, any temporary spikes in a specific sensor's reading (like PM2.5) would trigger alerts for immediate action.

Centering Measures on Users Instead of Rooms

Unlike heart rate or other health measures from products like Apple Watch, air quality is both a shared health metric and one that is measured from a fixed location. This creates additional challenges to determine exposure over time on a per-user basis and to provide relevant recommendations. The solution requires several different needs and scenarios:

Home size - homes may range in size from single room studios to gigantic mansions. To use HVAC parlance, this can be approximated to different zones or common air spaces. For smaller homes, a single HomePod would be sufficient to approximate a user's exposure over time to indoor air, but for larger ones, multiple HomePods would need to be located across zones to ensure proper coverage.

User location - to measure per-user exposure, the system would need to understand approximately where users were in the home. This is where Apple's ecosystem can come to the rescue. By measuring the bluetooth & ultrawide proximity and wifi strength of personal devices to the HomePod, it becomes possible to infer roughly what zone each user is in and thus what air they are exposed to. If users are not carrying or interacting with their devices, data recording can fall back to behavior- or home-based approximations.

Outdoors - for the times users are outdoors, Apple could partner with localized external data sources like PurpleAir or fall back to government sources like AirNow to approximate exposure levels.

Children, pets, and others without devices - for anybody (or anything) that is not a part of the Apple ecosystem, measuring per-user exposure in the Health app is not possible. This is where the system would need to fall back to providing awareness around the next broadest measure, which would be whole home air quality. Therefore, indoor air quality data would need to also appear in the Home app. Perhaps in the future, the Health app will include health records for anyone the user is a caretaker of, such as children and the elderly.

Different indoors environments - since air quality is a shared health metric, these data need not be limited to the owners of HomePods. Instead, anybody with a personal device in proximity to a HomePod could receive its broadcasted air measurements which would be automatically added to their personal Health app record. This also opens up the opportunity for Apple to create a standard for air quality data sharing, so that offices and other indoors environments could have sensors that broadcast their data to nearby devices.

Additional Considerations

Another unique challenge in making air quality measurement useful and actionable is how users are made aware of air quality issues and what recommendations for action they are presented with. Since users are less likely to have agency over their indoor air environment than they are other typical health or fitness measures, awareness and recommendations must be scoped to a broader decision environment rather than merely sensor readings. Additionally, several air-specific use cases must be considered. Again this requires considering several different needs and scenarios:

Daily & long term exposure - while extremely important to track, this concept needs a simple science-based recommendation along the lines of 'stand 12 hours' or '10000 steps' to be able to ground recommendations and user psychology in. Otherwise users are left to wonder what to feel about a wall of numeric or color-coded data, when reassurance is the goal.

Interventions available to users - most users will not have access to a full range of air quality interventions, such as HVAC, air purifiers, fans, good quality outdoors air, or other methods of ventilation. This will affect what recommendations the system can send.

Moderately bad environments - users may be stuck in moderately bad environments such as the office where they have no control over the air. If these environments go over their thresholds for being notified, they may want to baseline to a new norm just for that location (e.g. PM2.5 at 90), which would mute notifications. However they might also want to know if it that location gets a certain amount worse.

At home vs away from home - users may not want air quality notifications when they are not at home; on the other hand, if they have pets they may still want to know.

Returning home automations - users may wish to have automations run prior to returning home, such as running an air purifier.

Data gaps - users may have large gaps in air quality monitoring coverage in their home, so the system must take this into consideration.

Changes over time - recommendations may change over the course of the day based on indoor and outdoor air quality. Recommendations would need to strike the right balance of helpful vs bothersome.

No good option scenarios and emergencies - sometimes the air is just bad and there are no good solutions or actions to take to make it better. People on the west coast right now understand this well - air quality outside is terrible, so people shut their windows, but this only makes CO2 and VOCs inside rise, which is also bad. When stuck between a rock and a hard place, the last thing users want is are braindead notifications telling them what they already know.

User initiated air quality drops - users may take actions such as vacuuming, cleaning, or cooking that cause predictable drops in air quality. Users may not be interested in getting bothered by notifications about these.

Portability - users may wish to have some ability to measure places that are either out of reach of power cables or where they will be temporarily. Many devices on the market are portable and battery-powered for this reason.

Awake vs asleep - users may have differences in optimal comfort for temperature and humidity when asleep vs awake. This directly intersects with more traditional house thermostat use cases and products.

Solutions

The above were a laundry list of possible things to consider about indoor air quality, however the solution offered need not be that complex, and for a first version may choose a more chatty, simpler notification logic. I do think there are a few opportunities for adding an Apple touch to things though:

Sensitivity adjustment - different people have different levels of sensitivity to things in their environment. Therefore it would be important to include as an option the ability for users to adjust the thresholds for when they are notified about the environment they're in. Users with chemical sensitivities or allergies would likely want more aggressive thresholds for notifications than others. Adjustments should be possible both generally (a slider) as well as per-sensor. For best UX, this may be presented as questions around what sensitivities the user has (e.g. smoke is bothersome), which would set thresholds for one or more sensors that corresponded to.

Notifications - deliver various notifications such as when moving into a space with unsafe air, when instantaneous thresholds are met, and when exposure over time thresholds are met.

Daily and weekly summaries - provide these in the Health and Home apps along with more in-depth information on the effects of various air quality issues.

Emergency mitigations - provide users with recommendations for things they can do when in emergency situations, like put out water trays to capture smoke particles.

Exposure notifications - similar to Apple Watch standing notifications, a user's device could recommend changing locations if a daily exposure threshold was met.

Cooking & vacuum detection - one benefit of bundling air quality sensors into a HomePod is leveraging the HomePod's array of microphones. A vacuum cleaner's telltale sound would be easy for a HomePod to detect, which could mute any temporary spike notifications, and it potentially could for cooking as well. A more sophisticated version of this could remember the last time you vacuumed, and if air quality in that room slowly worsens, and it's been a long time since a vacuum was last heard, the HomePod could recommend vacuuming.

Space mapping - a more out there idea is that users could map out their home using the new LIDAR sensors offered in iPad Pro and iPhone 12+, creating a digital representation of their space. Combined with air quality data, this could generate understanding of airflows throughout house, and recommendations for specific actions (e.g. close the bedroom door).

Product Lineup

Here's my take on a new HomePod lineup that would include home health measures and provide customers with several options to build peace of mind about indoor air quality in their homes:

HomePod Pro - larger smart speaker with all air quality measures, room sound shaping, and fuller sound. Pretty much the current HomePod ($399-$499).

HomePod - moderately sized smart speaker with all air quality measures, but not as high quality sound ($199-249). This would be the mainstream one, or the one you get for secondary rooms.

HomePod Mini - portable USB-C speaker with more limited air quality metrics of temp, humidity, and particulate detection to meet the use case of carrying a sensor around to different places ($149-199).

Summary

Adding indoor air quality monitoring is a perfect way to re-position HomePod toward home health, leveraging Apple's ecosystem and work in health to help the product find its market.

Although an argument could be made that you only need periodic, localized air monitoring to determine how to set up a healthy environment, the benefit of continuous and broad coverage of indoor air monitoring can have both direct and indirect positive benefit on our health. This is because each additional health metric gathered paints a more complete picture of our environment and state of being. This picture is important for us to gain greater insight into why we feel the way we do, and what we can do to optimize their health and live our best lives.

So often in physical and mental health the things that ail us are difficult to piece together, and interventions wind up being a waste of time. But, if we cover the basics like vitals, exercise, air, water, and food, we can go a long way toward delivering more precise interventions to help us live the healthiest lives we can. Right now is a really exciting time in digital health as new innovations are doing just that, and helping us build greater awareness of both our own health and that of our environment.

Thanks for reading!

This post was written during an especially tragic time on the west coast, where millions of acres of forest are burning, the health and wellbeing of everyone is threatened, and cities and towns are experiencing some of the most dire air conditions in their histories. Though this post is mostly inside baseball about product positioning and features, I want to be sensitive to the immediate conditions we're all facing, as well as recognize that these environmental catastrophes are likely to grow as time goes on and climate change worsens.

As a product creator, the tools available to me are primarily within the scope of products and their effect on human psychology and understanding, which is just a small part of a much larger set of solutions that address the threats to future generations and life on earth. We're all in this together and we all need to do our part here. Participation is not optional.

In regard to this product concept, it's my belief that building a well-positioned and successful product that builds greater awareness around air quality will have a positive impact on discussions about climate change. We should be aiming to give a large number of people knowledge of how our environment works and the threats of greater carbon dioxide concentrations on our environment. Building a product that takes that abstract concept and personalizes it, in my mind, is quite a good way to do so.