What is a smart building?

Jake Tauscher
G2 Insights
Published in
8 min readFeb 3, 2021

You have probably heard chatter recently about “smart buildings”, and for good reason.

First, commercial buildings are a major source of emissions, typically accounting for ~16% of US emissions (although likely somewhat less than that this year, with offices empty). And, second, commercial buildings are hugely wasteful — the EPA estimates up to 30% of building energy use is wasted. But wait — third, there is hope. So-called “smart building” technologies have demonstrated energy savings of 25%+ in typical buildings, seemingly eliminating much of that waste.

So, problem solved? Well, not quite. These smart building technologies are not deployed in many buildings today, with estimates of penetration at under 10% of buildings (and potentially as low as 1%). So, what is going on here? Why such a gap between potential and realization?

Well, to understand the promise and challenges of smart buildings, we must first understand how a commercial building is managed today. And, to understand that, we should start with understanding the “Building Automation System”.

What is a Building Automation System?

The Building Automation System, or BAS, is at the core of building management. Most buildings over ~50K square feet will have a building automation system, and while this is only ~5% of commercial buildings, this is ~50% of commercial floor space.

A BAS is a control software that integrates with key building systems (usually HVAC, but also lighting, elevators, security, fire) to monitor and control them. The set-up is pretty basic. Consider the graphic below, which is a good overview of a typical BAS set-up. There is a computer (at the top) that is talking to sensors and controllers throughout the building (at the bottom), receiving input and telling them what to do.

h/t to Nexus for this graphic. As a side note, if you are interested at all in smart buildings, go subscribe to Nexus’s newsletter. They have a ton of great info on this space.

So, how does the BAS know what to tell the building to do? Well, it uses what is referred to in the industry as the “3 S’s”: Setpoints, Sequences, and Schedules. So, essentially, when the BAS is first set-up, the engineer will program in certain settings. These could be things like “turn off the lights at 6:00 PM” or “turn on the heat when the building temperature drops below 65⁰” — basic “if/then” style commands. And, that is basically it. The BAS can run for years without anyone really interacting with it.

If and when there are issues with the building (and there are often issues — industry experts estimate that BAS performance slips by 5–10% per year), then the installer of the BAS is called-in, diagnoses the problem, and updates the 3S’s to account for it. This is a process known as commissioning.

So, that is how most buildings work today! These BAS’s can be pretty expensive up front (estimated at $2.50–7.00 / square foot in hardware and installation costs) and usually require service contracts on the back end (to address issues when they inevitably arise). However, between those issues, there is very little daily variance in how a building is operating!

Emerging Technology

So, now that we have a good primer on how buildings traditionally operate, we can ask — what sort of “new tech” might make a building a smart building?

Well, (spoilers) there is no one definition of a smart building, but there have been a few key pieces of tech that have been introduced into the building industry over the last 15 years.

Energy Information Systems (EIS): The first is real-time energy information. Until maybe 2005, most building operators only learned their energy use when they received the bill in the mail a month later. So, some companies started integrating with the energy meters of the buildings, and then offered this data to building operators. Now, building operators could not only see their energy use in real time, but they could also easily compare it to other similar buildings to see if their building was as energy efficient as peers.

Fault Detection & Diagnostics (FDD): Then, around 2015, fault detection and diagnostics (FDD) came onto the scene. To proactively identify issues in the building, companies would pull and analyze data from the building BAS (or even direct from systems, like HVAC). With this data, they could identify simple things like “this thermostat has stopped talking to the BAS — it must be broken” to slightly more complicated things like “this cooling unit seems to be using too much energy — it could be leaking”. Rather than waiting for tenants to complain, FDD software can find issues proactively and notify the building engineer to solve them before they waste significant energy or affect tenants.

Advanced Supervisory Control (ASC): Today, we are entering into a world of advanced supervisory control. These systems will integrate with (or supersede) a BAS and provide not only fault detection, but also more dynamic operation of the building. So, you could imagine a system that detects (from an occupancy sensor) that the 3rd floor is empty. Rather than wait until 6:00 PM to turn down the AC, as its schedule dictates, it turns down the AC right now. This process is referred to as continuous commissioning.

So, this is all pretty cool stuff! And, fairly cutting edge — advanced supervisory control tech can use heavy duty AI/ML to optimize building operations. However, we must remember that this tech, particularly advanced supervisory control, is not present in most buildings today. So, what holds these technologies back?

Challenges for Smart Buildings

There is no one answer to why this market has been tough to penetrate — if there were, I think the issue would likely already be solved! However, there are a few barriers that I find compelling.

1) Difficult Access to Data: In today’s open-source world, it is sometimes easy to forget that you cannot always integrate directly into any system. Many BAS’s are “closed” systems, in that they will not provide access to their data without a fee (and if you try to circumvent that fee, they will void the building’s warranty). Then, once you have access to the data, you have to understand what the data means (e.g. this data is coming from a thermometer, this piece of equipment is an air conditioner, etc.). Each BAS may use a different system for organizing its data (there are 80+ languages for building data), so this process can involve significant manual effort.

2) Misaligned Incentives: Tenants pay electricity bills, but tenants do not pay for these projects or technologies — the building owner does. So, for a building owner, how do they underwrite this investment? They sometimes are able to share the benefit of energy savings with a tenant, or raise the rent when a tenant leaves — but it is not straightforward, and may depend on how the lease is written.

3) Risk Aversion: Simply put, buildings last for 30+ years. Many building owners are nervous to partner with a company that is only a few years old, and many of the companies offering these smart building technologies are start-ups.

Tailwinds for Smart Buildings

However, there are certain tailwinds that may mean the commercial building industry is ready for a major shift toward newer technologies. These include:

1) Standardized Data Ontologies: I just talked about how data access is an issue in this market. Well, many major industry / non-industry players have recognized this issue. So, there have been efforts to introduce new, standardized, open-source data structures.

This actually started back in the 90s with BACNet (designed to allow the BAS to talk to different systems, e.g. elevators and fire) and the Niagara framework (designed to allow applications on top of the BAS). These were semi-open systems, in that they helped some distinct systems work together, but they are not fully open systems. They also can be fairly pricey to implement.

More recently, there have been efforts to create open-source standards. In 2014, Project Haystack was launched by a group of industry participants to create standardized naming conventions around building data. Then, in 2016, Brick was launched to correct some failings with Project Haystack. Most recently, in 2020, Google launched Google Digital Buildings, designed to be compatible with both Project Haystack and Brick, while Microsoft launched its “Real Estate Core”. These are all efforts to create more standards around how buildings organize their data and metadata, but as is evident, the industry has not yet coalesced around a standard.

2) Regulation: Many cities have recently announced regulations on buildings that should motivate building operators to more seriously explore smart building technologies. For example, in 2019, NYC passed Local Law 97, which sets standards for “emissions / square foot” in buildings, with steep fines for non-compliance. This law will require 75% of buildings in NYC to take steps to reduce their emissions by 2030.

3) New Data Sources: If a building is operating consistently, it could be assumed that there are diminishing marginal returns to “dynamic” building control. For example — if I can reasonably predict that most people will come in at 8 AM, and leave at 5 PM, why do I really need to adjust my building in real time?

Well, new external data sources, such as hyper-local weather APIs, allow dynamic building technologies to generate more value. Using this external data, for example, the system could anticipate periods of changing demand in the HVAC system, and optimize the building accordingly. This sort of external data increases the value of smart building systems.

So, how are start-ups approaching this market?

There are a few different approaches start-ups are taking to try to drive value in the commercial building market.

The first is a “Rip and Replace” approach. These start-ups would make the following argument: “The BAS market is broken — we need to start from scratch. Let’s put in modern hardware, modern software, and do this thing right”. This approach actually highlights another tailwind for the smart building market (which could end up being the most important one): the increasing ability of “off-the-shelf”, commoditized hardware to offer high capability compute and control.

The second is a “Build on Top” of what is already there approach. These start-ups will draw data from the existing BAS, and then use this data to either push suggestions to the building engineer, or to automatically push commands back down into the BAS.

The third approach I will characterize as “Lite BAS for SMBs” (there is some overlap between categories 1 and 3 here). You will remember from earlier in this blog, that most buildings smaller than 50K square feet do not have a BAS today. Historically, they have been just too expensive. However, with sensor / computing hardware getting cheaper, some start-ups are trying to penetrate the SMB market with a “light hardware + software” BAS approach.

In summary, though, the players in this market still have a lot of work to do to deliver on the promise of the smart building. But, there is just so much opportunity for the companies that crack this code. So, it is definitely a space worth watching over the next few years. And, if you enjoyed learning a bit about the BAS system here, and want to go deeper, I will reiterate my recommendation to check out Nexus — it is a great source of insight on the smart building space!

Sign up to discover human stories that deepen your understanding of the world.

Free

Distraction-free reading. No ads.

Organize your knowledge with lists and highlights.

Tell your story. Find your audience.

Membership

Read member-only stories

Support writers you read most

Earn money for your writing

Listen to audio narrations

Read offline with the Medium app

Published in G2 Insights

G2 Venture Partners is a venture and growth capital firm investing in transformative technology companies at their inflection points to build a sustainable future. | www.g2venturepartners.com

Responses (1)

What are your thoughts?