Article
15
min read
Brad Bonavida

Case Study: KIPP DC Schools wants Autonomy, Deploys IDL, Supervisory Control, FDD, and CMMS

March 26, 2024

Welcome to our Case Study series, where we dive into case studies of real-life, large-scale deployments of smart building technologies, supported by the Nexus Marketplace.

I emphasize “real life” because this isn’t a marketing fluff story. We're here to share real lessons from leaders who have done the work to integrate smart building technology into their operations. I also emphasize “large scale” because we're not here to talk about pilot projects. We're here to talk about deeper commitments to changing how buildings are operated.

---

Case Study Data:

  • Technology Categories Mentioned:
    • Smart Buildings Service Provider
      • Master Systems Integrator
    • Application Layer
      • Advanced Supervisory Control
      • Fault Detection & Diagnostics
      • Asset Management
    • Data Layer
      • Independent Data Layer
    • Device Layer
      • HVAC
      • Metering
      • Indoor Air Quality
  • Key Stakeholders: Nathan Morris - Director of Facilities, KIPP Public Charter Schools; Stephen Dawson-Haggerty, Normal Software, CEO; Terry Herr, Intellimation, President
  • Vendors: Intellimation - Master Systems Integrator; 3rd party facility management company; multiple mechanical and controls contractors
  • Number of Buildings: 8 campuses, 20 buildings, 1,000,000 sqft
  • Project Dates: 2020 to Present

Case Study Outline:

  • Introduction
  • Background
  • Technical Overview
  • Challenges & Lessons Learned
  • Conclusion

---

Introduction

KIPP Schools is a network of public charter schools throughout the United States that is open to all students. KIPP stands for Knowledge is Power Program, a program with a proven track record of academic excellence.

Four years ago, Nathan Morris joined the Washington, D.C. branch of KIPP as Director of Facilities. KIPP DC consists of eight campuses comprising twenty buildings, serving students from K through 12. With the United States’ student test scores recently suffering compared to their international peers, there has never been a more critical time to focus on making facilities as comfortable for learning as possible. Yet, like many other schools, KIPP DC struggled with facilities products and service providers that weren’t making things easy.

“I’ve yet to meet anyone who works for a school system who hasn’t dealt with a vendor who’s not meeting the bar, or is not paying for way more than they are actually getting.” —Nathan Morris

When Morris arrived at KIPP DC, they infamously had more building automation systems than campuses, creating more noise than value. Additionally, KIPP DC needed better data to quantitatively compare the condition of different buildings and systems for capital planning.

Luckily for KIPP DC, Morris had previous experience with a trusted master systems integrator, Intellimation, with whom KIPP DC was able to partner to create new value in a historically challenging facilities environment. Intellimation immediately began working to converge data using an independent data layer (IDL)

As a more solid foundation of information was built within the IDL, new doors of opportunity opened up for the KIPP DC. Data models led to data normalization across multiple building automation systems. Data outliers highlighted building optimization opportunities to lower energy usage. Fault detection and diagnostics (FDD) and computerized maintenance management software (CMMS) applications were deployed on top of the IDL to improve operational efficiency. KIPP DC gained vendor agnosticism, data to support the prioritization of capital projects, and faster and more trackable service work. 

We sat down with Nathan Morris of KIPP DC, Terry Herr of Intellimation, and Stephen Dawson-Haggerty of Normal Software (the IDL provider) to discuss how KIPP DC has taken the reins of their own facilities data and how all this behind-the-scenes effort inevitably leads to healthier environments for KIPP DC students and teachers to thrive.

Background

Nathan Morris has built a career around helping educate youth. He started as a teacher, ranging in various grades from kindergarten through middle school. He later transitioned to the facilities side of the business at DC Public Schools. He’s been the Director of Facilities at KIPP DC since 2020.

When KIPP DC started in the early 2000s, most of its schools were created out of new building construction. Hence, KIPP DC spent its earlier years more focused on growing than maintaining, as Morris explains, “Our buildings are what we like to call Frakenbuidlings, where we have continued to tack on and add square footage as we’ve needed to expand, which is nice in theory, but has created a real challenge.”

With this rapid growth, a clear facilities and operations plan naturally remained a back-burner topic compared to creating a state-of-the-art educational platform for their students. The lack of attention to facilities led to some less-than-ideal situations. “We actually have more building automation systems than we do campuses… one campus actually has two,” explained Morris.

KIPP DC has historically outsourced most of its facility operation jobs. It works with a third-party facilities company that employs full-time maintenance personnel for each building. KIPP DC also uses numerous mechanical contractors for preventative maintenance (PMs) and other service work.

Fast-forward through two decades of heat waves, cold spells, setpoint changes, and ambitious students. Suddenly, KIPP DC had systems that were no longer new and shiny but instead in need of serious service or replacement. As Morris came on board, KIPP DC needed to transition from development-focused to asset management-focused.

Unfortunately, KIPP DC wasn’t initially well-suited for an asset management focus. They had an arcane work order platform and tracked service work using clipboards, O&M manuals buried in basements, and “a bevy of nice spreadsheets.” 

The result was data that was more qualitative than quantitative, making it increasingly difficult for the school to determine what capital projects were most important and where to start. Too often, projects were triaged based on who yelled the loudest, which became a slippery slope to campus inequity.

Additionally, years of contracts and handshakes with different building technology vendors meant KIPP DC lacked autonomy over data and decision-making. For example, Morris mentioned how their work order platform is licensed to the third-party facility maintenance vendor they use. Therefore, they risked losing access to all work order data if they needed to change facility maintenance vendors.

Terry Herr is the president of Intellimation, the Pennsylvania-based master systems integrator (MSI) that Morris worked to convince his colleagues to trust and partner with to untangle the mess. Terry has spent over 30 years in the building automation industry. 2014 is about the time that Terry claims Intellimation was able to transition from a system integrator (SI) to a master systems integrator (MSI), shifting focus to data-driven retro-commissioning and optimization. 

When Intellimation came on board to support KIPP DC's journey to a complete digital facilities management solution, Herr quickly advocated getting an independent data layer (IDL) in place. Normal Software stood out to Herr and Morris as the best IDL option for KIPP DC for many reasons. First, there was the trust factor. Just as Morris had trust in Herr and Intellimation because of previous work they had done together, Intellimation had trust in Normal Software, an IDL platform they had been watching develop over the last couple of years. Additionally, Herr was well aware of KIPP DC’s data autonomy challenges and their aspirations of vendor agnosticism. Normal Software prides itself on being built on open-source technologies, giving the building owner the most control and autonomy over their data and systems.

Stephen Dawson-Haggerty, the founder of Normal Software, also joined us in the conversation. Dawson-Haggerty spent a portion of his earlier career developing the company Comfy, which initially “unlocked the thermostat” by allowing occupants to control the comfort of their room with the Comfy app. Through massive success, Comfy was acquired by Siemens Building Technologies and grew to be an occupant engagement app with numerous other features. 

After this journey, Dawson-Haggerty reset his ambitions and founded Normal Software. As Dawson-Haggerty puts it, “I felt there was a lack of great software actually doing amazing things. There’s a gap between the best practices people talk about and what you can actually buy and actually scale”.

Dawson-Haggerty emphasized his view of the IDL’s infancy in the market. Despite becoming well understood in the industry, there’s a difference between concept and actual success. Normal exists to take the IDL concept and determine the capabilities to make it work technically and as a sustainable business. 

With an IDL in place, new doors began to open up for KIPP DC in the form of user-specific applications that could be extensively evaluated for their performance and fit. KIPP DC slowly transitioned its facilities practices and capital planning into a data-driven mechanism - capable of determining what was most important and how to keep students and faculty happy.

Technical Overview

Phases of Development

Understanding what KIPP DC has done to its tech stack differs from understanding what it takes to actually get there. Morris, Herr, and Dawson-Haggerty walked us through the phases of deployment.

KIPP DC applied many of the concepts that Nexus Labs emphasizes through our Smart Buildings Buying Process Cheatsheet:

More info on the buying process cheat sheet HERE

Phase 1 - Defining the Objective

Morris recalled the project's origin, “The first phase was working backward to determine what we wanted to achieve. What sort of things did we want to get out of it, and what was going to help us run our campuses and network more efficiently from a facility perspective?”

It’s often overlooked, but Morris is describing the strategy developed before any action was taken. Many unsuccessful projects immediately jump to the right column in the image above: procurement. Before engaging vendors, Morris and his team at KIPP DC spent enough time in the strategy stage to understand what they actually wanted and needed. In the case of KIPP DC, the strategy centered around achieving data autonomy and obtaining better insights into maintenance and capital project opportunities. 

Regarding preliminary market research on what categories exist in the market, the KIPP DC team outsourced much of the effort by bringing on Intellimation to help them determine which products would best help them reach their goals.

Phase 2 - Harvesting the Data & Centralizing the BAS

Herr described how Intellimation started working once they were brought into the project: “It all starts with trending the data.” According to Herr, most BAS trend only about 10% of the available data. Intellimation immediately deployed the Normal Software, which acted as a central hub to gather as much trend data as possible.

It always starts with an independent data layer, or a middleware layer, to bring all the data together, trend all the data in a neutral, non-BAS fashion. And it always includes a central and typical BAS front end, but a single one instead of five or six different ones, which is what KIPP DC had.”
Terry Herr

The Normal Software IDL stored the trending data from multiple systems in KIPP DC’s Amazon Web Services (AWS) account. This meant that KIPP DC, and no one else, owned the data gathered. If KIPP DC were to part ways with Normal Software or Intellimation for any reason, they would still have full access and control of this data.

In parallel, Intellimation worked to slay KIPP DC’s eight-headed BAS monster. Intellimation helped KIPP DC evaluate three different BAS options as a single front end to their building automation.

Referencing the buying process cheat sheet above, Intellimation was shortlisting BAS providers based on KIPP DC’s goals and use cases. Check out the Nexus Buyer’s Guide to HVAC Control for more support on shortlisting BAS providers.

Herr described how the BAS vendor evaluation mutually benefits the MSI and the building owner. Intellimation always strives to stay informed on which products and features lead the market. Every chance Intellimation gets to evaluate vendors is a chance for them to rethink their preferred partners for future projects. This continuous vendor evaluation is the same reason Intellimation was able to bring Normal Software into the project quickly and confidently.

Once a single BAS was selected, Intellimation began the tedious task of developing all graphics into that single front end of building automation. A universal graphical representation of systems, regardless of the campus where the system is located, immediately improved the operational efficiency of the facilities team.

Phase 3 - Data Models and Optimization

Not all data is created equal. After centralizing trending data into Normal Software, Intellimation helped KIPP DC develop data models that provided a visual perspective of what was occurring in the buildings. The team utilized Grafana, an open source platform, to start creating logical stories around data gathered. 

Intellimation’s approach typically focuses on building optimization once the IDL is in place. Starting with optimization is often the easiest way to show some quantifiable results around the system overhaul. Morris pointed out, “Optimization is how we get our organization on board. With a combination of cost savings from energy reductions to building energy performance requirements in DC, like BEPS, we can start to show results.”

Building Energy Performance Standards (BEPS) are regulations or policies that set specific energy efficiency targets or performance criteria for existing buildings to meet over time. These standards are part of broader efforts to reduce energy consumption, lower carbon emissions, and improve the overall energy performance of the built environment.

A stakeholder outside of the building technology industry, like a CFO or board of directors, may not understand concepts like the IDL and can lose patience in these types of capital investments if results can’t be shown early. The data model and optimization of existing building sequences can be a time-sensitive outcome for facility managers trying to get buy-in on their approach.

Phase 4 - The “Bake-off” at the Application Layer
“With an IDL, you’re going to unlock a lot more deployment opportunities because it’s easier to try out different FDD products. Instead of doing a crazy six-month integration after a brutal RFP, you can just kind of do a bake-off between vendors, and the cost should be low enough.
—Stephen Dawson-Haggerty

As Dawson-Haggerty explained, one of the most valuable attributes of getting a reliable IDL in place is how it simplifies and reduces the cost of evaluating solutions at the application layer. 

As a reminder, the application layer sits above the data layer. It is where organizations can apply use-case-specific views and tools on top of the communal data so that each stakeholder can get what they need.

When a robust data layer is in place, vendors at the application layer have minimal effort to pull the necessary data into their software. Therefore, building owners can quickly test out application layer software with minimal effort and cost. This is the “bake off” concept the Dawson-Haggerty explained.

In the latest stage of this project, KIPP DC has been working closely with Intellimation to perform bake-offs for FDD and CMMS applications. 

As the KIPP DC team evaluates applications, they have gained an upper hand in contract negotiations. They have fewer worries about getting locked into license agreements for products or services because the data is owned by KIPP DC, in their own AWS platform, and accessed by the vendors. 

Normal Software tries to differentiate itself by supporting this concept of data autonomy, which is part of why they were such a good option for KIPP DC. Herr describes Normal Software as “truly open source” because it operates on the BACnet stack, programs in Javascript (a language college students learn in class, not a proprietary commercial development), and talks from the edge to the cloud in MQTT.

Additionally, Normal Software’s IDL is a non-SaaS offering. KIPP DC does not pay a recurring fee to Normal after the IDL is stood up, compared to competitors which require an annual licensing fee or hardware upgrades after old hardware gets phased out. However, Normal does compete at the application layer by offering unique applications on top of the IDL to meet a customer’s specific needs.

What about the IDL and MSI bake-off?

One thing that may stand out to readers of this case study is the emphasis on vendor comparisons to help you get the right product or service. However, KIPP DC quickly jumped into a partnership with Intellimation as an MSI and Normal Software as an IDL. What about vendor selection for the IDL and MSI?

There’s a fine line between the speed of deployment and results versus analysis paralysis in vendor selection. Spending too much time on vendor selection for the IDL or an MSI could have resulted in stakeholder impatience. Morris was fortunate to have previous experience with Herr and Intellimation to the point that there was already an established level of trust. Similarly, Herr and Intellimation had prior experience with Dawson-Haggerty and Normal Software to feel confident it would address the needs of KIPP DC.

Moreover, Dawson-Haggerty explained how the concept of an ROI can fall apart when it comes to an IDL. If an IDL finally gives a building owner visibility into their building data, allowing them to make better decisions, how can you assign a value to that? Dawson-Haggerty equated this to asking someone the ROI of their email system, ERP, or CMS; if something is fundamental to getting a job done, it can be challenging to assign a value to it.

The Nexus community has long debated when to apply IDLs and when to add application layer software. Not all agree that the procedure KIPP DC took is the best option. For a broader range of opinions, we recommend you listen to our SME Debate on the IDL.

Quantifying Success at the Application Layer

Despite the challenges with tagging a value to the IDL layer, KIPP DC is undoubtedly experiencing positive results with its new autonomous and converged data layer. KIPP DC continues improving its FDD, CMMS, and even access control system applications on top of the IDL.

Morris describes the number one benefit as the end-user experience. KIPP DC facilities finally have metrics to gauge how quickly they can resolve issues that are coming up. Morris equates this to the famous Dominos Pizza Tracker, but instead of responding to a pizza order, it’s responding to an uncomfortable occupant.

KIPP DC is finally confident in its ability to make data-driven decisions regarding capital investments. “There’s always too many projects and not enough capital to support them. Unfortunately, it’s been a very subjective process to date. A lot of times, the loudest voices win out,” said Morris. KIPP DC is proud to have improved fairness and equity and defend decisions with data. Morris brought up one specific example of kitchens with commercial cooking equipment in rooms that were not appropriately retrofitted to be kitchens. They had staff members experiencing temperatures in the high 80s. Precise data highlighted the poor conditions of these rooms compared to others and allowed KIPP DC to prioritize investing in mini-split systems to provide a better space to the occupants.

Lastly, KIPP DC’s third-party service providers benefit from richer and more accessible building data sources. All these partners are getting trained to use the new applications, like FDD and CMMS. Herr brought up the specific example of technical data on the compressors of VRF systems that were previously unavailable without a trip to the roof. Having this data accessible in the cloud allows service technicians to be significantly more efficient while servicing these VRFs.

Challenges & Lessons Learned

These case study articles aim to share success stories in a way that can benefit similar building owners by helping them make accelerated decisions toward improved tech. Morris was willing to share the largest obstacles to getting to where they are today, hoping other building owners can learn from their mistakes.

Lesson #1 - Start by determining what you actually need

The IDL unlocks visibility into all of your data, but it can be challenging to benefit from that data if you don’t know what you are looking for. Morris noted the importance of determining what data you need to make informed decisions before starting the project. Otherwise, you may spend too much time chasing ghosts within your data.

Morris and KIPP DC determined two overarching goals early in the process that helped with their decision-making:

  1. They needed data to help provide quantitative analysis of the state of their assets, which would allow them to fairly prioritize capital investment projects.
  2. They needed to be the sole owners of their data so they were no longer beholden to old contracts and agreements and could achieve the freedom associated with vendor/product agnosticism.

Lesson #2 - Do your research and find trustworthy partners

Morris had the humility to recognize that he doesn’t understand all the concepts of a building technology stack and probably doesn’t need to. He was able to trust Intellimation’s technical knowledge so that he could focus on all the other vital aspects of his role. Instead of trying to understand every intricacy, Morris prides himself on having a decent bulls*** detector, asking the right questions, and reading the fine print. 

Morris shared a horror story from his past of partnering with a vendor on a performance contract, only to later learn that the performance contract was based on the vendor’s projected energy savings, not actual energy savings. Additionally, most energy-saving measures were behavior-based, like wider setpoint ranges, which put his organization at risk of having to pay the vendor more than they were saving on utilities. Reading the fine print and relying on trusted partners is paramount to avoiding these situations.

Morris and KIPP DC also benefited from their close relationship with KIPP Texas, a larger school district that has experienced many of the same growing pains from a facility/operations standpoint.

Lesson #3 - Learning how to navigate the political landscape

Understanding all the complicated relationships between stakeholders is critical, especially in the municipal sector. Morris spends a significant amount of time understanding the goals of the school board and the local politicians to stay ahead of ever-changing compliance and regulation goalposts. 

Speaking to a project's bottom line is critical to winning support for a project, but it can be challenging. In the case of KIPP DC’s IDL deployment, they started with little data, so it was complicated to show hard numbers behind the benefit of the project. When that is the case, Morris recommends focusing on “how much time are we spending on this now, and how well are we doing it,” which, although qualitative, can help stakeholders wrap their minds around what improving facilities can do for the organization's overall goals.

Herr used indoor air quality as a specific example in this instance. With math testing scores at an all-time low across the US, anything that can be tied to a positive impact on student learning is of astronomical importance. Gathering IAQ data was a quick win for the project and attracted the attention of stakeholders.

Conclusion

KIPP DC set out in the early 2000s, determined to create the best learning environment possible. Each new campus became a new opportunity, and the mindset was focused on growing rather than maintaining. When tasked with shaping the minds of the next generation, there’s much more to focus on than the condition of a building. Soon, a tangled mess of building products, services, and vendors grew. But when Morris and the vendor team set out to right the ship, they succeeded by applying values no different than what was being taught to the youth in their classrooms:

Building Trust: KIPP DC put significant trust into Intellimation and Normal as their partners, but the trust was not blind. KIPP DC focused on finding partners who can help them achieve their unique goals and relied on the previous experiences of others to feel confident in their chosen relationships.

Making a plan and sticking to it: Too often, the method for deploying building technologies is “point and shoot.” KIPP DC could’ve jumped to an unlimited number of “solutions” to improve operational efficiency, but instead, they invested in getting a functioning IDL. Although the IDL can be difficult to attach an ROI-style value to, it acts as a platform for enhanced decision-making.

Going from siloed building systems to an IDL is like an entity that was once blind gaining site. With data now available to KIPP DC, what comes next is limitless. But it all starts with untangling the mess, one knot at a time, and building the correct foundation to grow upon.

Upgrade to Nexus Pro to continue reading

Upgrade

Upgrade to Nexus Pro to continue reading

Upgrade

Welcome to our Case Study series, where we dive into case studies of real-life, large-scale deployments of smart building technologies, supported by the Nexus Marketplace.

I emphasize “real life” because this isn’t a marketing fluff story. We're here to share real lessons from leaders who have done the work to integrate smart building technology into their operations. I also emphasize “large scale” because we're not here to talk about pilot projects. We're here to talk about deeper commitments to changing how buildings are operated.

---

Case Study Data:

  • Technology Categories Mentioned:
    • Smart Buildings Service Provider
      • Master Systems Integrator
    • Application Layer
      • Advanced Supervisory Control
      • Fault Detection & Diagnostics
      • Asset Management
    • Data Layer
      • Independent Data Layer
    • Device Layer
      • HVAC
      • Metering
      • Indoor Air Quality
  • Key Stakeholders: Nathan Morris - Director of Facilities, KIPP Public Charter Schools; Stephen Dawson-Haggerty, Normal Software, CEO; Terry Herr, Intellimation, President
  • Vendors: Intellimation - Master Systems Integrator; 3rd party facility management company; multiple mechanical and controls contractors
  • Number of Buildings: 8 campuses, 20 buildings, 1,000,000 sqft
  • Project Dates: 2020 to Present

Case Study Outline:

  • Introduction
  • Background
  • Technical Overview
  • Challenges & Lessons Learned
  • Conclusion

---

Introduction

KIPP Schools is a network of public charter schools throughout the United States that is open to all students. KIPP stands for Knowledge is Power Program, a program with a proven track record of academic excellence.

Four years ago, Nathan Morris joined the Washington, D.C. branch of KIPP as Director of Facilities. KIPP DC consists of eight campuses comprising twenty buildings, serving students from K through 12. With the United States’ student test scores recently suffering compared to their international peers, there has never been a more critical time to focus on making facilities as comfortable for learning as possible. Yet, like many other schools, KIPP DC struggled with facilities products and service providers that weren’t making things easy.

“I’ve yet to meet anyone who works for a school system who hasn’t dealt with a vendor who’s not meeting the bar, or is not paying for way more than they are actually getting.” —Nathan Morris

When Morris arrived at KIPP DC, they infamously had more building automation systems than campuses, creating more noise than value. Additionally, KIPP DC needed better data to quantitatively compare the condition of different buildings and systems for capital planning.

Luckily for KIPP DC, Morris had previous experience with a trusted master systems integrator, Intellimation, with whom KIPP DC was able to partner to create new value in a historically challenging facilities environment. Intellimation immediately began working to converge data using an independent data layer (IDL)

As a more solid foundation of information was built within the IDL, new doors of opportunity opened up for the KIPP DC. Data models led to data normalization across multiple building automation systems. Data outliers highlighted building optimization opportunities to lower energy usage. Fault detection and diagnostics (FDD) and computerized maintenance management software (CMMS) applications were deployed on top of the IDL to improve operational efficiency. KIPP DC gained vendor agnosticism, data to support the prioritization of capital projects, and faster and more trackable service work. 

We sat down with Nathan Morris of KIPP DC, Terry Herr of Intellimation, and Stephen Dawson-Haggerty of Normal Software (the IDL provider) to discuss how KIPP DC has taken the reins of their own facilities data and how all this behind-the-scenes effort inevitably leads to healthier environments for KIPP DC students and teachers to thrive.

Background

Nathan Morris has built a career around helping educate youth. He started as a teacher, ranging in various grades from kindergarten through middle school. He later transitioned to the facilities side of the business at DC Public Schools. He’s been the Director of Facilities at KIPP DC since 2020.

When KIPP DC started in the early 2000s, most of its schools were created out of new building construction. Hence, KIPP DC spent its earlier years more focused on growing than maintaining, as Morris explains, “Our buildings are what we like to call Frakenbuidlings, where we have continued to tack on and add square footage as we’ve needed to expand, which is nice in theory, but has created a real challenge.”

With this rapid growth, a clear facilities and operations plan naturally remained a back-burner topic compared to creating a state-of-the-art educational platform for their students. The lack of attention to facilities led to some less-than-ideal situations. “We actually have more building automation systems than we do campuses… one campus actually has two,” explained Morris.

KIPP DC has historically outsourced most of its facility operation jobs. It works with a third-party facilities company that employs full-time maintenance personnel for each building. KIPP DC also uses numerous mechanical contractors for preventative maintenance (PMs) and other service work.

Fast-forward through two decades of heat waves, cold spells, setpoint changes, and ambitious students. Suddenly, KIPP DC had systems that were no longer new and shiny but instead in need of serious service or replacement. As Morris came on board, KIPP DC needed to transition from development-focused to asset management-focused.

Unfortunately, KIPP DC wasn’t initially well-suited for an asset management focus. They had an arcane work order platform and tracked service work using clipboards, O&M manuals buried in basements, and “a bevy of nice spreadsheets.” 

The result was data that was more qualitative than quantitative, making it increasingly difficult for the school to determine what capital projects were most important and where to start. Too often, projects were triaged based on who yelled the loudest, which became a slippery slope to campus inequity.

Additionally, years of contracts and handshakes with different building technology vendors meant KIPP DC lacked autonomy over data and decision-making. For example, Morris mentioned how their work order platform is licensed to the third-party facility maintenance vendor they use. Therefore, they risked losing access to all work order data if they needed to change facility maintenance vendors.

Terry Herr is the president of Intellimation, the Pennsylvania-based master systems integrator (MSI) that Morris worked to convince his colleagues to trust and partner with to untangle the mess. Terry has spent over 30 years in the building automation industry. 2014 is about the time that Terry claims Intellimation was able to transition from a system integrator (SI) to a master systems integrator (MSI), shifting focus to data-driven retro-commissioning and optimization. 

When Intellimation came on board to support KIPP DC's journey to a complete digital facilities management solution, Herr quickly advocated getting an independent data layer (IDL) in place. Normal Software stood out to Herr and Morris as the best IDL option for KIPP DC for many reasons. First, there was the trust factor. Just as Morris had trust in Herr and Intellimation because of previous work they had done together, Intellimation had trust in Normal Software, an IDL platform they had been watching develop over the last couple of years. Additionally, Herr was well aware of KIPP DC’s data autonomy challenges and their aspirations of vendor agnosticism. Normal Software prides itself on being built on open-source technologies, giving the building owner the most control and autonomy over their data and systems.

Stephen Dawson-Haggerty, the founder of Normal Software, also joined us in the conversation. Dawson-Haggerty spent a portion of his earlier career developing the company Comfy, which initially “unlocked the thermostat” by allowing occupants to control the comfort of their room with the Comfy app. Through massive success, Comfy was acquired by Siemens Building Technologies and grew to be an occupant engagement app with numerous other features. 

After this journey, Dawson-Haggerty reset his ambitions and founded Normal Software. As Dawson-Haggerty puts it, “I felt there was a lack of great software actually doing amazing things. There’s a gap between the best practices people talk about and what you can actually buy and actually scale”.

Dawson-Haggerty emphasized his view of the IDL’s infancy in the market. Despite becoming well understood in the industry, there’s a difference between concept and actual success. Normal exists to take the IDL concept and determine the capabilities to make it work technically and as a sustainable business. 

With an IDL in place, new doors began to open up for KIPP DC in the form of user-specific applications that could be extensively evaluated for their performance and fit. KIPP DC slowly transitioned its facilities practices and capital planning into a data-driven mechanism - capable of determining what was most important and how to keep students and faculty happy.

Technical Overview

Phases of Development

Understanding what KIPP DC has done to its tech stack differs from understanding what it takes to actually get there. Morris, Herr, and Dawson-Haggerty walked us through the phases of deployment.

KIPP DC applied many of the concepts that Nexus Labs emphasizes through our Smart Buildings Buying Process Cheatsheet:

More info on the buying process cheat sheet HERE

Phase 1 - Defining the Objective

Morris recalled the project's origin, “The first phase was working backward to determine what we wanted to achieve. What sort of things did we want to get out of it, and what was going to help us run our campuses and network more efficiently from a facility perspective?”

It’s often overlooked, but Morris is describing the strategy developed before any action was taken. Many unsuccessful projects immediately jump to the right column in the image above: procurement. Before engaging vendors, Morris and his team at KIPP DC spent enough time in the strategy stage to understand what they actually wanted and needed. In the case of KIPP DC, the strategy centered around achieving data autonomy and obtaining better insights into maintenance and capital project opportunities. 

Regarding preliminary market research on what categories exist in the market, the KIPP DC team outsourced much of the effort by bringing on Intellimation to help them determine which products would best help them reach their goals.

Phase 2 - Harvesting the Data & Centralizing the BAS

Herr described how Intellimation started working once they were brought into the project: “It all starts with trending the data.” According to Herr, most BAS trend only about 10% of the available data. Intellimation immediately deployed the Normal Software, which acted as a central hub to gather as much trend data as possible.

It always starts with an independent data layer, or a middleware layer, to bring all the data together, trend all the data in a neutral, non-BAS fashion. And it always includes a central and typical BAS front end, but a single one instead of five or six different ones, which is what KIPP DC had.”
Terry Herr

The Normal Software IDL stored the trending data from multiple systems in KIPP DC’s Amazon Web Services (AWS) account. This meant that KIPP DC, and no one else, owned the data gathered. If KIPP DC were to part ways with Normal Software or Intellimation for any reason, they would still have full access and control of this data.

In parallel, Intellimation worked to slay KIPP DC’s eight-headed BAS monster. Intellimation helped KIPP DC evaluate three different BAS options as a single front end to their building automation.

Referencing the buying process cheat sheet above, Intellimation was shortlisting BAS providers based on KIPP DC’s goals and use cases. Check out the Nexus Buyer’s Guide to HVAC Control for more support on shortlisting BAS providers.

Herr described how the BAS vendor evaluation mutually benefits the MSI and the building owner. Intellimation always strives to stay informed on which products and features lead the market. Every chance Intellimation gets to evaluate vendors is a chance for them to rethink their preferred partners for future projects. This continuous vendor evaluation is the same reason Intellimation was able to bring Normal Software into the project quickly and confidently.

Once a single BAS was selected, Intellimation began the tedious task of developing all graphics into that single front end of building automation. A universal graphical representation of systems, regardless of the campus where the system is located, immediately improved the operational efficiency of the facilities team.

Phase 3 - Data Models and Optimization

Not all data is created equal. After centralizing trending data into Normal Software, Intellimation helped KIPP DC develop data models that provided a visual perspective of what was occurring in the buildings. The team utilized Grafana, an open source platform, to start creating logical stories around data gathered. 

Intellimation’s approach typically focuses on building optimization once the IDL is in place. Starting with optimization is often the easiest way to show some quantifiable results around the system overhaul. Morris pointed out, “Optimization is how we get our organization on board. With a combination of cost savings from energy reductions to building energy performance requirements in DC, like BEPS, we can start to show results.”

Building Energy Performance Standards (BEPS) are regulations or policies that set specific energy efficiency targets or performance criteria for existing buildings to meet over time. These standards are part of broader efforts to reduce energy consumption, lower carbon emissions, and improve the overall energy performance of the built environment.

A stakeholder outside of the building technology industry, like a CFO or board of directors, may not understand concepts like the IDL and can lose patience in these types of capital investments if results can’t be shown early. The data model and optimization of existing building sequences can be a time-sensitive outcome for facility managers trying to get buy-in on their approach.

Phase 4 - The “Bake-off” at the Application Layer
“With an IDL, you’re going to unlock a lot more deployment opportunities because it’s easier to try out different FDD products. Instead of doing a crazy six-month integration after a brutal RFP, you can just kind of do a bake-off between vendors, and the cost should be low enough.
—Stephen Dawson-Haggerty

As Dawson-Haggerty explained, one of the most valuable attributes of getting a reliable IDL in place is how it simplifies and reduces the cost of evaluating solutions at the application layer. 

As a reminder, the application layer sits above the data layer. It is where organizations can apply use-case-specific views and tools on top of the communal data so that each stakeholder can get what they need.

When a robust data layer is in place, vendors at the application layer have minimal effort to pull the necessary data into their software. Therefore, building owners can quickly test out application layer software with minimal effort and cost. This is the “bake off” concept the Dawson-Haggerty explained.

In the latest stage of this project, KIPP DC has been working closely with Intellimation to perform bake-offs for FDD and CMMS applications. 

As the KIPP DC team evaluates applications, they have gained an upper hand in contract negotiations. They have fewer worries about getting locked into license agreements for products or services because the data is owned by KIPP DC, in their own AWS platform, and accessed by the vendors. 

Normal Software tries to differentiate itself by supporting this concept of data autonomy, which is part of why they were such a good option for KIPP DC. Herr describes Normal Software as “truly open source” because it operates on the BACnet stack, programs in Javascript (a language college students learn in class, not a proprietary commercial development), and talks from the edge to the cloud in MQTT.

Additionally, Normal Software’s IDL is a non-SaaS offering. KIPP DC does not pay a recurring fee to Normal after the IDL is stood up, compared to competitors which require an annual licensing fee or hardware upgrades after old hardware gets phased out. However, Normal does compete at the application layer by offering unique applications on top of the IDL to meet a customer’s specific needs.

What about the IDL and MSI bake-off?

One thing that may stand out to readers of this case study is the emphasis on vendor comparisons to help you get the right product or service. However, KIPP DC quickly jumped into a partnership with Intellimation as an MSI and Normal Software as an IDL. What about vendor selection for the IDL and MSI?

There’s a fine line between the speed of deployment and results versus analysis paralysis in vendor selection. Spending too much time on vendor selection for the IDL or an MSI could have resulted in stakeholder impatience. Morris was fortunate to have previous experience with Herr and Intellimation to the point that there was already an established level of trust. Similarly, Herr and Intellimation had prior experience with Dawson-Haggerty and Normal Software to feel confident it would address the needs of KIPP DC.

Moreover, Dawson-Haggerty explained how the concept of an ROI can fall apart when it comes to an IDL. If an IDL finally gives a building owner visibility into their building data, allowing them to make better decisions, how can you assign a value to that? Dawson-Haggerty equated this to asking someone the ROI of their email system, ERP, or CMS; if something is fundamental to getting a job done, it can be challenging to assign a value to it.

The Nexus community has long debated when to apply IDLs and when to add application layer software. Not all agree that the procedure KIPP DC took is the best option. For a broader range of opinions, we recommend you listen to our SME Debate on the IDL.

Quantifying Success at the Application Layer

Despite the challenges with tagging a value to the IDL layer, KIPP DC is undoubtedly experiencing positive results with its new autonomous and converged data layer. KIPP DC continues improving its FDD, CMMS, and even access control system applications on top of the IDL.

Morris describes the number one benefit as the end-user experience. KIPP DC facilities finally have metrics to gauge how quickly they can resolve issues that are coming up. Morris equates this to the famous Dominos Pizza Tracker, but instead of responding to a pizza order, it’s responding to an uncomfortable occupant.

KIPP DC is finally confident in its ability to make data-driven decisions regarding capital investments. “There’s always too many projects and not enough capital to support them. Unfortunately, it’s been a very subjective process to date. A lot of times, the loudest voices win out,” said Morris. KIPP DC is proud to have improved fairness and equity and defend decisions with data. Morris brought up one specific example of kitchens with commercial cooking equipment in rooms that were not appropriately retrofitted to be kitchens. They had staff members experiencing temperatures in the high 80s. Precise data highlighted the poor conditions of these rooms compared to others and allowed KIPP DC to prioritize investing in mini-split systems to provide a better space to the occupants.

Lastly, KIPP DC’s third-party service providers benefit from richer and more accessible building data sources. All these partners are getting trained to use the new applications, like FDD and CMMS. Herr brought up the specific example of technical data on the compressors of VRF systems that were previously unavailable without a trip to the roof. Having this data accessible in the cloud allows service technicians to be significantly more efficient while servicing these VRFs.

Challenges & Lessons Learned

These case study articles aim to share success stories in a way that can benefit similar building owners by helping them make accelerated decisions toward improved tech. Morris was willing to share the largest obstacles to getting to where they are today, hoping other building owners can learn from their mistakes.

Lesson #1 - Start by determining what you actually need

The IDL unlocks visibility into all of your data, but it can be challenging to benefit from that data if you don’t know what you are looking for. Morris noted the importance of determining what data you need to make informed decisions before starting the project. Otherwise, you may spend too much time chasing ghosts within your data.

Morris and KIPP DC determined two overarching goals early in the process that helped with their decision-making:

  1. They needed data to help provide quantitative analysis of the state of their assets, which would allow them to fairly prioritize capital investment projects.
  2. They needed to be the sole owners of their data so they were no longer beholden to old contracts and agreements and could achieve the freedom associated with vendor/product agnosticism.

Lesson #2 - Do your research and find trustworthy partners

Morris had the humility to recognize that he doesn’t understand all the concepts of a building technology stack and probably doesn’t need to. He was able to trust Intellimation’s technical knowledge so that he could focus on all the other vital aspects of his role. Instead of trying to understand every intricacy, Morris prides himself on having a decent bulls*** detector, asking the right questions, and reading the fine print. 

Morris shared a horror story from his past of partnering with a vendor on a performance contract, only to later learn that the performance contract was based on the vendor’s projected energy savings, not actual energy savings. Additionally, most energy-saving measures were behavior-based, like wider setpoint ranges, which put his organization at risk of having to pay the vendor more than they were saving on utilities. Reading the fine print and relying on trusted partners is paramount to avoiding these situations.

Morris and KIPP DC also benefited from their close relationship with KIPP Texas, a larger school district that has experienced many of the same growing pains from a facility/operations standpoint.

Lesson #3 - Learning how to navigate the political landscape

Understanding all the complicated relationships between stakeholders is critical, especially in the municipal sector. Morris spends a significant amount of time understanding the goals of the school board and the local politicians to stay ahead of ever-changing compliance and regulation goalposts. 

Speaking to a project's bottom line is critical to winning support for a project, but it can be challenging. In the case of KIPP DC’s IDL deployment, they started with little data, so it was complicated to show hard numbers behind the benefit of the project. When that is the case, Morris recommends focusing on “how much time are we spending on this now, and how well are we doing it,” which, although qualitative, can help stakeholders wrap their minds around what improving facilities can do for the organization's overall goals.

Herr used indoor air quality as a specific example in this instance. With math testing scores at an all-time low across the US, anything that can be tied to a positive impact on student learning is of astronomical importance. Gathering IAQ data was a quick win for the project and attracted the attention of stakeholders.

Conclusion

KIPP DC set out in the early 2000s, determined to create the best learning environment possible. Each new campus became a new opportunity, and the mindset was focused on growing rather than maintaining. When tasked with shaping the minds of the next generation, there’s much more to focus on than the condition of a building. Soon, a tangled mess of building products, services, and vendors grew. But when Morris and the vendor team set out to right the ship, they succeeded by applying values no different than what was being taught to the youth in their classrooms:

Building Trust: KIPP DC put significant trust into Intellimation and Normal as their partners, but the trust was not blind. KIPP DC focused on finding partners who can help them achieve their unique goals and relied on the previous experiences of others to feel confident in their chosen relationships.

Making a plan and sticking to it: Too often, the method for deploying building technologies is “point and shoot.” KIPP DC could’ve jumped to an unlimited number of “solutions” to improve operational efficiency, but instead, they invested in getting a functioning IDL. Although the IDL can be difficult to attach an ROI-style value to, it acts as a platform for enhanced decision-making.

Going from siloed building systems to an IDL is like an entity that was once blind gaining site. With data now available to KIPP DC, what comes next is limitless. But it all starts with untangling the mess, one knot at a time, and building the correct foundation to grow upon.

⭐️ Pro Article

This article is for Nexus Pro members only

Upgrade to Nexus Pro
⭐️ Pro Article

This article is for Nexus Pro members only

Upgrade to Nexus Pro

Are you a Nexus Pro member yet? Join now to get access to our community of 600+ members.

Join Today

Have you taken our Smart Building Strategist Course yet? Sign up to get access to our courses platform.

Enroll Now

Get the renowned Nexus Newsletter

Access the Nexus Community

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 Connect

Upgrade to Nexus Pro

Join 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