Solving the Three Layers of Predictive Maintenance in Commercial Trades | Podcast Ep. 291

Hello contractors and welcome to Toolbox for the Trades. I'm Amanda Salvatore, and today we are hearing from 2 people thinking hard about where the industry is headed. Evan Smith, Group Product Manager, and Leila Rookstool, Senior Industry Advisor for Commercial for ServiceTitan. Between Evan's product lens and Leila's 4 generations in the trades, this is a conversation worth your time. We're talking about reactive proactive and predictive service, where the industry sits today, what's standing in the way of true predictive service, and what it could mean for contractors when equipment failures get caught before they happen. Before we dive in, this is a vision conversation. We are talking about what might be possible for the trades, where the data, the technology, and the industry could go. Any specific features or capabilities we mention today are just that— ideas we're exploring out loud. This is not a roadmap or a commitment from ServiceTitan. I hope you enjoy this conversation as much as I did. Evan Smith, you are the group product manager, and Leila Rookstool, you are the senior industry advisor for commercial at ServiceTitan. And today we are gonna be discussing what the future of predictive service looks like for the trades industry and how we are moving it forward. Thank you both for coming on Toolbox for the Trades. Happy to be here. Thanks for having us. We're really excited to talk about this topic and learn more about it and what ServiceTitan is doing to move that initiative forward. But we always like to start off with a little icebreaker. And since we are going to be releasing this episode around back to school, I would love to know what was your favorite school supply purchase as a student? Your question is very timely, Amanda, because I was up till midnight last night ordering school supplies for, for my 3 children. So little PTSD there. But yeah, I mean, I, I was kind of a nerd in the sense that I loved all the school supplies. Um, but I would probably say the, you know, picking out the backpack whenever I got a new backpack, that was always fun and exciting. One of the latest trendy things. Right. It sets the tone for the year. I feel like the backpack does. Evan, how about you?

I think I have to go with the TI-83 calculator for a couple of reasons. First, I'm from Texas myself, so Texas Instruments, shout out. The TI-83 had a game on it called Block Dude. And we loved that game. So I'd have to go with TI-83.

Mentioning that calculator gave me some PTSD. So before we get into the nitty-gritty of predictive services, I would love to hear how you both found yourself in this industry and with ServiceTitan. Leila, how about we start with you? Yeah, I, I guess you could probably say, Amanda, I was, I was born and raised in it. So my, my great-grandfather started what was originally a coal company over 100 years ago, and it transformed into, you know, coal is what people were using to heat their homes. And, and, you know, in the '50s kind of transformed into a heating and air business. And so proudly part of 4 generations of being in the trade, specifically mechanical, electrical, plumbing, and spent a, a good significant part of my career with a, a really great organization called Comfort Systems USA. In a number of varying roles. I worked at a local operating company here in Tennessee where I live for a number of years doing sales, marketing, business development, and transitioned into supporting the broader Comfort Systems toward the latter part of my career with them and really leaned in on the service side and working with our sales team. So I've had the experience of working with a lot of different contractors across the US, had a unique opportunity to join ServiceTitan here and, and super pumped and excited for all the things that we're working on right now on the commercial end. It's a really cool thing to be part of it and to kind of peek behind the curtain on how all this works from a product standpoint. Incredible. And Evan, how about you?

For the majority of my career, worked in B2B product management. Prior to ServiceTitan, was leading the conversational AI platform at Best Buy. Fancy title for the chatbot on bestbuy.com and in the mobile apps. There we were building out experiences for customers to troubleshoot their own issues with products they had purchased from Best Buy, for example, TVs. We were indexing TV manuals and returning steps by step to the user when chatting on the website, as well as helping folks schedule and manage their upcoming Geek Squad appointments, whether Geek Squad was coming to their house or they were coming into the store. Tons of overlap actually with some of the core functionality of ServiceTitan. My wife and I were moving to LA. This is about 5 years ago. Wow. Coming up on 6 years ago and made my way to ServiceTitan.

You've all had incredible journeys. Let's get into the nitty-gritty. We're going to set the stage. We have reactive, proactive, and predictive services. What do they look like and where does the industry actually sit today on that spectrum? I would say that this industry has seen, of course, a lot of transformation over however many, the past many years. I mentioned that, you know, my family's been in the business for over 100 years and you think kind of when you talk about reactive, that's kind of how we started, right? Where service calls were being placed 100% via phone, work orders were being completed on paper. There was no digital trail that exists today. And When I say that this industry is seeing transformation more and more, there is a, a gravitational pull to, to start leveraging the data that is now at our fingertips that was never achievable or possible before when we were operating on pen and paper, right? We find ourselves in this place of, you know, now that AI is, is even more and more, uh, of a subject in the industry of reexamining What the customer experience looks like, right, Amanda? And it's completely transformative to what we used to know. And I'm proud to be part of a team here. We're really starting to kind of push the boundaries and think beyond what we traditionally know of service and what that experience looks like. I think what you probably see within the industry, you've got contractors that have BMS or controls capabilities in place. that do provide some level of predictive service to owners that can access that data right there on site and on premise. But, you know, I think what we're going to be talking about today is even beyond that, what that looks like and how data and systems can work to support that, to even bolster that, to give more context and more predictability that wasn't, you know, possible before. For anyone who is hearing predictive service for the first time. What does that actually look like in practice, boots on the ground? I would say, in short, when we think about what this actually looks like in practice, Amanda, I mean, it's predicting equipment failures before they happen, right? That's the end goal, and that's what we're aiming for. But as we're having this conversation, I think we first need to set the stage that when we talked about predictive service, I really look at it through the lens of of the experience that we're creating for service agreements customers, right? Because those are the customers that we really have a vested interest in the health and the longevity of their equipment. So when I think about how this could tangibly unfold, you could think of a scenario where back office now has visibility into data or statistics that's leaning toward potential compressor failure on a particular unit, right? And that could be stemming from a number of different things, which I think we'll start to unpack probably as part of this conversation, whether that's, you know, direct from sensors or physical hardware on the equipment that we're monitoring, or it is historical data or context that's specific to that make and model. There's lots that we could leverage to help, you know, give us some insights as to, you know, why there, there could be a potential compressor failure on this piece of equipment. But I, I think even a, you know, a precursor to that, or maybe even a predictive light version That we could think about too is in how we support a feature to where a technician is dispatched out on a call. And even before that, that he or she arrives on site, they have visibility into the context around, as I mentioned, you know, make and model data and statistics that could support, you know, for this particular unit, 60% of the time what we see based on the information we've collected when we set up the service call is that this is going to be a condenser fan motor failure. We need to replace the fan motor. And so when the technician gets on site, the diagnostics are a lot more quicker, sometimes more accurate, especially as it relates to greener technicians. And so while that doesn't have the impact that we talked about previously where there's, you know, true remote monitoring of that building an even stronger case, that's still a scenario where you're using prediction to help with, you know, uptime versus downtime, and that creates a really more positive impact for the customer and a different customer experience than, you know, what a lot of contractors are able to deliver today. Evan, I would love to know why you think that this is important for trades companies and why it's important for the customer as well. I know, Leila, you went into a little bit of that, but I'd love to hear your take on it, Evan.

I think about What is service in the eyes of contractors and customers? I think for a long time it's been thought of as, as really the work order lifecycle. So there is kind of that initial scheduling, there is the technician doing some triage or diagnosis and proposing a solution, ensuring they have the right parts for that solution, closing that out, right? All the work order lifecycle. Why this is important is the predictive and the proactive aspects of this go beyond the lifecycle of a single work order. And so every step of that process I mentioned, there's an opportunity to do things faster or have things happen. So for example, on scheduling, historically, a customer will call a contractor. In this paradigm that we're moving towards, that in theory doesn't have to happen. And the contractor is instead calling the customer and saying, based on what we know about your location, your equipment, your past service history, all of the information that we have, we recommend we're going to come out and do X, Y, and Z, and here's why. Kind of being on the front foot even from scheduling. When it comes to triage and diagnosis, again, building on that data, a technician can walk in not with a blank slate or starting from scratch, but instead are starting from the most high probability recommendation of what they're probably going to find there, given historical information that they have about that location, given historical information that the contractor has across their customer base. Parts ordering, right? If we have a very high likelihood of knowing what the problem is, even in advance of being there, in an ideal world, we can have the right part then and there, right? So same-day service, same-day fix, reduced downtime or uncomfortableness, you know, for the end customer as examples of even within that work order lifecycle and beyond it, how contractors can take advantage of the data and the technology coming to life.

How is ServiceTitan moving the industry closer to predictive services with its shared data foundation? Evan?

In plain terms, some of the functionality coming to life within the ServiceTitan platform right now today is pushing us and contractors, pushing the industry in that direction. So for example, from the mobile device, allowing technicians to capture everything That they are seeing and saying, and using our Atlas platform behind the scenes, then structuring that data not only for that contractor, for that particular visit and customer, but for any other future visit that happens to look similar to that customer or have the equipment that's similar. So additional functionality for technicians to capture all of those things. To make their job easier the next time they go out there or make their coworkers' job easier next time they go out there. In addition to that, feeding information captured by technicians, the eyes and ears of the contractor, back to the sales team and the account management team to do a better job of proposing the right solutions to that customer to prevent downtime, particularly if it's a critical facility or high criticality equipment, having a contractor be in that driver's seat as a true consultant, a proactive consultant to the customer, becomes possible by enabling contractors to gather up that sort of data. The last piece that I'd add coming to life here is with those problem solution pairs, ServiceTitan has the capability to then suggest what should the estimate be, what should the fix be. And we can then move more contractors, more technicians who many by my trade are maybe more focused on the technical aspects of the job, right? They're mechanical experts, they're technician experts, in some cases maybe not as interested or trained in sales. The more that ServiceTitan can do to help drive that, and make some of those sales-type decisions enables more technicians to be more sales-focused with the support from the ServiceTitan platform.

Let's talk about the contractors themselves. They have this data in hand with ServiceTitan. How are they actually putting it to work right now? What are some practical examples? I think you heard Evan share about, you know, some exciting features that we are in the process of unveiling here at ServiceTitan and how they're actually putting it to work today. There's much that you can see right within our FieldPro product and what starts to surface for the technician in terms of insights when they scan the OCR, they scan the data plate with the OCR, and it pulls out the make, model, and serial. I love absolutely for a technician to have the data right there in hand in terms of the service history particular to that unit as they're working on it, because that's really important context for any technician working in the field. I'm proud to be part of a team that recognizes garbage in, garbage out. And so a lot of the solutions that we're building toward is enabling a strong data foundation, because we know any futures that you talk about with AI, you know, AI is fruitless without that strong data foundation. And so there's lots that I'm seeing in terms of what we're doing on the sales end to enable green salespeople that are brand new to the industry. They don't know what sort of equipment they're looking like. And so how can we support them to understand when I'm doing an equipment survey, this is the equipment type that I need to designate, the right capacity, and the trickle effect that it has from performing the PM in the field for the technicians and into, again, the strong data foundation that we're building, because that's what's going to make predictive service possible in the future, right, is building that strong foundation. And so there's much that's in place today, but there's a lot that's coming in the future that's really exciting in terms of how contractors will be able to use the data.

If I may, I have to speak to preventative maintenance, or PMs. We refer to them as service agreements within ServiceTitan, but really for that PM case and specifically in hand today, we go out there and we, as a contractor, we capture information about the equipment and the location for the purposes of proposing a great maintenance plan, bespoke tailored plan to that customer. So one example of how we're leveraging that today is based on the things Leila mentioned, we're capturing the right make model, the type of the equipment, the filter sizes, the belt sizes, right, the pieces of materials that are often included in a comprehensive maintenance plan. That data captured in the field flows directly into the proposed service agreement. The tighter we can create those loops within ServiceTitan, the richer data contractors are going to be able to capture and flow directly into creating a more personalized and more accurate a more prescriptive and ultimately more professional service agreement proposal. Not to mention the things— oh, I guess all of those great things I just mentioned, can't forget about speed. We've heard from contractors where in the absence of any differences in the proposed maintenance plan, same price, same visit schedule, same coverage, the contractor who delivers the proposal to the customer first is going to win, you know, not to mention the improvements to those proposals, but the tightness of that cycle, capturing the information and turning it around immediately to propose is another example of leveraging the data within ServiceTitan.

Well, and that much time that those salespeople could be focusing on additional opportunities rather than manually keying in all of the information. So yeah, speed and precision for sure.

100%.

You're talking about building on this foundation. How is ServiceTitan continuing to build and enrich that dataset that we're speaking about?

I think we're looking at this in 2 ways. First, it's internal or within the ServiceTitan platform, and then the broader ecosystem around or surrounding the platform. And Leila said it earlier, garbage in, garbage out. So we can, we can use that as a, as a way to frame things as well. Contractors are creating mountains of data just simply by existing as a business and doing their daily operations. And one of the goals within the ServiceTitan platform is without getting in a contractor's way, without getting in a technician's way, how can we do our best to help contractors structure that data in a way then that models can be built and it can be leveraged? For other use cases, so it's striking that balance between. For example, we don't want to put something in front of a technician that would require them to spend 20 minutes filling out a data labeling sheet. That's a non starter. But especially with the technology and LLMs available today, if we can help the dispatcher, the technician, and all of the players involved at a contractor produce. Data in a somewhat standard way, we can then normalize and classify that with the technologies available today to have that sitting in a very clean, structured, rich dataset that again can then be leveraged for, for other use cases. So that's how ServiceTitan is building, enriching that dataset within the platform. Speaking to the ecosystem around the platform, we have partnerships today with third parties enriching and augmenting the data we have and contractors are creating within our own system. That partnership ecosystem continues to grow. That could be on the technician side, supporting diagnostics and manuals and troubleshooting. That is on the procurement and supplier side. You name it, it's those additional third-party or external sources augmenting and building that foundation within ServiceTitan.

What are the biggest gaps standing between where we are today and true predictive analytics? The way I think about it, I almost process this in layers, Amanda. The first layer being what can we know and understand about a singular piece of equipment? So basically tying back whatever data we can to a serial number. In other words, trying to understand what the service history looks like for this particular unit. so that it can inform the next service call, bringing in the context, right? The next layer that I think about that supports this predictability future is the make and model level context. So looking at any sort of historical data or context that we can bring in that's specific to, you know, this particular manufacturer and this model that could tell a broader story, kind of going back to that similar scenario I brought up around You know, when a technician is going out on a service call, what data can we give the technician ahead of time with that particular context to say, look, you know, based on the data on what we know here, it could be 60% this or 70% that to help them diagnose even faster and more accurately. The third thing that when we talk about gaps and we talk about, you know, really putting our heads together as an industry is figuring about this third layer of physical connectivity, you know, whether that's through sensors. We talked earlier about, you know, BMS and control systems, how those exist today, and they do give levels of predictability to a building owner. But even within, you know, that particular scenario, what happens a lot of times is you've got a service agreement in place where the controls may be on a whole separate agreement under a different provider. And so even sometimes the contractor that's performing that PM doesn't even have access to that data. And even those contractors that do have a controls or BMS capability, that data is still operating disparately or disconnected from the FMS system, which houses, you know, everything that we're talking about from the sales and operations, the problem and solution that Evan is talking about. So when I think about it is the gap exists in being able to provide maybe even that physical connection through a sensor kit or whatever that may look like, or finding a connection even to the fault detection systems to where we can get another layer of predictability as it relates to assets and equipment. But we've got to tie all of that together. This is all still disjointed and disparate. That's probably where I would say the industry is thinking long and hard about that, that third layer and how do we remotely monitor in a more affordable and streamlined way to achieve even more predictive results. So we understand where we need to go. Let's zoom out and talk about the future. What does the future of predictive service look like? for the industry, and where does ServiceTitan see itself driving that forward?

I think it builds on what Leila mentioned with, with kind of those 3 layers. Speaking with Leila, speaking with really forward-thinking contractors, one way to express the future is in terms of uptime as a service, where the customer is not paying for on-demand service onesie-twosie, Instead, what the customer is paying for and expecting from the contractor is that their equipment is never down, their facility never has to shut down, their customers are never uncomfortable or unable to do, you know, what they need to do in their facility. And so ServiceTitan today, very strong in kind of those first early layers. We've been talking about the things we're building in kind of that second layer. Moving beyond that is where I think we look towards the additional data in the world currently outside of the ServiceTitan platform and pulling that into ServiceTitan for the benefit of contractors to realize what could be the future of what the market expects, which is my equipment should never be down for any moment.

I feel like the industry too is like at a really big inflection point right now because we've got such a tremendous amount of backlog flooding into data center space and hyperscaler work. And when you think about like the criticality of the equipment, particularly just within that vertical market, you know, predictive service becomes even that much more valuable. And so more and more of these, these buildings and facilities are entering our space, you know, service, construction's playing the role right now, but service is gonna come in right after, after all of these are, you know, fully completed and, and now have to be maintained. And I think as an industry too, it's a really great time for us to be talking about this. And again, a really good inflection point for us because the higher the criticality, the more value that this, that this predictability brings. And, you know, not just for data centers, but for particularly for manufacturing. Uptime is extremely valuable versus downtime in that scenario where, you know, the equipment could literally be supporting the process as a whole to keep things running and moving from a manufacturing standpoint. So. This has been a great conversation. There's so much to look forward to that is really going to change the game for a lot in this industry. Before we go, I would love to know what gets you personally most excited about where this is headed. Leila, how about we start with you? Ooh, it's probably the part earlier where Evan was talking about sales and operations being connected in one system. And, you know, this whole vision building toward let, let's stop operating in these disparate spaces and let's bring all of this together because, you know, that, that's what's gonna create a more positive future for our customers, for the industry. I would just say let's kill the disparity. That's the part that gets me really, really excited is bringing all this together within a, a central hub, a central brain. to make all of this possible and achievable?

Maybe a little biased as a, a ServiceTitan livery here, but, uh, the Atlas platform within ServiceTitan, that's effectively the surface area that is materializing those algorithms, those recommendations that, that we've been referring to. And I think specifically is, and I would love this in my, my own job, but thinking as a business owner of a contracting business, And coming into the office and instead of asking the question, what should I pay attention to today? ServiceTitan and Atlas effectively telling the business owner, here's what you should pay attention to and why. Completely turning that interaction daily on its head is super duper exciting. The second answer, and this is maybe more non-biased, but as a homeowner, As a side business owner, I get excited about the level of service that contractors will be able to provide to me as a consumer. And just the level of comfort, the level of service, the level of knowledge that contractors can provide to me as either a homeowner or a business owner is exciting from that perspective.

Evan, Leila, I want to thank you both for coming on Toolbox for the Trades. What a great conversation.

My pleasure. Thank you.

Yeah, thanks for having us.