Natan: Lisa, welcome to Augmented Ops. This is going to be an exciting episode. I'm really glad to have you on. Welcome. 

Lisa: Thank you, Natan. It's a pleasure to be on Augmented Ops. Thank you so much. 

Natan: Yeah, and for those of you who don't know Lisa, and she will introduce herself in a second here, Lisa is the CEO of Seeq.

What is Seeq and how did you get to start working on that? 

Lisa: So, very quickly, I'm Dr. Lisa Graham, and as CEO of Seeq Corporation, I have had a career across many process manufacturing industries around getting more value out of data to drive better products. So, I'm a chemical engineer by training and a registered professional engineer.

And I tell people that I've had the privilege of serving on the front lines, creating products we use every day, like from the pulp and paper plants, silt and wafer manufacturing, to be on the chemical plant floor, and then 12 years in pharma and life sciences, developing and launching new medicines. And so I've been connected to Seeq since early 2016 as the first partner and user of Seeq software.

Natan: Oh, wow. Really? So you ended up coming on board from a perspective of a customer. That's right. That's amazing. 

Lisa: Thank you. I had had an exit and was thinking, Oh, you know, I think I'm going to go focus on my family and my husband's business. And I started an engineering consulting company and the thought was, well, given the format of Could Seeq be useful in pharma?

And so for three years, that's what I did. I used Seeq myself to do good advanced analytics work, and then over time decided to join Seeq in 2018, and then I've just come up to the ranks a bit. 

Natan: So that's a great Genesis story, but for those who are listening who don't know what Seeq is doing, can you describe What is the product and why is it so special given, you know, that the landscape of business intelligence tools is very wide and rarely people understand the verticalization of this space.

What is Seeq doing differently in this space and why is it so important for people on frontlines? 

Lisa: Okay, well, maybe by way of starting to answer that question, it's important that we recognize data and analytics are a critical part of operations. I mean, flat out from operations to process, equipment, monitoring.

Natan: Hey, you don't need to convince me, you know, I'm, I'm totally with you. Yeah. 

Lisa: So that's what enables the key intelligence for driving what everyone's talking about. We need better products with less energy in a profitable way that is sustainable. And so at Seeq, We are first and foremost providing access to that data to the people who really need to use it.

Natan: And those people, you often hear things like, well, who needs the data? Everybody needs the data, you know, and then you still go into large and small companies and you find 20 data tools. So when you talk about your people who need the data, who are they? Who are those people? 

Lisa: You know, it's interesting. When we first started, we were focused on engineers, because at the heart of what we People who are changing how things are run tend to be the engineers, whether they're in energy or pharma or chemicals or any number of industries that we support.

What we've recognized over the last several years is that to really make change, we are supporting all persons, all functions. So that is. Where workflows change, that means we need involvement with the information, engineers, operators. It's why we look to partner with companies that serve other people, because together we can do it more broadly.

Natan: That's great background, Lisa. Can you perhaps give us a few examples of Seeq customers and how they're using Seeq to create unique value with data? 

Lisa: Sure, a couple examples. So first, we've seen international pharma organizations deploy Seeq as an operational equipment effectiveness. So otherwise usually called OEE platform.

Natan: Yeah, we know OEE very well. 

Lisa: Yeah, and through deploying Seeq for that. They're using it to understand production losses and have achieved more than 8 million in annual savings from those insights for just one deployment on one site. And the second one, enterprise monitoring, is another area where we've seen very recent success of companies that have been working with Seeq for a year or two now.

Seeing the opportunity to roll that pilot solution out globally and Flint Hills resources. You Seeq to run more than 80, 000 surveillance tests across its assets. 

Natan: Oh, wow. What, what's a surveillance test for those who don't know? What does it mean? It sounds scary. 

Lisa: So it's making sure that every asset is operating within the band of expected parameters.

So it's a chance to proactively understand where you might do predictive maintenance as opposed to even Time intensive, costly, cost based maintenance. 

Natan: Sometimes I hear people call this machine health. So understand how well is the machine. And now can you predict when it will fail, for example? 

Lisa: That's right.

Exactly. So that avoids unnecessary downtime, unplanned outages, and ultimately, uh, seeing customers get more throughput in the facilities because they have much more control over when and how their equipment's operating. 

Natan: Yeah. And they have visibility. 

Lisa: Yeah. Exactly. 

Natan: Yeah. And you know, when I was playing around with Seeq and thank you for letting us play.

One thing I noticed is that there's a lot of emphasis in the design on making it easy for people to access and explore the data. So I'm an engineer too. I'm not necessarily a data scientist. What I think both our companies and products are kind of focused on is democratizing this access, you know, like letting people figure out how to get stuff done.

And in a way, in your story, I'm hearing it's a tool that you've used as a customer. It's really hard to build. I mean, when you perspective change from the engineer using the tool to one of the leaders and now the CEO of the company, how do you maintain that? How do you keep it such that you're staying close to the customers and keeping it simple and easy to use so everybody can do it as you mentioned?

Lisa: I mean, there's lots of things I could say at the core of that is staying very connected to the customer. It's really being intentional to listen, hear what they're saying, hear what they're thinking, understand what they're trying to accomplish, knowing that that's evolving over time. And I think, well, I know Seeq has evolved well in response to customers saying, Hey, We love what we're doing with Seeq.

We're getting this value out of it. Could you help us do X? 

Natan: Do you have a story like that you want to share? 

Lisa: Yeah, probably one of my favorite is around supply chain management. This came to us from a C suite business persona. So this is a person who's Wanting to actually drive their pricing strategy and what they need to be buying when, and in order to do that, you need effective plant forecast, how much raw material is going to be used, how much of this is going to be produced.

So we have evolved Seeq such that that business owner who's trying to mitigate supply chain disruption can work with a process engineer. to quickly identify what their pricing and strategy should be. And the engineer turns around and makes the most financially vetted changes to the process, what could be most effective at being rewarding changes from a financial perspective.

So we've evolved it to connect to the data that would provide not only the engineering data from the plant, the time series data, but all that context that's needed so that those financial questions and those business questions can now be answered. 

Natan: And, you know, last time we met. We spent some time during our global annual event in Operations Calling, which was really focused on an ecosystem, and we'll talk about that in a little bit, but, uh, I've learned a lot on like, why is it important to think about the vertical that you're serving?

How would you define, like, you're, you know, in operation, which is very wide, who needs the type of BI that Seeq is providing? And how are you thinking about this? In other words, like, who should immediately, like, drop everything they're doing and go check out what you guys are doing? Because otherwise, they're going to be less competitive without your tool.

Lisa: I love that. So, frankly, I think every company and the people who are trying to lead in those companies and, and not only keep up with those competitors in their space, but outpace and set that knows that there's wasted time and efficiency because spending time and having people do things, that's not what they went to school to do.

So for example, engineers, they didn't go to school to spend hours and hours a week. Cutting and pasting in Excel only to show up in a report that is a week old in the operating space says yeah We don't believe it. It doesn't have all this context. So instead companies who want to envision they show up and Shifts are making changes to the process to drive efficiency and profitability and they're doing it with visibility to the information The engineers are there And leadership can see it as well.

So it's really a shift towards people who want to have a more team approach, and cutting out the wasted days of things getting reiterated. Good example, uh, we have an amazing customer who spent years trying to create intelligent alerts. Essentially, where there's a logic built in for, you know, why is this alerting and what should I do with it?

So if you picture a company that maybe has 3, 000 wells, what they're doing before Seeq is maybe have eyes on a handful of those, maybe in as much as 100 or 500, which means they might be able to optimize those, but they don't have the insight to optimize the rest. You implement Seeq and empower people now to create those intelligent alerts.

They've moved to exception based surveillance monitoring on all the wells. So anybody that would like to be able to go say, hey, I can see that now, should come chat. 

Natan: Okay, that's a good call for action. So, I'm sure you've heard this a million times before. Customers say, look, we're data rich, but information poor.

Lisa: Yeah, sometimes called DRIP. 

Natan: DRIP, yeah, exactly. And I found that a lot of it, you know, has to do with how do you get data into whatever you're using. Okay. Say for example, Seeq or other BI's like tool to figure out what you need to figure out. What are the advice you can give to organizations who are trying to get out of DRIP and to get this first, you know, layer of data kind of set up?

Because it's never one source and it's never one asset and it's never just the historian and it's never just the MES, you know, and the list is very long and complicated, right? How do you approach this and how do you guide customers today as The landscape is actually shifting because it's not the same conversation like five years ago, certainly not like what it was 10 years ago when it was like, oh, we got a couple servers and a few Excel files and an SQL.

How do you guys approach this? 

Lisa: As you asked that question, what I started thinking about is what's the reality of managing the massive amounts of data. 

Natan: Right, right. 

Lisa: I mean, 25, 30 billion just in infrastructure alone on sensors and just imagining the millions of data points every minute, right? Per line, per process, per plant, whatever you want to look at.

And the reality of companies that are routinely divesting and reacquiring, to me, it really sets the stage for this. There's going to be data in lots of places all the time. And an effective IT strategy that is essentially taking that into consideration such that the end state isn't the, hey, wait a second, operational folks, you'll get the data when we all have it in one place.

Here's our IT strategy to continually evolve, making sure this data is safe and secure and accessible. Recognizing that there's not an end state for that. Seeq was intentionally designed to connect to data where it lives today, because there's immediate value that we need to be getting out of it, there's not time to waste.

To actually adapt our processes to be especially sustainable and flexing them with the IT strategy. Those, those experts, you know, Hey, I want to move this information here because this is where, you know, security and, and taking advantage of the cloud and all those other technologies that are coming along. So we were designed to be very Malleable, let's say, and work with both the needs of operational technology people today, and then be a key part of the architecture strategy that the people responsible for keeping that data secure can depend on.

Natan: I find that pretty unique because you're basically crossing between the IT people and the OT people. They kind of need to see where you fit in this architecture, but I guess your users now also span both sides, right? Maybe not necessarily pure IT, but people who are more on the back end of the operation as opposed to just on the front line.

Lisa: That's right. And what I'd say is while OSIsoft's PI historian was quite popular and we see it nearly everywhere, the reality is, you know, we have a lot of customers and the majority of customers have 10, 20, 30 unique data connections. And there's multiple of those connections. So it's not uncommon for us to have customers with 200 different sources.

Now, a third of those might be PI the rest are something else that provides context. You know, really for this democratization to really make change, we have to provide the best experience for users. And so there's this growing. I don't like to think of it as pressure. I like to think of it as an opportunity for IT and OT worlds to collaborate on how they get value today.

Yeah. And I think, you know, since you mentioned OSI PI who's obviously been a key player in the historian space, it really evokes, you know, a broader discussion that I think we touched on briefly. What is the historian today? You know, I mean, I'm really interested in this perspective. Let me set it up for a second, because traditionally historians have been filling up very specific role in the traditional operational sense.

Natan: But like you're saying, now there's like 20 different connections. And I feel like the historian is, uh, is a term used, but like in the real world implementation, now elements of the data or all sorts of islands of data exist in various System, places, clouds, lake, oceans, you name it. And so the, the real sort of full historian, I don't know how to call it, the all encompassing historian is kind of in obviously more than one system, but it's like, it becomes like the historian is almost a process, kind of like the way you think about information security.

So like, when is it done? Never. Because there's always More vulnerabilities you have to fix, more compliance you have to address, you know, et cetera. So is data historians becoming a process and an architecture than a product? How do you think about that? Or, or am I kind of imagining stuff and just in my Natan brain of not accepting the status quo?

Natan, I think what it comes down to is how do we envision how action gets taken and then what feeds in? So I believe the PI historian and SAP and LIMS systems and all of that, what we should be thinking about is in order to make good decisions on changing a process, that involves not only the time series data stored in a traditional historian, but also the additional context needed.

Totally. And you know, this is, this is, now that you're saying it like that, I didn't think about it like that, but maybe I'll try this on you. It's almost like that the emphasis becomes like, what is the tool I'm using to access the data which becomes the historian as opposed to like what we used to call historian in the past?

Lisa: If we come at what the problem is that we're trying to solve or the success we're trying to create, put it positively, then we'll say, what are the pieces that we need from that? And then where does the cloud play in? Where does this historian play? Where does a combination of, say, a couple of us in the ecosystem working together deliver a comprehensive solution for somebody?

Natan: Yeah, so this is back to what we said we'll circle back to in Operations Calling when, you know, we were launching our ecosystem and it was very exciting to see everybody here and we had a discussion. How do we embrace openness and interoperability and kind of open up the tech stack such that it creates value to the customer?

How do you see that in action? Do you have some examples or strategies that you would suggest customers and partners adopt to make the best out of what the ecosystem can offer? 

Lisa: To some extent, making sure we're starting with what, what is the outcome that they are trying to achieve, as opposed to presupposing which solutions fit that outcome.

I believe Tulip and Seeq and, and many other companies are in a better position to be responsive if we understand truly what it is that's trying to be developed or implemented. And then because of partnerships, uh, that we're growing, like the fact, you know, we're working together now more, we can show up and say, all right, you're trying to do that. Here's how-

Natan: Here's pieces of the solution coming together. 

Lisa: That's right. We should not leave it as a burden to the customers, the companies out there to try to sort it out and put it together. We can do a better job in 24 and 25 partnering both with, I'll call it more of the the historical vendors that have been in the space and very effective for years in combination with many of us that are a little bit newer on the scene in the last 10 years. And really there's a lot of, a lot of very good synergy. Even though I usually hate that word. 

Natan: It comes with a lot of baggage, but like in a way, you know, it's accurate because, you know, preparing for this, I was kind of browsing on the, the Seeq add on gallery and kind of seeing all the cool stuff that is there.

What's your favorite example of something that you've done with, um, with a partner, uh, just to give an example of the synergy that it's not just a word. It's actually, Hey, this is, this is creating real value to the customer, moving, moving it to the hands of the people doing the work. And otherwise it's, if you don't have open system and galleries like that, Tulip library, for example, on our end, you just can't bring it to the customer. So what's a good example from your perspective? 

Lisa: Well, in addition to our work with Tulip, and I love the fact that you and I got connected and three weeks later, we're at your event showing we're working together, okay, well, that's Agile. That shows we're fast. And it was all people, operators, engineers, managers. I'll say in addition to that. I'll call out Databricks as one of our fastest growing partnerships in this new partnering model. Yeah. And put very specifically, Seeq brings the OT and operational context, so we know the people, the data, the problem set. The path to a faster ROI. And then Databricks is really bringing that IT and data science context, scalable platform, ML workflows, on demand pricing.

I mean, as I think about that, and then there's a customer, you know, that we put out with our press release that says, Hey, here's how I see these technologies. 

Natan: This combination. 

Lisa: Yeah. Right. So I think we should work together more with those people that are leading in our, in our joint customer base. Who see the biggest thing they're trying to do and we can lean in on that and ultimately have more impact, frankly, a year from now.

Natan: Yeah, I think that's a really good thing to tie in this ecosystem point. It takes time to build good things, including the ecosystem. So while it's moving fast, you know, getting the right stuff in there takes time, so I agree. You know, as we're kind of coming up on time here, and we're sitting at the end of the year that would likely go down in history, for better or worse, as the year of the generative AI.

And so no conversation is complete without it. And I think with, um, strong emphasis on data, its importance, uh, making it accessible to everyone, as we've discussed here today, we should tackle this, this topic a little bit. And I'm kind of curious about two things. One is what are the misconceptions that you're seeing about the tech?

Because, you know, we, we all are very well exposed to the hype and trying to figure it out. Of course, we're building product in this space as well, like everybody, and uh, trying to figure out like what it would mean to customers and partners. And then how do you think customers will actually get value out of generative AI specifically when you compare it to the more traditional machine learning, deep learning techniques, and why is that with a perspective and your command of the data space?

Lisa: To start, I'll step back a little and just say in the next years, especially in the next 24 months, the role of analytics is going to continue to become even more mission critical. So as we think about machine learning or generative AI or everything else that might be coming, it comes back to the role of analytics as the data driven decisions continue to drive gains in productivity and sustainability.

So we're going to keep coming back to The fact that analytics will even be a stronger rule. And so when I think about your question about the hype and some of the misconceptions, Gen AI promises the potential for significant improvement in the future, but it is not magic. Gen AI is not magic. And I'll even put it in quotes.

Organizations must acknowledge its limitations and associated risks, including data challenges, a lack of transparency and data privacy concerns. And Gen AI results need to be validated. Gen AI is only as good as the data and models that are used. And as far as misconceptions, despite popular discourse, Gen AI requires human oversight to function effectively, doesn't replace the need for domain experts.

But instead, it complements their expertise and Seeq was found on the premise that there is a significant amount of data and information that's not captured anywhere in the people that are engaging on the process. So we're a long way from fully accurate, I'll call it models that live outside of human interaction or control.

And then you asked me, do I think in the future consumers will get more value out of generative AI? 

Natan: Yeah, and comparing it to machine learning. But, you know, quick remark before we get into that. First of all, I totally agree. And it's kind of like humans are still needed. That's the key point. And, and it is, you know, pretty exciting and to see what, what it can do, but kind of standing on its own, you know.

We already kind of know it's insufficient. So what I'm excited about in this is like building it into tools and contextualizing it to people who can say like, Hey, you know, I built, we have our frontline copilot. You can ask it questions like compare line three and line five and which one is faster and things like that.

And it mostly will get it right. And because we have structured data and it's like generating SQL underneath and it will. Generate like a chart so people can actually build things using natural language, which is pretty cool. I mean, I started playing around with it. I also think it will come to any BI near you any moment now, if not already.

And like our main UI/UX modality to deal with BI tools will change over time and it's still human required. So I, you know, anything that has human in it, um. I'm excited. But yeah, but let's end on this note. It's like, how do you think it will compare to the more traditional sense of machine learning and, and how would it impact the value, uh, manufacturing organization can get out of it?

Lisa: Right now the world's pretty enamored with Gen AI. 

Natan: Yeah. Honeymoon? 

Lisa: I will say just doing a quick, I need to leave a review of a, of a VRBO and I used it and it was the best review ever written. So, you know, I see the fun, right? I see where it's like, Oh, wow, I get to focus on what's most important. I have another tool in my toolbox. So I think Gen AI is already bringing amazing human to machine interface possibilities, that human to machine interface. And traditional machine learning, it still has a role. It's continuing to demonstrate amazing business value, including in time series and analytics. So if we think about AI and we define AI, the perspective that I have and I share with many of my colleagues is it's about advanced analytics, machine learning, and generative AI and the idea that, for consumers anyway, the emergence of this is even more revolutionary because it's yet another tool in the toolbox. That humans can use and it's useful to everyone. So you're right. I mean, it's going to be pretty much, I think, in every technology near you.

Natan: Yeah. 

Lisa: And I think that's exciting. And I'm excited that Seeq is a part of that and is a leader in this. And we're excited to see Tulip doing the same thing. And many of the others are like, Hey, let's embrace it. And then back to your comment earlier, how do we show up together holistically with an even stronger, better solution so these companies can really make progress in stuff we know they need them to do?

Natan: Yeah, totally. It'll be interesting to see how it unfolds because, you know, we, when we talk about integrating products, it's like, oh, we talk about API and REST and, you know, this, that, the other, all sorts of things like protocols, stuff that makes products talk to each other, right? And it wasn't so simple 15 years ago.

And that's like the maturity of internet and open source and blah, blah, blah. But now it may be that we will have integration bots that know how to talk to each other and we just let them talk and they like recommend to us. And then the engineers go and figure out the integration. You know, who knows? I mean, maybe not so crazy, huh?

Lisa: Not so crazy at all. I mean, if we think about it right now, what we're doing is. If you think back 20, 30 years ago, we'd go to textbooks and libraries to find ideas for what to do differently. So that information exists. What I think about right now is we get done with a bunch of, call it, intelligent alerts. It's not a big step to say, hey, here's some ideas of what you want to focus on first, and then here's- 

Natan: Yeah, next best step. Yeah. 

Lisa: That's it. Like, that will be available as we turn the corner into 2024. What I'm most excited about is seeing organizations making tangible action already without it. So imagine, right, I'll take AllNex developing new methods of total steam demand modeling.

That's significant environmental impacts reduced. It's equivalent to like a thousand vehicles from the road per year for one application. So I see companies like Allnex or Syngenta or others that with what we are all providing them in the broader software ecosystem is already making significant impact.

Imagine if now we leverage all that extra knowledge and like you said, there's these bots that give suggestions. Essentially upskilling workforce at the day. I mean, if we think about another major change that's happening and driving a need to engage in this, it's Realities of our workforce transitions that are happening over the next 10, 20 years.

Natan: Yeah, and you know, I don't think we have it figured out at Tulip, just to be clear, but at least we're trying, and I believe it's a joint mission, which is operation folks have been so deprived from the past, whatever we call the digital transformation revolution, you know, industry 2.0. We can't afford them losing the, um, sort of the, whatever become post honeymoon.

So like productive, real change in the UI/UX landscape that makes what you defined as the human machine collaboration completely different. So they have to be included. And I think the work you're doing is awesome and will help usher that. And so I, I can't wait to see how you will use it in your product.

And, um, I sure hope that Tulip can catch a bunch of the alerts that you're getting from all this critical data, so we can trigger some good apps and work instruction workflows to help people get stuff done. 

Lisa: Absolutely. 

Natan: And thank you so much for coming on Augmented Ops, and I hope to have you again soon. So thanks, Lisa. 

Lisa: Yeah, thanks, Natan. It was fun. Until next time. 

Natan: Take care.