Ken: This is Ken Forster, Executive Director at Momenta. Welcome to our Digital Thread podcast produced by, for, and about digital industry leaders. In this series of conversations, we capture insights from the best and brightest minds in the digital industry- their executives, entrepreneurs, advisors, and other thought leaders. What they have in common is like our team at Momenta; they are deep industry operators. We hope you find these podcasts informative, and as always, we welcome your comments and suggestions. Good day, and welcome to episode 198 of our Momenta Digital Thread podcast series. Today, I'm pleased to host Michael Risse, former Chief Strategy Officer at Seeq. Michael's focus is big data technologies and analytics. For over 25 years, he has worked as a consultant, advisor, speaker, and founding partner at Seeq Corporation, transforming industrial process data into actionable intelligence. Before his recent work with startups and big data, he achieved successful results in a series of product and sales leadership roles at Microsoft, contributing over 13 billion in revenue, as well as incubation efforts for Microsoft Visual Basic, Office, BizTalk Server, and the overall Server & Tools Business Unit. Michael, welcome to our Digital Thread podcast today.
Michael: Thank you, Ken. It's great to be here.
Ken: Great to have you. It's already warming my heart because I was an early user of Microsoft Visual Basic. You talked about early in the day when your plan to do Windows UI was difficult. It was a great tool. We call this the Digital Thread podcast, which talks about one's individual digital industry journey that has led them to where they are now. What would you consider to be your digital thread?
Michael: My digital thread is software. I remember what typewriters and White Out were like; I remember the first time I saw a spreadsheet and the first time I saw a browser. The first time I saw Visual Basic - what you just mentioned. My passion and focus have been on software - enabling, empowering, inspiring, innovative software that helps people make better decisions and improve business outcomes. In a recent podcast with you, Jane Arnold talked about the importance of empowering the frontline, the operators, the engineer, and bottoms-up transformation. That’s what I am talking about. Software has been my focus before Microsoft, at Microsoft, obviously, at startups, and then with Seeq. At Seeq, for example, the NPS and CSAT numbers for Seeq are outstanding. Seeq customers love using Seeq, and the level of enablement and empowerment it provides it's just phenomenal. It gets them promoted and makes them successful.
Software is the root of my digital thread and continues to be my enthusiasm, including software business models. We are the heirs of a publishing business model that goes back 600 years. Books, magazines, newspapers, other publications, records, and now software. Trying to be a student of the business and licensing models for how intellectual property gets monetized through various means to take advantage of the fact the marginal cost of the next unit is near zero; that's the publishing business. The publishing and business models for software, and then the power and ability of software to transform people's efforts, have been the center of gravity for my efforts in the last 25 years.
Ken: You have certainly been a veteran. As I've calculated, you started your tenure at Seeq with about 20 years of experience at Microsoft, leading key developer and enterprise offerings, as we discussed. How did this time prepare you to play a leading role in industrial analytics at Seeq?
Michael: Ken, that would be a bimodal answer because one side prepared me for Seeq because of all the version 1.0 products I worked on. For some reason, that's what attracted me; what was interesting was the first version of Visual Basic, codenamed Thunder. The first version of Visual Studio, the first version of an integrated applications suite, Office. The first version of Biztalk, the first Microsoft Enterprise offerings in the server domain. That heavy lifting and traction and the determination to bring a version one product to market with all the decisions, all the required clarity, and communications- were important and transferable to the Seeq experience
The other thing that was applicable was all the learning from the different audiences, business models, licensing types, and routes to market that Microsoft provided. Thus, Microsoft was a portfolio of different experiences in one company. Audience by audience or product by product, influencer by influencer, etc. That is very helpful training for separating ourselves from a market and seeing it at arm's length; who is the audience? Who is the influencer? What are the routes to market? What are the right messages? What's the market framework? For all of those things, Microsoft provided an outstanding training ground.
The gap, the other side of the bimodal distribution, was on the industrial part, and with all apologies to any Microsoft employees listening, this was long ago. But no. Microsoft's focus was IT products and audiences- a network was a network, a database was a database, and an email server was an email server for whatever audience. When we would sometimes get questions about the manufacturing space, we’d smile at the customer and say, "Let me find a partner who can help you with that." Then we’d leave in case there was another question. There was just no manufacturing fluency in the organization during those years, the 1990s.
And compare that to today: Microsoft and Amazon are investing massively; they have multiple vertical manufacturing organizations, oil and gas, energy, pharmaceuticals, healthcare, etc. It's very different today. But back at then in the 1990s, no. The learning, the abilities, and the fluency I now have in this manufacturing and industrial space that's all been on the job with Seeq.
Ken: I'll call it a side note; I turned down a job offer from Microsoft in the early '90s. It was to be what they considered to be the technology evangelist for WinSEM, which was Windows Science, Engineering, and Manufacturing. My primary reason for turning it down is that I saw nothing of manufacturing at Microsoft at that very early stage, which led me to believe that they would do much there beyond HMIs, what Wonderware was doing at the time. I can relate very much to your topics there.
Michael: That may not have been the right decision from a stock price perspective.
Ken: Mike, I've done those calculations more times than you can count. But yeah, that happenstance in life, we always tend to think it would have been better than it does. I convinced myself that I would have stepped off the bus the first day on campus and gotten hit by a car or something. That's equally plausible, so that helps offset my anxiety about stock options. You joined Seeq in August 2013 as a founding partner. Tell us a bit about the origin story of the company.
Michael: In 2013 ten people were around a picnic table. Well, how did they get there? About half of them were developers. They were the start of a world-class software development organization and had worked with the CEO at his previous company.
The CEO, Steve, had an entrepreneurial background, saw an opportunity in industrial analytics, and was excited about working in this space. He was joined by the Chief Technical Officer, Brian, who had a background at Honeywell and in startups. At Brian’s latest startup, they needed remote analytics, diagnostics, and monitoring for a particular asset class. When Brian first met Steve, they talked about how hard this was, why it was so hard, and why it could be easier given all of the innovation in computer science, data management, big data, etc.
These two – CEO and CTO - were joined by John Peterson, a longtime OSIsoft executive with a sense of the market and the opportunity to work with the data storage systems platforms.
Then my contribution was covered a moment ago- let's bring an IT software approach to this OT space and span verticals. Let's span different vendors, let's span geographies, let's get scale through partners, and let's work with the cloud vendors as a route to market.
That was the group in 2013. In the fall of 2015, Seeq released the first version, which did not have a product-market fit (laughs). 2016 was chasing feature sufficiency and adding capabilities. I tell you; we turned the corner in 2017, and the world just changed. It was just different Seeq was important; Seeq was something that got reactions. It's been a rocket sled from 2017 to the present.
Ken: Well, you played your part certainly in helping to steer that sled. You led key roles, including Chief Marketing Officer and Chief Strategy Officer, going from what I consider demand creation to demand anticipation. This is unique. You mentioned your background in IT and Microsoft particularly, so it's almost apropos to what you would have done before. However, jumping from one to the other is still a unique cross-functional skill. What enabled you to execute both roles successfully?
Michael: I think the variety of experiences mentioned earlier. I have familiarity with market analysis and competitive analysis. I have familiarity with demand generation and ran the mid-market business for the Microsoft US subsidiary. Just a variety of experiences, and we needed all of them done. So, it was one person, me, until it was two people because I could hire somebody to outsource some of the work, accomplish something else, and then hire somebody to be that part of the team. So, one person became two, became three, became four, etc. But as an example, my first meeting with Microsoft was in June 2016, and we announced Shell as the first SaaS customer on Azure in September 2018. That was two years of effort, or as I call it, kissing frogs. Just trying to make something happen, trying to make one of them turn into a princess.
We finally hired somebody to lead our cloud integration efforts in March 2019. Almost three years of elbow grease and frog-kissing before we had proven the need for the resource. That, by the way, was a Microsoft approach to headcount. You don't hire in anticipation of what you think you might need; you hire when you've proven the need and got the right person, and then you scale the organization. It's a conservative approach to hiring and resourcing based on what's proven when you're overwhelmed, as opposed to, "You know, I think this might be interesting. Let's hire somebody to look at it." That was not our model, not my model, not my history.
Ken: What are you most proud of at Seeq?
Michael: Well, to start from nothing, literally nothing- 10 people from different backgrounds around a picnic table, to identify a market opportunity- to deliver software, to learn from that software, and then to succeed with it as a business, that’s something… From a startup perspective, if you look at the numbers- and it depends on the cohort, but if you look at the number of companies that get through Series A, B, and C funding rounds, which is what Seeq did raising a total of about $115 million, only 6 to 7% of startups make it there. We got there from scratch to a going business, a going concern, with hundreds of customers and employees. That's what we did.
And the how is equally important: taking care of customers. If you care of customers, the business will take care of itself. You don't win by watching the scoreboard; you win by playing the game with all the little details in execution and effort and by taking care of customers by getting a product out, getting painful feedback, taking the feedback and improving, and iterating. Very quickly, Seeq ships a new version about once a quarter. That led to business success, which led Seeq to where it is today. What we did, we achieved something special. Something that very few startups do. How did we do it? Again, I mentioned the Seeq NPS and CSAT scores: by listening, taking care of customers, a tight loop on feedback, enabling them to be successful.
Ken: Seeq represents industrial analytics, as you said earlier. What do you consider to be state-of-the-art in industrial analytics? Who do you see as some of the best practitioners of it?
Michael: For industrial analytics, one thing I repeatedly saw from a customer perspective, inventor perspective, market perspective- is there are four important audiences in the market, four end audiences and product sets that need to be considered or included in the customer industrial analytics journey or roadmap.
One is certainly machine learning and data science. Whether it's the big enterprise ML Ops tools, like SageMaker, Azure ML, or open-source Anaconda, or one of the startups- Raven.AI, FERO Labs, or Quartic.ai, there is going to be a machine learning component to analytics. There is no argument; there will be a machine learning component to improve the types and accuracy of analytics in manufacturing datasets. There has to be the ML checkbox; how do you enable that audience?
Second, how do you enable self-service process engineers? Excel is 30 years old and was never meant for this type of work; there's got to be better offerings. Seeq, Trend miner, that's the self-service, advanced analytics component for process engineers in production plants.
The third audience is more corporate based: the business analysts that want to use Power BI, Spotfire, and Tableau. They need their data very right. It's got to be structured. It’s got to be assembled. It's got to be contextualized. It's got to be in rows and columns. It's got to fit those tools, so it needs to be assembled and made ready for those tools.
Then the fourth category- maybe that's the kitchen sink bucket that says 'other.' That's where your statistics packages, OEE metrics, other custom IT apps, and dashboarding apps come into play.
So state-of-the-art industrial analytics is two things. One, it's recognizing those four categories and audiences and ensuring they're all on your roadmap for supporting the different roles and requirements.
Then there's the second part, which is okay; now we have to read the data. From a market perspective, one of the biggest, most interesting, 10-billion-dollar conversations is, how do you get that data ready for use? This is where you hear about industrial data models or industrial data layers, or you look at the work of a Cognite, or maybe an Element, an Imation, which Emerson acquired- to get the data aggregation and contextualization and access to enable those four user categories and audiences.
State of the art analytics recognizes the four audiences and supports them all, including ad hoc interaction, and surfacing and rationalizing, organizing the data to make it accessible for those four audiences.
Ken: Well, I appreciate that. It's interesting as our conversation has stepped through, you almost seem to be a bit of an archetype in that your IT meets OT personally, and the influence you brought from Microsoft and your impact now that you've had on OT. It's interesting because one can't go too far in operational technology, or OT these days, without considering the impact of the so-called hyperscalers. Azure, Microsoft, AWS. You and I know Andrew Obin from BofA, and he just completed their industry forecast at the end of the year. He listed one of the five key trends for OT being the impact of the hyperscalers effectively. You've been in both worlds and brought one world to the other. What impact do you see coming in terms of this?
Michael: I like that Andrew doesn't use the word hyperscalers because I don't like the term for a couple of reasons.
If you and I were analysts- and analysts profit on confusion- so if we were analysts, we would use the word hyperscalers as much as we could. People would have to ask us what we meant, which would benefit our business. But as vendors, as investors, the market is clear. AWS and Azure, Microsoft, and Amazon; there is no third place beyond those two. It is an Azure and AWS world, Andrew calls those two out specifically in his report, and it is in our interest to we recognize in open software markets there is consolidation because of the network effect, economies of scale, and price advantage So there's a first place, there might be a second place, but there's no third. Today's most interesting business competition is Azure versus AWS and who wins the marketplace battle. It's not Burger King and McDonald's; it's not Coke or Pepsi. It's AWS and Azure.
The importance of starting any conversation about what AWS is doing and what Azure is doing is key. If you're a customer, which are you leading with, if you're a vendor, which one will you go to market with?
But they are dominant which leads to the growth rates Andrew is talking about in his report as they continue to succeed and aggregate more of the market. That's the first point.
The second point is how they will win; it's because of how they sell; I call this the dark arts of the cloud. Both AWS and Azure incentivize customers to buy more from them as a single source through their MACC and EDP agreements. Those agreements enable IT, procurement and others to get better discounts the more they buy, which means they will be more involved in every software decision and discussion. Which means IT will gain purchasing authority through these arrangements over OT purchasing. The (IT) pendulum swings back and forth, but right now, it's swinging to IT in terms of the authority and the ability to make those purchase decisions, which will impact downstream what people can buy.
Ken: Thanks for the clarification on hyperscale; you're right. He calls this the Cloud vendors, so I'll remember that going into the future because I don't get paid for mentions of hyperscale. But what you just said is interesting. One of the other trends he's listed in these top five trends is what he calls the SaaSification of OT vendors. He says that's the industrial software turning SaaSy; it's his term there. But it's interesting because you mentioned the growth number. He's saying industrial software will likely exceed 250 billion annually by 2027- a representative CAGR of 15% plus. Given everything you've just said and mentioned which way the pendulum swings, if you're a betting man, who do you see as the winners and losers in this expansion?
Michael: Andrew mentioned some workloads specifically in his report; what did he say? First, data, data analytics and AI. The Cloud is where manufacturing data is going to go. I would love to do a podcast with you on the ten reasons why that's true. But it will go to the Cloud for analytics, data storage, machine learning, and otherwise. Who are the winners and losers? As the report says, the competition will increase for incumbent automation vendors. The barriers to entry that have existed in the past for years to protect the automation vendors, the proprietary systems, and the hardware-centric and software-attached to hardware sales, the Cloud will bulldoze them. When that data gets to the Cloud, it gets opened. It gets opened to other vendors, it gets opened to other analytics opportunities, and so there will be pressure on the automation vendors to figure out their value propositions relative to what AWS and Azure want. What AWS and Azure want- they want the data, data has gravity, and data attracts other high-value services. Do you want to be in the data storage business and compete with the Azures and AWSs of the world or double down on the high-value applications and insights you can enable specific to your verticals on top of those services? That's a rethink for the automation vendors. It may be overdue, but it's certainly coming as more data gets to the Cloud and becomes more obvious there's a new industrial analytics and insights model.
Ken: We have to do this podcast you mentioned because this is a topic that we could spend a lot of time on. But if you look at Schneider Electric and Aviva, Emerson and Aspentech, or Rockwell and Flex/ PTC. Certainly, all of them have software moves afoot, especially with Emerson's recent Pure Play moves in the space. Andrew says in his report that it will be a very interesting next five years. For the astute listeners- they've heard Seeq as present and past tense. Just for the record, I know that you recently left Seeq. What's next for you?
Michael: It'll be software, I can promise that. That's been my digital thread to date and will be my digital thread going forward. Based on experience, it will be around incubation, version 1.0s, and getting something started related to the intersection of the Cloud and analytics.
I think about this as the Purdue model needs to split. The Purdue model- so level one and level two, great that stays on-prem, but then the data splits. What data needs to be real-time and stay local versus what needs to go to the Cloud as the natural storage space. Then when the data gets to the Cloud, it needs to split again. What data do you need short-term for statistics and analytics reports? Maybe that's days, weeks, months, vs what, can you store because you'll want it at some point, but you want to store it cheaply, so you want that off into cold storage.
What does this new model look like? What does the new Purdue model look like where there are a couple of forks between what stays on-prem and what goes to the Cloud? And then once it’s in the Cloud, what’s needed in the short term versus what's not? How does that become an easy-to-manage and operate experience for customers? That's just a super interesting question.
Ken: It goes hand-in-hand with these top reasons for the Cloud and IoT. We'll look forward to putting that on the agenda in our podcasts, and maybe we can even convince you to do a white paper there along the way. The final question I always like to ask is, where do you find your inspiration? Any good books you're reading lately, articles, podcasts, etc.?
Michael: One, I am impressed, and have been for years, with the writing and insights that McKinsey provides. My journey in Big Data was started in May 2011 by what? The McKinsey report on Big Data and analytics. That's a steady source of quotes and insights. They continue to be a source of inspiration and opportunity for an analytics and advantage perspective in vertical markets, and I appreciate the folks there and the writing they do online when I am looking for information and updates.
Then on the personal side, the thing about scaling businesses is you're really talking about an idea in your head and success means you are getting that idea into other people's heads and then through them into other people's heads. Sometimes directly in one-on-one discussions or indirectly through the web, writings, or releases.
The result is thinking a lot about communications. I talk about thinking through words at the syllable level and precision because once the words leave you, they have to be understood by your audience and then understood by that audience's audience to scale businesses and impact. Again, we're talking about a global footprint that Seeq has right now.
I read a lot about writing. I just finished a fantastic book about the history behind "The Sun Also Rises" and I'm reading a book right now which is an analysis of Russian short stories. It's just pounding on the syllables, the words, and the sentence structure; how do you communicate effectively? If you're going to grow something or incubate a business; it's about getting an understanding in lots of other people's heads quickly. Whether it's Ted Talks and how to present or it's writing, the importance of communication success- and communication is the responsibility of the sender, not the receiver- is paramount. - How do I communicate more effectively to get the kind of reach and impact that I'm looking for?
So, McKinsey on the analytics and manufacturing side and books on writing on personal.
Ken: I love it. That's quite an interesting topic overall that you're discussing, and it also explains the concern over the hyperscalers comment earlier.
Michael: Pick your words with care.
Ken: Alright, especially your TLA is along the way, too. Michael, thank you for sharing this time and insights with us today. It's been great.
Michael: Very much appreciated. Thank you for the opportunity, Ken.
Ken: Thank you for taking the time. This has been Michael Risse, former Chief Strategy Officer at Seeq and soon to appear in probably an exciting role; I have to imagine. We'll stay tuned there. Thank you for listening, and please join us for the next episode of our Digital Thread podcast series. Thank you, and have a great day. You've been listening to the Momenta Digital Thread podcast series. We hope you've enjoyed the discussion, and as always, we welcome your comments and suggestions. Please check our website at momenta.one for archived versions of podcasts, as well as resources to help with your digital industry journey. Thank you for listening.
Connect with Michael Risse
What inspires me?
As an evangelist for analytics, I am inspired by stories of data and insights in any field. Great reads include books about the use of data in sports (Moneyball, Michael Lewis), dating (Dataclysm, Christian Rudder), and populations (Everyone Lies… , Seth Stephens-Davidowitz). For something shorter, the best magazine articles that describe the intersection of data volumes, algorithms, and insights are “The Petabyte Age” in Wired (June 23, 2008) and “The Perfect Milk Machine” in The Atlantic (May 1, 2012). Although both are over ten years old, they are approachable to anyone and clearly articulate the benefits and opportunities in analytics.