PTC Inc. (PTC) Earnings Call Transcript & Summary

We certainly hope this finds you well in having a wonderful Friday thus far. I am Lauren with the National Association of Manufacturers and the Manufacturing Leadership Council, and I am delighted to welcome you to our MasterClass series tech talk, delivering 5 years of value in 12 months. digital continuous improvement at enterprise scale. Now before we begin, just a few housekeeping items. [Operator Instructions] We will do our best to address your questions at the end of the session. So now let's get into our content. Please welcome our end moderator, Mr. David Brousell, Manufacturing Leadership Council's co-founder. David, over to you.

Thanks, Lauren, and for that nice introduction. And welcome, everyone, to our MLC Tech Talk for today. I hope you've been having a great week. And I'm sure, today, we'll put a capstone on the week for you because we have a great topic to discuss with you. But before I get into that, let me welcome our panelists today, Craig Melrose from PTC. Craig leads PTC's thinking on digital transformation and manufacturing. And I'm sure, as many of you know, PTC is a longtime member of the Manufacturing Leadership Council. So welcome, Craig. Great to have you with us today. We've got a really interesting subject, as I mentioned. It's how to achieve continuous improvements in operations at enterprise scale faster than ever before. It is one that will be of great interest to any manufacturing operations executive who was trying to drive value from digital investments in their plant floor and factory operations and better align their operational disciplines. I'm sure that's music to the ears of our attendees. Everybody is trying to optimize their M4.0 investments.

So let's get started. Craig, my first question to you is that MLC Research, which you know we do quite a bit of has clearly shown that manufacturers are accelerating their investments in digital technologies and digital transformation, in part spurred by the pandemic putting a premium on greater agility and flexibility to deal with disruption but also as part of a long-term transition to the digital model of production. However, traditional improvement and efficiency discipline, such as continuous improvement and operational excellence, haven't really kept up and are still largely manual. So my first question to you is why has this happened in your view?

Sure. Sure. Absolutely, David. And David, always great to see you. Thank you for having me. Always great to be a part of Manufacturing Leadership Council. And like you said, exciting topic, we'll dive into a lot of it. But to your first question, my experience over 30 years across almost every vertical industry on the face of the earth and driving improvement in all of those industries. This is similar across all those industries. And I think part of the reason and rationale behind it is everyone has a different starting point, either a different starting point because of the cycle times of their products or the skill level of their workforce or the customer requirement and customer demand. But certainly, once you get below that, it's also different in the sense that some people have 30-year-old equipment, 10-year-old equipment, brand-new equipment. Some people are using MES systems. Some people are not. ERP has been around for quite some time, and everybody has evolved that or ERP systems. And so whatever the different data systems are, everyone is trying to understand how do I just deliver today's performance. And not all of that information is easy to get a hold of, to get your arms around or there's so much data. Nobody knows what are the few pieces to pull out to give me the one insight that says, go do x and...

Lots of data but few insights.

Very few or almost no insights to be provocative. And so I think everyone is looking at that and saying, I don't quite know how to get my arms around it. I'm just going to circumvent the entire thing and do manual exactly what I need and use whiteboards and checklists and audits and what have you. And that's a great Band-Aid, but it's not a cure. And I truly think digital can be a cure.

Yes. So how do we apply continuous improvement techniques to accelerate continuous improvement, if I can put it that way in a better way?

Yes. And I know we'll get into it in this conversation. Part of what we've tried to do at PTC is create a system that is much closer to what we just described. What is it that you're solving for? For example, I know you know and an operator on any shop floor for everyone that's on this webinar today knows what an hour is. But I might look at an hour and say, hour is 60 parts for me. You might look at it and say an hour is 1 part for you, but it's still an hour, right? And so we've got this common denominator that we can use. And I think if you start to build the digital systems around some of these common denominators, stack the data such that it helps you to disaggregate or break down into bite-size pieces, what's inside that common denominator. You start having a much higher chance of success of then leveraging digital versus manual. And in that higher chance of success, it actually is very intuitive. There's no training required. I don't need to explain to you what an hour is or you don't have to explain to me. It's a bit like the smartphones in our pockets. Most people did not go to a class to learn how to use it. They just started learning because it's very intuitive. And how do you get digital to be that same level? And at PTC, we call that digital continuous improvement for continuous improvement. Continuous improvement is a great process, great process. Sometimes some companies can't see the forest through the trees because they're so into -- I have 1,000 little actions. Each action is worth $10,000. Collectively, I've got my $1 million a year target. I'm going to get $1 million every year. That's 1% of my cost, which was $100 million. And all of that is wonderful. What we're trying to do is change that and say, well, what is a one activity that might be worth $5 million. How do you get 5 years of impact in 1 year or in 12 months and really change it because you've got digital giving you those 3 or 4 big ideas versus hundreds or thousands of little ideas. Does that make sense to you?

Yes. No, I see what you -- where you're going with this. You mentioned before just a moment ago about all the data coming in, and the data volumes are growing and growing and growing. Our research here at MLC has shown that manufacturers in the next couple of years are expecting up to a 500% increase in beta volumes. So finding the nuggets and really focusing on the few things that can really drive substantial value versus many things that may drive less value is -- can be quite a challenge. So how should manufacturers tackle that focus and prioritization problem?

Sure. Sure. Absolutely. So I'll give an example. I spoke to a customer 1 time and they were saying that now that they have more data, they're measuring micro stoppages. And I was like, I understand what that is, but help me understand your definition of micro stoppage. And they were saying less than a second, hundredths of the second, tenths of a second. And I was like, okay, that's interesting. I could absolutely see how digital is allowing you to do that. And then I was saying, well, how many micro stoppages do you have in a week? And they're like, "Oh, thousands." I'm like, wow. Okay. That's good to know and good to measure, I guess. If you total up the full time lost of all those micro stoppages, what's it come to in a week, and they were like, minutes. And I'm like, why are you worrying about all the energy and effort into measuring thousands of microstoppages to understand 10 minutes of loss a week when I'm certain there's something you have that's maybe 3 hours or 6 hours or all of your changeovers, what's the variability between them. And it's a total of 10 hours to change over, and maybe 3 hours of that 10 hours is variable and could be compressed. I'd rather be going after 3 hours, 2 hours, 1 hour, not 10 minutes 1,000 times. And they were like, "Oh, I see your point." Like just because I can doesn't mean I should.

Yes, yes. And that's where a better strategy comes in better model or implementing that strategy. Yes, that's a key issue. We all get inundated with these things. But to focus down on the 2 or 3 things that will really enable you to improve things and execute better is the challenge for sure. Let me ask you another question now about this whole idea of continuous improvement and making it much better so that we can accelerate the delivery of value around it. Many manufacturers rely on operational equipment effectiveness. OEE is an important way to measure how operations are performing. But of course, it's not perfect. What do you think are OEE's limitations?

That's a loaded question, David. I will do... .

We specialize in them.

Yes, that's right. I may make a lot of friends. I may make a lot of enemies here in the next few moments. For me, my biggest concern with OEE over the last 30 years of doing this in every different vertical industry, the challenges around measurement and accuracy of measurement and standardization measurement. Now what do I mean by that? OEE, by definition, is percent-based. So it's a percent of something else. So it's not even a direct measurement. It's an indirect measurement or a secondary measurement. Then it's a secondary measurement of things like availability, performance, quality. So now I've got 3 secondary measurements, all in percent, then I'm going to multiply all of these percentages against each other to create a fourth percent. At that point, that's like an index. It's not even a real number. And so as a result of all of that, simplistically, OEE becomes a reporting tool, not a problem-solving tool. So it's Craig is a 65%. David is a 72%. Well instantly, everyone is saying, David's doing a good job; he's 72%, and Craig is doing a bad job, he's 65%. That may or may not be true. I might be including all planned maintenance. You might be excluding some of the planned maintenance because you're like it's planned. I knew I wasn't going to run during that time. I'm not going to penalize myself for it. So now even though we work at the same company, your measurement is different than my measurement, but you can't really know that by just looking at the percentages. Maybe we're measuring it identically the same. Your 72% is not meeting customer demand. My 65% is meeting customer demand. Suddenly you're the bad plant that needs to help and needs work. I am not the bad plant. But you wouldn't know that just by looking at the numbers because human nature, 72% is better than 65%. And so as a result, it really becomes complex to do the right problem solving, to understand is it the right level of performance or not, is everything included in the calculation or not. And so there's a lot of I think, frustration around that metric. Although it's a great metric, it's more the issues around how it's used and how it's measured, not that there's a problem with the metric itself. Does that help to better...

It almost sounds like you're saying it's good but not necessarily sufficient.

Yes, although what's happened to it over time has made it less and less valuable and yet it's the best thing we have. And so that's where our earlier conversation -- something like OEE, more intuitive, more helpful and -- to do so rather than keep doing it the same old way, expecting a different result.

Yes. So let me circle back on something you briefly mentioned before. You and several of your colleagues last year authored an article in our Manufacturing Leadership Journal that advocated that manufacturers should move away from percentage-based measurement models to a time-based model. Is that the better approach that you see here? And how do you see that? If you do that, how do you see that improving things?

No, absolutely. Absolutely. I think we have couple of graphics that will help us here. Here's an example of time-based to help everyone, right? This is just a typical week. Now this week could be 6 months' worth of data, David. But in that 6 months, take the normal performance and normalize it down to 1 week, 1 168-hour week. And in that week, this company is running 5 days, 2 shifts, 80 hours and 1 shift of over time on the weekend, 88 hours. And their output is 45 hours of productive time because they made 4,500 units products and they're basically making $100 an hour for simple math for the simple example here. 45 divided by 88, 50% OEE. So you still have that percent that you've always had because you don't want to go away from that metric that you've measured for decades now. But what's between the 45 and 88 is presented in a very different way and all the details below this, this is just the high level, are presented and nested in a very different way. So quickly, you look at this as a frontline operator and say, "Wow, we're losing almost 2 shifts to unplanned downtime, 1.5 shifts to change over, a little over a shift for speed loss. Wow. Let's go in and problem solve speed loss, let's say. And so very quickly, it turns into a problem-solving tool plus the reporting, you still have the 50%, when you go into the problem solving and say, let's try to improve speed loss. And so then on the next page, just assume that we were able to do this. The next page would show those 8 hours are improved out of speed loss, and we can turn that into revenue or we can turn that into cost. And so then the revenue example here would be those 8 hours turn into 800 more parts because we are making 100 an hour with 8 hours, 800 parts just again for simplicity. Everything else stays the same, 53 divided by 88, 60 or a 9 percentage point movement. Now also, we could have said the target was 5,500 units to meet customer demand, 6,000 units to meet customer demand. And you would have gone back and said, okay, how many additional hours do I need to meet that target number. Here, we just used 8 hours versus 800 units or 1,000 units or 1,200 units. But the problem solving target setting is very easy here. And then where am I going to capture that savings to achieve that target is very straightforward. Now they're detailed in the problem solving, absolutely. But here, we capture that. And again, this is just 1 week. So that 800 units in this example is probably worth thousands or hundreds of thousands of dollars, 50 weeks a year that ends up being millions, multiple lines in a site that ends up being multiple millions and then multiple sites across the network that ends up being tens of millions of dollars for a company, and we're seeing that. We're seeing that in our customers that are deploying like this. They're getting millions quickly, same year, and they're getting tens of millions across their network, and it's really changing the way that they're engaging with the data, and digital is allowing this to happen. This is just a revenue example. The next example is cost. We can take those same 8 hours and take out that overtime shift. Same type of math, 45 divided by 80 now instead of 88 because we took the overtime shift out, that happens to be a 56 or 5 percentage points. But again, it's tens of thousands or hundreds of thousands per week weeks in the year is millions, multiple lines, multiple millions, multiple sites, tens of millions. And so very quickly, that multiplier gets very large when this becomes your standard across all your lines and across all your sites in the network. And so I think that's why the time base helps more than the percent. You still get the reporting of the percent, but the time helps you to see and problem solve and establish targets and engage. Maintenance can look at that and say, "Wow, 14 hours of unplanned. I need to go look at that. operations can or a continuous improvement can look at changeovers and say, "Wow, 11, is that compressible. Operations can look at the speed loss and say, wow, is there a way to solve that." All different groups can engage on different parts of the data and problem solve together against one common target and one single source of truth. Does that help to clarify it a little bit better, David?

Yes, the numbers are very compelling. What's involved in a company making that switch to the time-based? Is it a steep climb to do that? Or is it relatively easy? What would you advise there?

Sure, sure. So I think, like anything, change management is the hardest part, right, right? The behavior and the mindset is the hardest part. Let's assume you know how to change mindsets and change behaviors. The rest of the changes are very straightforward, right? So this example and this approach, what we call it DPM, digital performance management, and is built on the back of ThingWorx, an IoT platform, so by definition, IoT is meant to grab data from any system and then contextualize that data. So it really doesn't matter your starting point. It doesn't matter what systems you're using. It doesn't matter where the data comes from. You can tag the data, ingest the data, contextualize the data and get it to say the right things, and that's weeks of effort. And then we've had customers go live and within weeks, literally 6, 8 weeks there are already 10, 15 percentage points of improvement worth millions, like what we just walked through. So that was a general example to help everyone here, but it's a real example in the sense of what we're seeing customers achieve in just a matter of weeks to get to that point. I'd also say we've built it were it to be 100% manual. So we've had customers run it in parallel to their existing systems just to test them side by side and do it manual, even though they're little bit extra ask of the operator for a short period of time to enter it in manually. But if you had a 50-year-old machining center that's not automated, you could leverage this manually and still have a standard answer across all of your assets. You could use it in a brand-new plant that's fully automated and it's 100% automatic data. You can use it in a 10-year-old plant that's maybe 80% automated and 20% -- entry on some of the items that aren't sensored or an instrument yet. And you may actually use it to say, wow, we should instrument this item because it's requiring a lot of manual entry, and we don't want manual entry anymore. We want to automate that data. So different companies are using it in a lot of ways, but there's a lot of flexibility to allow you to embrace it, deploy it with the added element of mindset and behavior change on change management. But I do think it's a major change. We've had customers look at it and say, We already have something that we love, and then they look at this and the operators are like, wow, I didn't even know that." And suddenly, they're like, "Wow, our existing system didn't measure that way or didn't make it that transparent." And even just that, Aha! I see something I haven't seen before, helped with some of the behavior change and some of the mindset change to say we want to adopt a different way.