Employee engagement: it’s a challenge that can either excite or frustrate a manager. In the fast-paced environment of a manufacturing facility, it can feel nigh impossible. We interviewed Martin Cloake, CEO of Raven AI, for his perspective on how managers can improve their employee engagement through technology. Read on, or listen to the entire interview on the Augmented Management Podcast.
Martin: Raven is a platform designed to support manufacturing facilities in their op-ex and continuous improvement goals. Raven does that by collecting data, clarifying it, and then driving action on the shop floor. With Raven, our clients are able to deliver on time, reduce unwanted downtime, and increase profitability.
Martin: When you don’t have technology, the easiest thing to do is to have one metric: profitability. And everybody looks at that one metric. But that’s not a fair metric, and that’s demotivating. When using technology to recognize performance, often technology is used to highlight poor performance in kind of in a sort of broader way. But often the keys to driving huge improvements are by looking for and identifying abnormally good performers.
Martin: When you look at both negative anomalies and positive anomalies, sometimes you find people who’ve just found a trick to do something really, really well. There’s a really neat story from one of our manufacturing clients where, for whatever reason, this one guy was doing set-ups in half the time.
This is a process to set up a big industrial machine, and he was way, way better at it. And so, when they went to go see what he was doing, they saw something super scary: this guy had taken this giant spool and had tipped it over on its side to be able to do two steps at the same time. Super dangerous, but kind of ingenious.
So the leadership at the plant was took the good parts of his process, and built him a jig that allowed him to tip over this big spool so that he could do this neat process in a safe way. And this is something that they actually rolled out to the other cells, other lines. This is just the perfect example of stuff that pops out in the data when you look at it in the right way. Whereas if you just look at the summary, in general, they weren’t doing that well at setups. This one guy was doing really well. And that’s a really exciting moment.
Martin: A supervisor might know that there’s problems in their data based on how their team responds to when they show their KPI dashboards. And really, usually data quality is not a surprise. When I worked in manufacturing, it was pretty clear when the standards were on or off, and at some points operators know right away if things aren’t aligned. There’s a lot of knowledge on the shop floor. So if the KPI becomes an inside joke on the shop floor, you know something’s off.
And I would say for the most part, for those of us who’ve spent time in manufacturing, we just take for granted that the data is maybe directionally okay, but in detail it’s wrong. And if on occasion, everybody gets to the point where they get so frustrated with data that they say, “All right, I’m going to get my Excel file and I’m going to book off a Sunday evening to dig into it,” and you find a whole bunch of mess.
And then maybe you tell somebody, but at some point there never is that time to go through everything. So I think it’s pretty universal that a lot of the data in manufacturing is just wrong. And I think manufacturers hope that it’s directionally correct so that they’re more or less working on the right thing.
Raven collects, captures and cleans data from multiple sources — from processes, people, and planning software — so teams have clearer data for quality KPI’s.
Martin: But yeah, I would say that the way to know is based on how if you’re in a supervisory role, how your employees respond to the metrics, even how they’re presented at management meetings. I’ve often seen plants where there’s a different set of books for operational metrics and for financial metrics.
So at some point, there’s this secondary data set on the shop floor that they’re using to figure out how things are going that’s not even the one that’s communicated up to management. So effectively, there’s these two messy data sets, and usually the plant manager is the one who’s responsible for stitching them together in an ideal world. But in many cases, they just have to manage these two different datasets that don’t even connect.
Martin: I would say that the concern that digital technology is like a big brother applying undue pressure on people on the shop floor is completely founded and completely warranted. And if people are afraid of this tech coming in to do that, in many cases, it is. And digital tech is often just a way to crank the pressure even more to get people to produce what’s required.
But this is not the way to get value out of tech that’s available. The problem is that pressure is being applied in the wrong direction. Somebody working at a process–an operator–if everything’s working well at their station, they do not need technology. They need technology to stay out of the way. When they do need technology is when something’s going wrong. Then, the best application of technology on the shop floor is to give them the ability to apply intense pressure on support and leadership of the organization to work on the right things, to make their job easier.
Martin: Yes. In some cases, the most progressive manufacturers are ones that don’t even present these classic KPIs and performance metrics on the shop floor. They exclusively put in the ability for the operators to just raise that flag and apply that pressure. And in some ways, the technology is there to give them a tool to tell the story of their day.
It’s always neat to see the first time operators and managers see the data and see what’s happening, really happening for the first time, because you know you got it right when the operators have this big smile on their face and they’re nodding their heads going, “I told you so. I told you this is what was happening.” And the managers go, “Yep, we’ve got to work on that. Yep.”
Martin: I think that that’s the moment where it’s almost like the hierarchy gets flipped upside down and everybody in the organization should be supporting those who are standing in front of the machines where, at the highest level, if you’re not standing in front of a machine actually producing product, you should be doing something to make it easier for them to do that.
Martin: I would say that the first thing that anybody on the support side, whether it be a manager or an engineer, supervisor, needs to do is to listen to their operators. And if we’re getting initial data that describes what’s happening, see if they agree with what you’re seeing. And often what will happen is they say, “You know what? This isn’t exactly what’s happening here. When the system says this, this is what’s actually happening.” And say, “Perfect.” If the tool is not effectively helping you to tell the story of your day, fix the tool.
Raven’s operator-friendly hubs let operators share data in real-time with their managers. In turn, managers can pull up real-time reports and review things like downtime and through-put numerous times a day or week.
Martin: Once you are all agreed that you now see a kind of how, when things are going well and when things are not going well, the next step is to demonstrate that you’re going to do something about it. Say for example, and we’ve seen this for several of our clients, one of the biggest issues is just waiting.
Say the operator is spending an hour a day waiting for their supervisor to come by and sign off on a sheet, or they’re waiting for material to come from the warehouse. So say for example, that the thing that they see is waiting, a report that describes a problem provides no value. And in this case here, how do you reduce waiting time? Well, maybe you bring in the operator and the maintenance tech, and you can work together.
Martin: One of the things that we often get wrong in manufacturing plants is how we present performance information. And I think we don’t, when I say how, I don’t think we take into consideration how presenting performance makes people feel. And the fact that we often do one size fits all, where everybody sees everything all the time, is demotivating for those who are in the middle of the pack. I think, especially with technology today, there’s a lot more flexibility to create more of a personalized performance management system that provides feedback in the way that works best for each individual.
I think outside of manufacturing, you can even think in a tech company, there’s no way that you would be presenting people’s performance reviews in public. And what we’re doing, in fact, on the shop floor is exactly that. Not only is it in public, it’s up for everyone to see. You can see if you and I were working on the shop floor, our names would be on a board, and what we achieved for that week would be on that board.
And the plant manager, when they’re bringing by a stranger, they’d say, “Oh, this is what Jordan did this week,” or, “This is what Martin did this week.” And that is maybe motivating for the top ones, but it’s a pretty quick way to lose credibility with your teams. The difference between an engaged team and a disengaged team is massive. No amount of technology is going to take a disengaged team and get them to perform as well as an engaged team with our technology.
Martin: For the last few decades, it’s been hammered into our heads that data has value; so then you often have leaders spending so much time with Excel creating reports. It’s amazing how that’s pulled leaders away from the shop floor. If you were to think about how effective a leader was 30, 40 years ago on the shop floor versus today, just by the nature of the fact that they were spending more time on the shop floor, I would say that there was probably a stronger connection to their teams than there is today.
So the key is to find ways to allow people to spend more time doing what we do best, which is to engage with one another, communicate, solve problems, and spend less time doing what we’re not particularly good at, which is collecting, interpreting, cleaning data, and playing with Excel.