Episode 190: The Fleet Manager’s Guide to Burning Old Data Without Burning Bridges

This is the Fleet Success Show dedicated to helping you and your fleet succeed using the experience of fleet professionals and Hall of Fame fleet legends. The show that explores how the pillars of stakeholder satisfaction, intentional culture, resource efficiency, and risk management all combine to lead your fleet to success. We're glad you're here. Now let's get into the show. All right, welcome back for another episode of the Fleet Success Show with your hosts, Josh Turley, Mark Canton, back in studio together again.

Yeah, man.

Been— I think it's been 2 months since you and I've been in studio together.

Yeah, it's been a while.

It's been a bit.

Yeah.

Um, yeah, so for those of you who are just joining us for the first time, Josh Turley, CEO over here at RTA Fleet. I've been running this thing for, well, since 2015 anyways, but been here part of 21 years helping fleets do fleet management software, which is going to be relevant today. Uh, Mark, you've been in fleet for 30-some-odd years, going on 40, I'm sure.

Not quite. Almost 30.

I'm turning 40 this year. And, you know, yeah, but yeah, tell our listeners about yourself.

Yeah, sure. I ran a fleet at Fordham University in New York City and their end transportation for 20 years and then been a consultant for the better part of 10 years with a heavy focus in analytics, right, and data and leveraging data to help you run a better fleet. So Then I got here a few years back. Uh, love it because I feel like there's such an opportunity for software and technology to help change the world and really help fleets succeed. So, um, here we are, man, and I love it. Let's do it.

So I know, uh, you know, we, we help fleets do a lot of stuff. Uh, recently we've been in the consulting business. We've brought on a whole stable of consultants, some incredible resumes. Uh, obviously as we talk about like Tony, Scott Rood, Scott Conlon joined the team, we've got Chuck Kramer on the team, Steve Salzgeber, who's been a podcast host for many years for us. You know, then we get the opportunity, we bring on new clients all the time with the fleet management software, right? And usually one of the biggest questions always comes up is what about all this data, right? What do we do with this? You know, how many years of data do we bring over? Things like that. And same thing like when we're doing consulting projects, you know, say we need this data, you know, 'cause anytime you're trying to analyze things and it's, you know, how far back do we need to go? And it's a hodgepodge usually. We talk about garbage in, garbage out, but like it's usually just a big garbage pail, you know? And I think about the average tenure of a fleet manager and how long fleet managers have been there, right? So talk to me about, let's just start with the first one. Data retention, right? How long should fleets keep their data?

Not long. Look, I understand that especially like in government settings, there'll be— well, really even in corporate settings, there's usually some sort of rule or law or regulation that has been set that this is how long we keep things for various reasons. Risk management's involved, lawyers are always involved and so forth. So I get if there's some sort of obligation there, go ahead and keep what you're obligated to keep for as long as you're obligated to keep. But most of the stuff in our world does not fall in that category. There's some, you know, you want to make sure you're PM compliance, you're inspection compliant, etc., but most of it does not fall into that world. And, and to be honest with you, like from an analytics perspective, when you're, you know, which is, which is looking at historical data to try to gain some insight about the future or a decision I want to make about the future, right? It's not relevant more than 5 years old. It's just not relevant. It's no good to you. And you'll have to— they'll have to take my word on it for now. But just think about like your everyday life, right? Think about the pace at which technology changes, right? You know, when you look at vehicles, for example, let's talk about capital costs of vehicles.

All right. Yeah.

It does not follow the typical cost index of inflation. No. Always goes faster.

Yes.

Right. Because you have that inflation factor, but new technologies are always being brought on and they're always being added to the vehicle, which makes the overall expense go up much more quickly. Right. So, you know, you'll see fleets all the time, they'll inflate data for 5 years. It's like, yep. And then they have this shortfall in their reserve fund because They didn't account for how those things are going up in price. So, you know, if you want to look at— let's say you're doing a life cycle analysis and you want capital costs. Capital costs more than 5 years ago are useless to you, right? Because you have then the same thing in the reverse, right? And COVID is a good example of that. You know, what are residual values going to be? You need to look at trends that are recent. It does you no good to look at, you know, what— how residual values changed over 5 years, 15 years ago. You want to know how they're changing over a recent time span, right? So when you're thinking in those terms, you know, M&R, look at what's happening right now. We talk all the time about what is actively happening in the industry. Well, there's a technician shortage. Do you think that drives cost or reduces cost? It probably drives cost, right? So looking at something from, you know, we have a customer recently or a prospect who I think is about to be a customer. Who's saying, yeah, we want to keep our data from 26 years ago and for the work order transaction data from, you know, what?

And you have to ask why.

Why? Yeah, why? And I— that's the interesting part.

So I'm sure the sentiment is like, well, I've still got that asset in service and I need to go know that I replaced the water pump on that vehicle. And how many times have I had to replace the water pump on that asset? And which water pump did I replace it with? And, you know, whoever. But yeah, going back, you know, 13, 20 years, at some point, you know, like, just doesn't make a difference anymore. No, who cares? Yeah, you know, in the words of Hillary Clinton, like, what difference does it make?

Yeah, exactly.

So it, it's interesting to me when you look at that, uh, I think about— you, you mentioned already the capital cost. Yeah, when my wife and I first got married, we bought a 2012 Oh no, we would have first been married. So it was a 2006 or 2007 Toyota Corolla.

Okay.

Toyota Corolla S. One of the nicer ones with the sporty steering wheel. The only S thing that was about that car and the spoiler. But I love that little car, whatever. But, and they last forever.

Right.

And I want to say I had 12,000 miles on it and we paid, I want to say $15,000, maybe, no, it was less than that. It was like $13,000.

Okay.

We're now looking for a car for my daughter. So I'm like, okay, well, let's go look at a little Toyota Corolla. One, you can't find any that are, you know, like this was 2008 we bought this car, so it was only a year old, had 12,000 miles on it.

Right.

Well, I'm seeing now Toyota Corollas that are 2021, so they're 4 years old already. They've got 25,000 miles on them and they're going for $20,000, $30,000.

Right.

It was almost triple the cost that we paid, you know, back in 2008.

Right.

Which definitely way outpaces inflation. Sure. So just even that curve is totally different. So you can't look at recent trends there. You already mentioned the, obviously the cost of technicians, cost of labor is totally different. So the M&R cost curves are there. But I just, I think about like that transactional history and you don't even know if it's the same part number anymore. You don't know like why does that part number even matter? Who cares that it was done on that day and this, you know, by this technician.

You don't need it.

Unless the NTSB is asking for it. Which they're not going to go back. I can't imagine anybody asking for more than 7 years of repair history on an asset.

Exactly.

You know, even if you've had that. So because I know this is going to come up, and especially in some of our government fleets where we may actually have an asset that's 30 years old.

Right.

And there, I think there's a mistaken belief, and maybe I could be wrong. If I'm wrong about this and there's some legal precedent and you've got your counsel that's told you otherwise, I don't think you need to keep, you know, down to the transaction-level detail for the entire life of that asset. Correct.

Exactly.

I think it's only like you got to talk to your legal counsel to find out about your retention policies. But 5 to 7 years is probably legit. And that's enough for, you know, any kind of legal counsel, legal defense that you might want to mount. In fact, I bet you if you asked your legal counsel, they would tell you you can't keep more than 7 years.

Get rid of it.

Yes. Because if you keep more than 7 years, you're actually liable for that data.

Right.

Okay, so we talked about those things. What we didn't talk about, and this is the thing I think is always, it's the thing that makes it easy for me to make this argument to people that are coming over.

Okay.

Most times when we have new customers, they are also new. New to the position, new to the job.

Yes, at least relatively speaking.

Because one of the first mandates is, hey, we need to get a fleet management system in place. We need to change how things are going, otherwise I wouldn't have been brought in.

Right.

Like they're brought in because something wasn't working. The previous fleet manager. Well, you think about how many times has that job turned over. And when you're doing that job and you're bringing in 13 years of data, the best way I can think about this is, well, I'm gonna finish the story and then I'll put another analogy on this. When you're bringing in those 13 years of data, you're also bringing in 13 years of past fleet managers. Right? So maybe it's, maybe it's only been one, right? It's been the guy that was there forever and did things his own way. And you're doing things a different way. That's not as bad, right? If at least if he was consistent in his job, then, then you know you're in good shoes. You know, at least you know what the garbage is, you know where the skeletons are buried. Whatever the, whatever bad practices they were doing, they were probably doing it the entire tenure.

Right.

But that's usually not the case. Usually you're taking over and it's been 3 or 4 fleet managers in that time span. and each fleet manager has had a different way of doing things, which means that the data's been getting recorded differently.

Right.

Which means there's no consistency in that dataset. Absolutely. Some of 'em were using, you know, mechanic time tracking, others were doing paper processes, others weren't reporting parts, some were just doing manual chargeouts. Like there's a million different ways that we've seen fleet manager turnover affect the integrity of the data in the system.

Great point.

Let alone all the processes with how do we handle downtime and driver pull-ups and, you know, stockouts and things of that nature. So not only are you bringing over 13 years of vehicle history, but you're bringing over 13 years of decision-making and, you know, we'll call it organizational garbage, right? Like the organizational health garbage and the dysfunction that was in that fleet department. You're bringing that all over too. It's great because the data just reflects the reality of what was happening in the moment.

Yeah.

And so I'm gonna take this one step further, which is you're bringing in all this stuff and it's sunk cost, right? If you don't know what sunk cost is, sunk cost essentially is like, you spend, let's say you spent $50,000 on an asset and you have a decision is, do you spend $20,000 more on it or do you spend $10,000, start completely over, but you get, you know, you start over with a brand new asset, right? Sunk cost fallacy is, yeah, but we've already put $50,000 into this project. It's just another $20,000 more. We can go, we can do this. Instead of just looking at future outlays and saying, oh, well, it's $10,000 to do this or $20,000 to do this. It doesn't matter where you spend it. You spend it somewhere.

Yeah, it's gone, it's gone.

Well, it also applies to organizations and people is that the decisions you made 5 years ago were made for a future that this person didn't know would exist. And so you're letting that person, you're letting your past self hold, right? And so in this case, we're letting the past version of the fleet dictate—

hold this fleet hostage—

dictate how the future fleet is going to operate, right? Because we're using that data from, you know, more than 5 years ago.

Yeah.

Um, and so there's this sunk cost fallacy that applies to data, which I had never thought about until this morning. Uh, I was listening to a book called The Psychology of Money. He was talking about this. Uh, fantastic book, by the way, if you ever want to, you know, learn more about that. But, um, he talked about this whole idea of sunk cost in our decision-making about who we are and who we are as like an organization and how the decisions we make are also a form of sunk cost.

Yeah, it's interesting.

Right?

You're trying to like fulfill or justify something you did in the past.

Yeah.

So you keep following that path. Yeah, that makes a lot of sense actually.

But then the data reflects those decisions, but then now you're making future decisions based off the decisions you made 5 years ago.

Right.

Based off of data because of how you were doing things instead of saying, you know what, like, that's sunk cost, it's sunk data, right? And so, but there's just an emotional tie that we have. Like, we hoard data.

That's what it is. Yes.

Yes. Uh, there should be an episode of Hoarders on this.

Well, there's a fear there. So that— so you, you're touching on a few things. I mean, number one, that's my intuition. I think it's like you don't know what you don't know and you fear what you don't know, right?

Yeah.

And it's like this legitimate fear of, oh my God, suppose there's something that's in there that some higher up or some lawsuit or some whatever is going to ask me. And it turns out I should have had it and I don't.

It's the unasked question.

Yes.

You don't know what you're going to get asked. And so you hoard data as a defense mechanism.

Right, right. As a just in case. Yep. And meanwhile, all it does is make it really difficult for you to use the good data, the relevant data that you have.

Yeah.

Effectively. Right. Because you get skewed outputs and you make a really good point too, like You know, think about, you know, let me go back a step. Even, even to your point, like you said, oftentimes it's changing positions or changing the guard, or even when it's not, you're going through some sort of transformation, right? You, you're changing systems or you're acquiring a system because you want to transform the way you do business for the better, right? And it's your opportunity to make improvements. And we talk about this a lot as well, process improvements. And one of the process improvements that you're going— or one of the types of process is how you track data and what you use that data for. That's going to be a major factor here. So to your point, something you just triggered for me is like, you know, you can think about how many like different ways a field can be used in a system. You know what I mean? Like, like we have a thing and as we see this, we try to sort of combat it and help people. But, you know, you have a field that says lease. Well, we— most of our customers, they see lease and they track the lease payment that Fleet is paying to the leasing company in lease. We had another customer come in and see that and immediately think, oh, that's what I'm charging my, you know, my customers for their wet lease, for what I'm providing to them. Not one day, right? So you don't know what your predecessor put in, right, that system. Or even if you know what went in, it might not be how you want to use it the next time.

Oh, heaven forbid, what if they never got trained, right?

Right, exactly.

That's half the problem is they never got trained in the system in the first place, correct?

Yes.

Oh, we've got one client that, and I know this for a fact because they actually got up in front of, they were part of a panel, and they talked about all the different ways that they bastardized the system essentially. If there was an empty field, they didn't care what the label said next to it. They're like, there's a spot for data, we're putting it in there, and then I know where to go get that in my custom report writer so I can dimensionize this data any which way I can think of. And if I need that data dimension, and I wanna say this is a longtime customer, so they started with us probably in 1988. Yeah, probably before we had unlimited custom fields.

Exactly.

Well, and even before we had the first 20 that we built, right, they were making their own custom fields out of pretty much any field.

Right.

But then they would do it differently. So they had to do all this if-then logic is like, well, if it's a vehicle, then the engine make isn't actually engine make. And then if it's a trailer, the engine make is used for this, right?

It's the refrigerator or something.

Oh my gosh. All over the place. And so we used to call it rod logic.

Right.

Is you had to know the rod logic is how did he use that field that one time?

And then, right.

Where's it documented? Well, the only place it was documented was in the customer port.

Right, exactly.

So when you took over for that guy and somebody did have to take over for that guy—

lost it.

And then it was like, okay, well, now that's not the right way to do it anymore.

Yes.

You talk about changing technology. The system that he started with in 1988 was green screen. Then he went to Windows. Right. Let's talk about this. Then he went to the Windows version of RTA, and then finally they're going to Fleet 360 on web.

Right.

Completely different user interfaces, completely different capabilities. You mentioned the unlimited user-defined fields.

Yeah, right.

You know, that's something we've had for probably 10 years. But if you're only bringing over 5 or, you know, if you're bringing over 30 years of data, well, now you've got to figure out where to put all that.

How do you map all that?

And you got to remap it and migrate it, you know. So yeah, we always have fun when people come to RTA and they're like, yeah, I want to bring over all my data. It's like, hold on, right? We have an opportunity here.

Yeah.

And we've talked about this on the podcast before.

Yeah. Yeah.

Right. There's an opportunity coming up every time because there's this tendency, right? And it's— I don't know, is it a scarcity mindset?

Is there—

there's a fear mindset for sure.

Yeah, for sure. But scarcity could be an element there too. And it's funny too, because oftentimes we'll advise people, you know, there, there are ways to quell your fear. And still not make the mistake of, you know, pulling all this data into your system.

You know, keep a flat file, go get a copy of the database, dump it out to Excel, you know, dump it out to Access database if you want to.

Exactly. You can always go get it now, but it's not muddying, you know, the good stuff and the new stuff.

In fact, if I, if I were going to do that— now I'm a little more technical, but, you know, I'm assuming everybody's got an IT department— if I can get a copy of my old database I would just go grab a copy of SQL Server Express, load that up somewhere. Hopefully we've got an enterprise version of SQL Server and you park the data there so you can query it, throw Power BI at it, Excel at it, whatever you need to.

Exactly.

And then if you do need to get any data out, it's still there. You just don't have to bring it all into the system.

Right. Which by the way is a good practice anyway, in my opinion, right? You should be creating periodic copies of the database anyway, you know? But in any event, that's one option. Obviously we talk about rolling up, yeah, you got a 20-year-old vehicle, and you— because that's going to be the first objection people listening to this. But Mark, I got a, I got a 15-year-old truck. What, do you want me to not keep the history on the whole? And I'm saying, yeah, don't. Yeah, keep the last 5 years from a sort of line item perspective, right, from a detailed perspective, and roll the rest of it up.

Okay, consolidate it.

Yeah, I know before 2020 this is how much we spent in the M&R, uh, you know, this is how much we've depreciated it on the books, etc., etc. And yeah, you can have some notes. You know what, hey, we We rechanged the hydraulic arms on the bucket, you know, you know, in year whatever, so that you have that note and you know it was done and whatever. I'm not saying you can't account for it in some way, right? But you don't need that level of transaction. And again, it's just going to hurt you.

You can have repair histories, right? So this is— we separate— you've got transaction-level detail, which is we posted this part on this date for this cost, right? Which that cost is garbage anyways.

Yes, exactly.

10 years ago, that water pump cost $45. Now that water pump costs $115.

Right.

The warranty, the metallurgy is completely different. It's probably made in a different factory, right? It's just, it's, you can't, it's useless.

Right.

But what you can do, you can roll it up and then if you want to keep line-level detail, so I know that, you know, hey, on this date we replaced the water pump and I know who the technician was that did it, but it's one line of data instead of three lines of data.

Right.

You know, there's, there's other reasons to do this as well, which is it can improve the speed of your system.

Right.

Right. If you're only querying, you know, you think about that example, If we roll, let's say, 3 to 5 transactions into one single record, that's an 80% reduction in the size of your data. That's an 80%, that's massive when you talk about scale and performance. And this is where I like, my technical background always comes in. If I reduce the number of records I have to read by 80%, I can improve the speed by 80%. That's just simple linear math. So your system speeds up all of a sudden. So there's some like benefits to cleaning up your house here.

Exactly. It's a good analogy as well. Like, it's just like your house, you know, you walk into your house, you haven't done a spring cleaning in a while, it feels cluttered, you're not as comfortable, you feel grumpy going in instead, right? And, and it's harder to find the things you're looking for that you really want. Yeah, you know, I can go on and on and on. You do a nice little cleanup, you get things organized, you get— you keep exactly what you need and what you're actually going to use. Throw out that shirt in the closet that maybe you love because your uncle gave it to you when you were 6, but you're just never going to put it on again.

You got to go KonMari on this stuff, right? Like, you just gotta— does this bring me joy? No, get rid of it. Okay, so take that further. Maybe I don't need the line level detail. I don't need to know that, you know, we had this many hours on this line and this many hours on this line. We consolidate it like you said, turn it into notes, right? And I can just see that, yeah, there was a work order, work order 18567, and it was open on this day, closed on this day. And then here's a summary of what was done on that work order, right? That's it. Done. Well, now I've taken 5 transactions, consolidated it to 1 line. I've taken 5 lines, consolidated to 1 work order. That's not an 80% reduction. That's like, you know, yeah, you're getting close to 90. It starts exponentially. Exactly right.

Yeah.

Because it's 80% times 80%.

So, and we've had to do this. I mean, we have customers who have 20— I mean, they've been with us forever. God bless them. Thank you. And, you know, 20-plus years and they're looking at their dashboards and they're going, yeah, but I don't know how to make sense of that pie graph. And it's like, well, and then I'm looking through their filters and I'm like, do you need— do you really need 1981 on this? Because if not, we can really help this go much faster and much better and actually be easier easier to use if we just kind of cut this puppy off and say, let's just look at the last, you know, 5 years, that sort of a thing.

When we were coming up with our, you know, because one of our goals at RTA is to help fleets save $100 billion, right? Not $100 billion, $1 billion. $1 billion.

$100 million. $1 billion.

We had started with $100 million, right? We're at $322 today, by the way.

Yep.

But we had set this goal of like, hey, let's save $100 million. And the question came up was like, since when? Right. Wow, that's a great question.

Yeah.

Well, so we made the decision, it was like, well, let's do it since the day I took over the company, right? Which is July 15th, 2017. And so that was the day we set, but we could have gone back further.

Sure.

And maybe 2017 wasn't, you know, maybe it was too far back. Maybe we should have only gone back 5 years. It just felt a little more emotional for that date.

Yeah, for obvious reasons.

For obvious reasons. Yeah. But it's definitely something like I wouldn't go back further than that 'cause like that wasn't, It wasn't you. It wasn't me. And it wasn't part of the, you know, like the system we were in. But it's like, "From this day forward, this is how much we've done." Right. But I think about that, I'm like, but the system was different back then. There was different time savings, like the way, like we're just making assumptions about how that was calculating back then.

Right.

And so from this day forward, it's like, no, every time we do this again, like we're gonna do new time studies.

Right.

And we're gonna say from this day forward, it's gonna be a different time calculation than it would be be, you know, based on what we did the study last year.

Right.

So we'll refresh that and keep those, those methods up to date.

Yeah.

But we're not going to go back and apply tomorrow's logic to yesterday's data. Yeah, this doesn't work.

That doesn't make sense. Yeah. So yeah, that's a good point. I could totally see us maybe one day saying, all right, well, let's, let's only do 360 and, you know, we'll start with that date or something like that. Right.

So yeah, we might purge and say, okay, well, now let's save a— let's save $1 billion, but only since the day we sunset Classic or something.

Yeah, right.

Yeah. We know for a fact how long it takes to do a work order in Paperless Shop in Fleet 360. So everything from this day forward is just that number.

Right. We'll get there.

If you're looking for fleet experts and former fleet managers sharing their best practices and advice, networking opportunities with hundreds of fleet professionals, and live hands-on training sessions and workshops designed to help you take your fleet to the next level. Then you need to register for RTA Connect 2025. It's the fleet event of the year. Join us September 22nd through 26th in Las Vegas and get the tips, tricks, and tools you need to help your fleet succeed. Visit www.rtafleet.com/connect to learn more.

And let's talk, you know, because I think the key thing and like, I think we're taught, we are implying this, but let's call this out a little bit, right? Like, Why do you have an FMIS? And one of the things we talk about here is there are a lot of organizations, unfortunately, and you know, it's my pet peeve, they use it as a digital filing cabinet.

Oh, for sure.

Right.

Yep.

And so it's just compliance tracking is really all it is. Right. And I get you have to do that. That's important. You do got to do it. But really what you're trying to do is you're trying to leverage data right, to gain some sort of insight to help you make better decisions so that you can address all of the 4 pillars of fleet management that we talk about. Right. So that's it. It's about being able to use the data. So think about that. If that's the goal, well, what do you need? Right. If your goal is to go out and change a tire, right, a hammer is no use to you. You know, what's the tool that you need? What's the information you need? What's the etc., etc.? And if you, if you just kind of pause and use that and reset on that every so often. Now, from the risk management perspective, again, you're going to have compliance issues. All right. This is what I have to keep. This is what either regulation or the lawyers or whoever. All right. Keep that. That's your risk management piece. Right. But from your resource efficiency, man, you know, from your stakeholder satisfaction, you know, who cares if they loved you 10 years ago? Yeah. What did they say about you last month? Yeah, right.

Exactly.

So So it really, again, history is important. And if you want to file some things away and, you know, got no problem with that. But when you're looking at the today and to tomorrow, you really, if your scope is too wide this way into history, there's no way you can well see what's going on here on today and in the future, right? You got to close that enough so that you can have a proper focus there, right? Like, trying to think of other ways to tell people It's the same thing like with cameras, you know, cameras and telescopes.

You gotta narrow the aperture so you can focus on the right aspect, otherwise it floods the camera lens and you get all this washout and like sun flares and stuff. Yeah. I think about, so like our financial data, we've obviously been in business since 1979 or whatever. I think I have financials going back to 2011, but we don't have transaction data going back that far.

Right.

Right, what do we do? Well, we collapsed it. And so now I've just got a bunch of journal entries and I think I have annual data from like 2011 through 2017. 15 or 17. And then I go monthly. You know, all I'm doing is just changing what we call granularity, right? Is how granular is the data, right? So I've got annuals and then I go monthly and then I want to say the last somewhere, it's either 3 or 5 years, we have full, you know, detailed transaction-level data. But every year we go through this process of, okay, it's time to collapse, right? You know, collapse that year into monthly. And then collapse those monthlies into annual just again to make it cleaner, make it easier. And because it just doesn't matter as much anymore.

Right.

But I still like being able to go back into 2011 and see, hey, where were we? In fact, I had somebody ask me this question because they were interviewing me for just, you know, hey, where's the company at versus where was it when you took over and, you know, like when your dad died? And I wish I could go back and find out, hey, when's the day that we hit $2 million in revenue? When's the day that we had our first million in revenue? I don't have that level of detail. I'm sure Dad's got that somewhere, but I can't ask him. So, but I went back and I was like, oh, well, when Dad passed away in 2014, this is where we were at in revenues. And then when I took over in '17, and then like, I just, I was able to kind of benchmark all these different events in the history and align them with some of this historical data. So I like being able to do that.

Sure, sure.

But I don't need all the granular. I don't need to know that Dad went to Office Depot and spent $279 on a computer monitor that year.

At 6:16 AM at— Yeah, like I just, this is garbage.

I don't need the credit card statement. Statements, right? So like, you know, you collapse it, you look for the big stories, reduce your granularity so that it's, you know, it's easier to see and just easier to look at in the system.

Yeah. And if you do know the history, oftentimes it's going to tell you, to your point earlier, like, you know, what is it that you were even keeping? You know, a lot of times the maturity of these organizations in those yesteryears was, you know, it's almost like mind-boggling. But, you know, what is it that you're even keeping? You know, And you see it too. It's not just maintenance and repair. And like, sometimes it's motor pool. You— why do you— you don't need motor pool reservations from 2005, man. You just don't. Nope. You just don't. I don't know. Yeah, you want— like you said, I like keeping the history. Hey, we had 30 assets then, now we got 130 assets.

Keep the big stuff, right? And that's where dashboards are nice, right? You can do snapshots with dashboards.

Sure.

Just get some snapshots and, you know, just keep the big metrics. Yeah.

Scan in the newsletter.

Maybe I want to. Yeah, right. Like keep the historical artifacts. I've got like some really cool pictures, things like that. Yeah. But like, okay, maybe it's not motor pool reservations. Maybe you want to just know how much in revenue did you pull in from your motor pool?

Sure.

For 30 assets, is your revenue per asset going up since then? Hopefully, you know, so you can kind of compare.

Summary data was great for that, right?

Yes.

Hey, look, we had 30 assets. This is the revenue. This was our expenses. I couldn't, couldn't, couldn't agree more with that. We advocate that. Keep that.

But, but yeah, no, you don't need to know that. Well, these ones, this was checked out on, you know, this on this Friday and went to this Friday or whatever. Like, you just don't need that level of detail, right?

So, so, so let's help folks kind of address this, you know, like, first off, if you have the fear element hitting you, for one, contact legal, contact leadership, right? Hey, are there any rules? What are the rules? What do I have to keep? And then you can make sure you're in compliance there. And, you know, and you don't have to have that fear, right? I've done my diligence. I know this is what I, you know, and again, how do you have to keep it? Because it might be totally fine to just keep that flat file as long as you can read it when you need to, like you said, throw it out to a CSV and put it in a SharePoint somewhere, right? In which case now, now this is this new system you're about to get into is completely your own and you control it and you, you know, and you can make the decision. So that's, that's probably step one, right? Go, go control the fear, right? And then step 2 is something along the lines of, well, what is it that I'm going to need to do that data analysis? Or what is it that I'm going to need to make good decisions about my future? Right. And, you know, maybe you talk to, you know, your— maybe you have an analyst on staff, you know, maybe IT has some analysts they can help you with. Again, go get some expert help, you know, if that's what you need to do. And when I say that, I'm not talking about a $30,000 consulting engagement for or something like this. I'm talking about just get some advice. Like, you know, what do I really need, right? What else? What else would you say is, you know, I think those, I mean like the fear thing is the key thing. I think, I really do think that's like the big, how do you get over that?

Well, and I'd say that should go down the longer you're in your career, right? If you're, you know, somebody who's long in the tooth, and this is something that's interesting to me is that people who have been in the career longer tend to hold onto more data.

Yeah.

You know, and I don't know why that correlation exists, but. And if you're younger in your career, maybe you just don't know better. But it's interesting is that that goes complete opposite to the thing which you said, which is I don't know what question that would be asked. Well, the longer you're in your career, you've been asked a lot of questions. Think about those. Right. What are the questions I haven't been asked?

You've seen a few accidents of you though.

And this is where I like to be really proactive about, well, this is what I'm planning on showing. What do I need to show every month? What do I need to show every year? You be intentional about the data you're showing. Most of the time that's going to be sufficient enough, right? And you pick the metrics and you, you know, share more than you need to for sure, but that will drive the conversation. And if there's follow-up and little things like that, then you address and say, well, we can start tracking that going forward, right? You don't have to go back and be like, well, I, I don't have that data historically. Let me go see if I can fudge it together, right? I'm going to hold on to this forever just because I don't want to look like an idiot. No, we're not tracking that. Yeah, you know, so like it's just have that mindset of things is, well, I'll start doing that from this point forward instead of I'm going to keep this forever just in case that one day I get asked. Because otherwise you're going to keep all this data for that one moment that will never come. And you're doing it for probably 20 or 30 of those moments when instead you can just wait for that to happen and then start collecting the data for that thing. Yeah, be proactive, be smart and say these are the 20 or 30 things that I want to know. These are the things I think they should know.

Right.

And you start with that.

Yeah.

And then everything else, just if it's not pertinent, it's not necessary. Let it go.

Let it go. Yeah. And you know, something you said earlier too, I think is, is interesting, which is, you know, what do you know about the past version of the organization? If you know nothing, that's an indicator.

That's a huge— I don't think you can understate that.

Right.

It's a huge indicator that that data has little relevance to you other than telling you what the other guy did wrong.

Correct.

And the fact, like, I would, I would almost like, that's like a poison pill. And I don't want my data getting—

exactly—

getting sullied by that. Right, right. Like it's almost tainted where it's, you know, it can— it's going to rot. It's rotten fruit, right? It's going to rot my good fruit.

Right.

So you got to keep it separated.

Yeah. Yeah. It's a great point. And that's kind of where I'm going with it. Like, if you do know that your labor rate hasn't changed in the last 15 years and it's currently $45 an hour, 13, you know, like, well, in that case, you could just throw out all the M&O. I mean, you still want to know your— maybe your parts turn, you know, like, yeah, but I mean, geez, right? So if you know that the practices were bad, well, that just makes all that stuff all the, you know, all the more reason to not hold on to it, hold on to it, right?

Yeah, you know, so I probably am looking at 2 to 3 years, uh, you know, if I'm trying to do any kind of analysis or trend analysis, like, I'll probably keep 2 to 3 years of data.

I agree.

Even if I didn't like the way the other guy was doing things, like I would, you know, 2 or 3 years, that would help me think things through and, oh, okay, I know why these are the problems. Or I'll go back and why is this such a problem? Go be like, oh, okay, there you go, right? Because this is how they were tracking it. We don't want to do that. We're going to do it this way instead.

Right.

But just know, like, if I'm running a dashboard and it's got a 12-month time span on it, it's going to take 12 months for any decision you make today to show up on that dashboard. Because it's obviously like you're reporting 12 months in arrears. So it's going to take that much time for you to get that process clean, right, all the way through for next year.

Right.

It's just how data and time works.

It's just how it works. Yeah. And in our experience, like, you know, I talked to Griffin earlier today, you know, he's our fleet management analyst and Christie, and we're blessed to have a few. And in our experience with the consulting engagements, like, we do usually need at least 2 years of decent data. Yeah. Okay. 3 years. If there's, if there's a critical mass in the data, solid 3, 3 years is like the sweet spot, right? Yeah. We don't ever want more than 5. Now, you made the point earlier about 7. 7 is a common number you see in governments and others where they have to keep that or whatever. So, so I get it. Like, I'm comfortable with that.

Yeah. But so you can hold on to 7.

Yeah.

We're never going to ask you for more than 5. And we might stop at 3.

Yeah.

But again, if you've got garbage data, I don't want 3 years of that. No, give me two.

Give me two.

I'll take two. And then we'll like, you know, because even then there's going to be holes in the data, there's going to be, you know, gaps, and it's going to be inconsistent, right?

It's easier to just clean it up and go get some industry or whatever you need to do.

And the smaller amount of data I have to clean up, the better, the faster the project goes, right?

Exactly.

It's not that counterintuitive, but you know, it's, it's one of those things that you're just thinking, oh, the more data the better, right? No, no, it's not. It's just more stuff that you have to go through.

Yes.

You know, it's like having more space in your garage. Yeah, it just means that you have more clutter.

Yeah, right. You know, another good indicator, just— I know I'm being it, but someone sends us a spreadsheet, right? And they got rows and rows and rows and columns and columns and columns, right? If your spreadsheet looks like this, here's your— let's say your asset along the first column, and there's blank spots. Even if you got a bunch filled, But there's blank spots all over. Swiss cheese. Swiss cheese. That's a good one. Yeah. Swiss cheese or, you know, Polkadot. Nope.

It's not consistent.

Don't bother.

Yeah. You want consistent data. You want— I would say, what, like 80%, 90% participation rate in any data field?

Correct.

In order to gain any kind of insight. That's reliable. If you're 50/50 on any data field, it's like, eh, it's not really going to give you anything. Maybe some vehicles that's really important, other vehicles it's not.

Right. Unless you find obvious lines of partition where it's like, yeah, The fleet's 50/50, but I got 100% of police.

Yes. Right? Well, now you can say something about police. 100% of this class code, right? And we've seen that, or this type of, you know, make and model or whatnot.

Right.

This department. That makes sense.

Yeah.

But you have to organize it that way.

Yeah.

But yeah, if it's just Swiss cheese all over the place and like—

It's a pretty good indicator.

Because when we were acquiring assets in 2011 and 2015, this fleet manager required us to put in these data fields.

Right.

From 2015 to 2018, this fleet manager didn't. And then in 2018, we got a new fleet manager and he required not only the first data fields, but then he added another set of data fields. When we talk about sunk costs, this is what I'm talking about.

Right.

Right. Is that's what creates Swiss cheese in your datasets.

Yes. So yeah, that's interesting.

Yeah. And that happens all the time. You know, it happens, you know, for anybody that's had to take over for somebody, it's just you're, you know, that data is literally just a representation of how they ran things.

Right. So.

Until you're in the job for a few years, then, then it becomes your data and it's all your fault at that point. That's a different podcast episode. So yeah, so you talk to legal, you think about proactively what data, you know, what metrics really do I want to drive? Because the metrics should drive your data collection policies.

Bingo.

If you're collecting data and you don't have a metric tied to it, you're collecting the wrong data, right? So start with the metrics, identify the data that the metrics need to support them. Then think, how far back do I need to go for this metric to be reliable and usable? Most of the time, anywhere from a month, 12 months, maybe 2 years. Like you said, you're not usually going to go back more than 3.

Yeah.

I don't know, where do you go from there? Figure out how to purge, you know, and I would say you don't have to switch fleet management systems to do this.

No.

A lot of them have the capability of archiving data or purging data.

Right.

I was like archiving. Maybe I'm a hoarder too, just in case legal comes back and says, hey, like, you don't need all that data. But if you need 7 years, then okay, well, after 7 years, then start purging, right? Um, or finding a way to like do a roll-up function in the data. I would much rather roll up than purge, right? Uh, if you are switching FMIS, that's a great opportunity. We've talked about that in past podcast episodes. It's a great time to clean your data Don't bring in the garbage just because that's the way it is and you don't have time. Make the time, make the investment. Your future self will thank you for the cost you're sinking today.

Right.

In fact, your future fleet managers will thank you for the time and effort you put into making this a clean dataset. It's a great opportunity because this system that you're going to has a completely different set of processes and it's difficult to compare. Well, is this a time saving because of the system? Is this a time saving because we implemented a new process or is this a just a time saving because I'm a new fleet manager and I have a different way of doing things. Right. You know, and so it's, it's very hard to be comparing apples to oranges when you're bringing in old data from an old system.

Right.

Doesn't mean you don't do it. It just means take the opportunity to clean it before you get there. So at least you're starting off on the right foot.

Yeah.

We've even gone so far as to say don't bring any over any data.

Yeah.

Just bring over the records, the master records.

Right.

The file cabinet.

The file cabinet.

Bring me your assets. Give me your parts. Give me your employees. Maybe your vendors.

Start from there.

Start from there.

And it's funny you say that, but man, the customers who choose to do that, they're so much happier. They're so much happier.

They're so much, and yet the customers that like, we fight this battle like tooth and nail and they just, you know, the ones that really, really wanna bring over the data, they're so frustrated, you know, even 6 months into it because the system's not working, this data's not showing the right way. And it's like, well yeah, because that system didn't show this data this way, right? We have a different way of doing things. Things.

Yeah, but they find that they're used— like, they get value out of the system so much more quickly, right? And again, they still have the data, it's just in a flat file somewhere, right? It's not that it's not there if they need it. Oh, I really need to know what we did 3 years ago to this thing, and there was a frame damage, you know, nobody here— nobody works here who was in the shop at that time.

And you're listening to Open Work Orders? Yeah, bring in your Open Work Orders. Yeah, let's not stop the shop.

Sure.

But yeah, that's pretty much You know, that's it.

But that's actually a really great point because when they come in flat, boy, they, they hit the ground.

Those customers are the ones who tend to hit the ground running and they set it up the exact way that they want it from the get-go instead of— because this is why it forces you to look at it, it forces you to clean it. And what you're doing is you're overcoming all the bureaucratic inertia of, well, this is just what was in there and I'm just accepting the status quo. It keeps the status quo from poisoning the next system, right? It's that sunk cost. It keeps the sunk costs from coming in and affecting the future self. Yeah, it's, it's a choice you have to make to do it though, because it does take an investment to clean up the data and massage and say, no, okay, we're going to invest and start over.

Right.

But from that day forward, it's you, it's your fleet, it's your data. And this new system.

Yeah, it's like a great point. You know, I never, I've never connected those two dots before this conversation. But one of the other things we always say is it's a great— when you got a new system, it's a great opportunity to clean up your processes, right? And it's true, it is. But you just, you just kind of said something that's like, boom, light bulb. It's like that attachment to the old data is a direct sort of, let's say, correlation to attachment to the old process. Yeah, right. And even, even when you get folks who are willing to think, and it's like, yeah, but we had this data, so we need to do it this way because we got to collect this data. What does this data do for you? What problem does it solve?

We have to keep that data relevant, right? You know, in order to keep that data relevant, that means that we have to do the process the same way, otherwise it makes that data irrelevant.

Yeah, which is like some subconscious thing, which means the work I did in the past wasn't relevant or something like that.

I don't know, but we're going to call it— it's not sunk cost policy now, it's going to be sunk self policy, right? Like, you know, it's the, the, you know, the past self of you. You're letting the past you dictate what future you does.

Yes, yes.

Isn't that crazy?

Very much so, very much so. And I, and I know we're all guilty of it a little bit, but Man, yeah, yeah, interesting. Don't do it.

So, so the other reason, like, you could start over clean, you start over fresh, but like, yeah, there's just nothing holding you back from adopting a new process, a new technology, a new piece of the software that you wouldn't use before. Maybe you couldn't use because the data didn't support it. I mean, there's some historical stuff that's really cool, like warranty tracking, whatever. But for the most part, just let the— like I said, 12 months of lead time and in a year you'll be further ahead than you would be if you try to bring over all this old data and then like shoehorn your new processes into the, you know, your old processes into the new system.

Yeah, I think we've hit it, man. I think we've hit it again. The data—

this horse is dead.

Yeah. And again, when we think about everything we've covered, right, like if you don't believe us now, I don't know, like even if it's great data, even if you are the best and you're his your organization's history was perfect, if such a thing exists, at collecting data and cleaning data and keeping data great, it still becomes irrelevant. It still becomes irrelevant. That's what we started with, obsolescence. You know, at some point it just— there's no value to it to you because there's nothing you can do with it, so to speak, from an analytical perspective. And again, in our industry, the way costs change, there's just no correlation. It changes so quickly that there's— I'm just trying to think of something I could accomplish with 15-year-old data, and I, I can't think of anything. And I can't remember ever using it for any purpose, you know. So number one, probably you're spoiling the well. Number two, it's probably creating some sort of resistance to, or harming your opportunity to improve your organization in some way, shape, or form. And number 3, even if none of that is true, it's still useless to you. So it's like keeping the old shirt that doesn't fit anymore. You know, I was a chubby guy. One day I'm going to fit in it again. Like, yeah, but by then the moths will have eaten it in half.

Or the opposite, right? Everybody's on Ozempic now and they're all losing weight, but they're not getting rid of the old clothes.

Yes, yes. In case I gain it back.

In case. Just in case. Because I don't know. Right. You're almost like counting on it to happen.

Yes.

That's even more reason, you know, Cortés, when he came over from Spain and he got to the Americas, he burned the ships because he was so committed to this thing.

Yeah.

When you do the same thing, burn the data, right? Get rid of it and say, nope, this is the new way we're doing things.

And let's start fresh.

Yeah, let's start fresh. Yeah, I like it.

Cool.

All right. That's going to do it for us. As always, if you have any questions, we'd love to hear from you. I know I got a couple emails in this week to podcast@rtafleet.com. Love to hear from you guys. Questions, comments. If you do know that I said something wrong about legal and retention. Yeah, I would love to hear that.

Educate us.

But I'm pretty sure go talk to your counsel and they'll tell you that, oh no, we want to get rid of that.

Yeah.

So you'd be surprised at what they say. Otherwise, until next time.

Okay, everybody.

We hope you enjoyed the show. If you'd like to help support the podcast, please share it with others, post about it on LinkedIn, or leave a review. Want to learn more about our fleet management software Fleet 360? Check us out at rtafleet.com. If you want to learn more about having your fleet evaluated by our consulting team, go to fleetconsulting.com.