Cold Open & Turnarounds 101
Welcome to the Major Project podcast, your inside. Look at the high stakes world of billion dollar projects.
Welcome to the Major Project podcast. I'm your host, Orion Matthews. Big builds get the headlines, but what happens after your project is done? Today we're talking turnarounds, those high stakes maintenance events that can make or break a billion dollar asset. Joining us is Grant Stead from Stead Global to give us that inside look at the riskiest and most complex event in a mega projects lifecycle.
Let's roll.
Dude. I have been, I've been like working right now. Yeah. I dunno. Are like, should we just get started or, yeah. I'm like, I don't know. I have been. Absolutely on the grind, dude. I have gotten, and I'm just so excited about what I'm working on that, I mean, I've been taking [00:01:00] naps in the evening. I've been going to bed at like midnight, one o'clock, and I've been getting back up at like three, four o'clock in the morning.
Yeah. After it again, I am on one right now, and so I'm, I'm gonna try my best to get out of that and be like high level with you. Yeah. But
well. What if we go there too, because, uh, sure. We gotta capture that energy, right? So yeah, we got, we gotta go there. If you can go there. I don't know if it's like secret, but
no, no, I'm, I'm, nothing I do is a secret to anybody.
All right. Well, thank you for having me, Orion. I always enjoy talking with you. Grant Stead, I've got a, uh, entrepreneurial spirit and I have, uh, done quite a few things. A little bit of a rolling stone. Got a degree in aerospace engineering. I've never used, I've done large turnaround projects, scale of, four to $500 million for a turnaround.
And have just always looked at [00:02:00] how do we make this better? And I'm pretty much perpetually dissatisfied. And so, so hopefully we can, uh, share some of the things that are bringing us satisfaction lately and kind of talk about that space a little bit. And so, I mean, I'm, I'm assuming a lot of your audience probably doesn't hear a lot about turnarounds then.
Yeah. So that was my first question. I think it was my, when we were talking beforehand, I was like what is a turnaround? You know? It's, it's, yeah. It's a, it's a very like, chill term that I think a lot of people in our industry don't even know what turnarounds are. Yeah. Um, so can you maybe enlighten us, just talk a little bit about like what is turnaround?
So really simply a turnaround, uh, in refining or an industrial facility is a, uh, it's a maintenance event. At its core. But of course we are creatures of opportunity. And so what you have to do is you have to shut down all the different units on, in, in a refinery that you [00:03:00] wanna do maintenance on.
And, um, and while you do that, you end up doing a lot of other things too. So you'll say, you know, because the units are operating 24 7, that's your only opportunity to do tie-ins for major capital projects. And so I'm sure all your major capital guys, you know, when you're doing brownfield work, how important all your time in planning is because you'll have a whole package of pre turnaround work that you're trying to get done.
You'll have a whole package of turnaround work you're trying to get done, and then you'll have a whole package of posts to actually finish the job. And so, it's a, it's a pretty critical moment in the refinery. It's, um. It's also, one of the biggest points of failure that you can have as well.
So, when you're doing a turnaround, one of the most important things you have a, a one-time startup. And so there's a lot of safety risk when you're shutting down units and starting up units and, um, it, uh, a turnaround is a very short duration, but high impact activity. And so, [00:04:00] for example, if you have a, a $500 million turnaround you're gonna spend almost all of that money within a 45 60 day period.
And so, you know, you just do quick math. You know, you got $500 million turnaround, you're gonna have five to 6,000 people at peak operating on that event. You're gonna have, you're gonna have a labor burn rate in the eight. $8 million a day range. You're working 24 7 on the project. And, um, and when you do that, you end up having your resources come in pretty much overnight.
You'll have a base group of people, you know, maybe 200 people kind of planning and, and working on the event for a year. And then pretty much within two or three shifts you'll go up to full staff of like 4,000 people. So establishing a culture like that and establishing your, how do you do business and what [00:05:00] tasks need to be done and when they need to be done to achieve your dates and milestones.
And then of course then at the end of these 45, 50 days, you then have to punch out the whole unit and make sure that when you start it up. You've got a one-time startup, you've got no leaks, you've got no compressor lineup issues. You've got no problems. And so when you do a turnaround, your typical scope of work, you are opening up every tower and you're crawling the towers and you're cleaning them, you're tearing compressors apart, you're replacing entire piping circuits.
You are basically touching every piece of equipment in that unit, uh, with thousands of people like Anton an hill and controlling that is an absolute effort.
Okay. I'm like I, my mind's blown, right? Because as a, as a project guy I might spend $500 million, but I'm gonna take a year or two to do that.
So trying to understand, like, how do you, [00:06:00] you, you said culture's really important. I guess I wanna touch on that for a sec. Like, let's say I'm part of that 4,000 person crew. So we're gonna spend $500 million in 45 days. Yeah.
We're gonna turn around this facility, which means I'm, everything's gonna get ripped down to the studs in many ways, and we're gonna turn it back up and we've gotta start it, restart it the first time.
And the reason we're doing this is because we're gonna lose money. Right. These refineries are burning cash. Not only are you burning cash investing in it, you're also losing money while it's offline. Yeah,
that's right.
You've got your
lost opportunity cost of not running it. And so some of these units you know, a cat crack or just that one unit, uh, it'll lose $500,000 a day in lost opportunity.
Wow. Because it's not producing product. Uh, the other backend on this is every one of these facilities have got tons of contracts. And so I've been on turnarounds that have gone poorly and they'll have to declare force majeure on a contract 'cause they can't provide jet fuel to Delta. Oh yeah. It, so it's very critical [00:07:00] because they only have so much storage.
So moving into an event, they start building storage of different products.
Culture Is Cadence
Yeah.
So that they can make it through the downtime, and it is a highly orchestrated
event. Okay. So take me from the perspective then. Let's say I'm like a welder and I'm part of a 4,000 person crew. Like when do I first intersect this project?
You've been planning it for a year. Yep. When do I get my paperwork? And how do you grant as one of the project leaders, what do you want for the culture experience of me as like a welder? Like what's my,
yeah. What do
I look like looking from the bottom up at this thing? Yeah.
So for me, and this has been kind of like my hill that I die on is that I believe that culture is cadence.
I believe that following. The, following the footsteps, right? But if you don't know where you're going, that creates cultural issues. [00:08:00] And so the main culture that I believe that you need to incite on these events is that when someone shows up, we can tell them, I need you to go do this. You a welder, what are you there to do?
You're there to drop weld material. That's your goal. You are there to burn rod. And if I can't effectively tell 400 welders on day one where to go put their stick, boom, you just blew up the whole turnaround because you are disorganized and you don't know what needs doing. And so what that means is that on day one, when 400 welders come up, I need to be able to hand every single one of 'em go make this weld over here using this weld specification.
Using this weld procedure, tell 'em exactly where it is, what work packet it's on, and why they need to do it. They need to be handed to go do it. [00:09:00] And so the type of planning that we have to do to achieve that is extremely detailed. And so we break down our scopes of work into every single flange, every weld.
Everything that you're touching, you have to track because you have to put it back together. It's like if you take a, if you take a little toy apart and you put it back together and you've got four pieces left over, you're like, uhoh, right? Yeah. And, and so for a so really for a turnaround on that scale, you're gonna have anywhere between 70 to a hundred thousand activities in a schedule.
Wow. In a pretty piece of schedule. And that's what you need. And so for me, you get good safety performance, you get good culture by knowing what you're doing. The, all the incidents, everything that happens is when you don't know, right? Humans, we're there to do something. These people are there to work.
They're not [00:10:00] there to be a thought maker. And so if you can't tell 'em what to do, they're gonna find something to do. And that is where the culture goes awry. And when everybody knows what needs doing, you're not standing over the guys ask 'em, Hey, well when are you gonna get that done? 'cause we gotta go do something else in rushing them.
And so you'll literally, I use the term no running by the pool. You'll literally be on turnaround events that are high scale, a lot of energy, people having low. Emotional IQ and really getting hot and bothered when they don't know what's happening or what they're supposed to be doing. Yeah. And so I, I really find that, I've been on events where a truck driver will come to deliver materials and because everybody else is running around and freaking out, they start running around too.
And we've had truck drivers trip over hoses because they're rushing and break an ankle. We've had, we, we had a truck driver one [00:11:00] time, and I've, I've heard, got stuck on some train tracks because he wasn't familiar with the facility. There's so many things that come just from, just from not having a controlled cadence.
And that all starts in the planning. And uh, and so that's pretty much one of our primary objectives when we go into a turnaround, is really taking the scope of work that we have to do, really validating that we have to do it, breaking it down into its components so that every day we can make sure that everybody knows what they need to do.
And that's critical. And you know, that energy you were talking about, so I used to run, uh, tech support teams, right? We do infrastructure support, and we had this term, we called it heard spook, where it was like if somebody would call in with an issue, you'd have this like medical facility or something, and all of a sudden you'd start seeing all these people calling in and like getting agitated.
And it was like this herd spook could happen and you could lose control of a system of people. Uh, yeah. Just because [00:12:00] they start. There, there's like a weird energy that builds on itself, and you gotta put it, you gotta push that down and that,
yeah.
So here's my question on all that. What how many levels of leadership do you have?
So
it's, um, it's executed kind of interestingly, and so you pretty much have your senior turnaround manager. And then for an event that size, you'll typically break it into blocks. So an event that size, you'll have two to four blocks, physical areas of execution leaders. Underneath those execution leaders, you'll typically have unit and discipline, specifically leaders where split by days and nights 'cause it's gonna execute that way.
And so you'll have, a piping leader over one unit. Now those guys all kind of our owner rep side. Then it pretty much kind of, you select your contractors. So you have main mechanical contractors, turnarounds are done, time and [00:13:00] materials. And so pretty much every contract's t and m, so you work with your contractors to have rates on file and then, uh, your contractors come in and they basically fill under those discipline leads.
And so, you know, you might have. On an event that size, you might have 30 or 40 general foreman and, um, superintendents and then kind of working down. And typically our crew mixes, we'll have one foreman in like a team of about six to eight. Okay. And so when you've got four to 5,000 people, it's uh, it's a big dissemination issue.
So you don't really, that's where the planning comes in. You don't really have time when all those people are there to have an idea at the top and push it all the way down. Right. You've gotta be ready to execute.
So, question, do you have daily standups with these people? Like what's your leadership? Yeah, man.
Is it scrum, for lack of a better word, for your project management? Like you're obviously not doing monthly status report meetings. Um, [00:14:00] yeah. How so?
Battle Rhythm, Discovery & Found Work
Yeah, so, so it goes through some phases, like anything, right? So during the planning phase, you know, you've got your long-term phase, you're doing your typically weekly reporting on where we're at with planning and what equipment we planned and what we haven't.
During the event we have daily meetings or every shift. So this is mirrored day and night shift. We typically have a 5:30 AM meeting where we get all the senior leaders together. So for a project that size, the last room I was in probably had like 50 to 60 people in that meeting. Okay. And so we have 5:30 AM and a 5:30 PM and that's kind of like a shift turnover.
And so most of those guys, they show up at 5:00 AM. They get a shift turnover in the field. They show up in the meeting at five 30 for a 30 minute standup to where we go through all the critical path and productivity and what we got done and what we need to get done that's critical for that day. Then we'll typically [00:15:00] have, there'll be like kind of a 9:00 AM ish that we talk about project controls and productivity and performance, and we'll do crew balancing and move crews from one unit to another unit or, move things around a little bit with our staffing levels.
And then we typically have a 10:00 AM uh, like change management meeting because one of the things that's critical on a turnaround is a lot of these vessels, you haven't been in them for. 10 years. And so when you open them, it's like a can of worms. You don't know what you're gonna find. So one of the most critical periods in a turnaround, one of our biggest checkpoints is called the discovery phase.
And so we have a discovery meeting where we say, what did we find? Oh, no.
Okay. Yeah. So that's early in the project. Like right after your shutdown. That's
every day. Yeah.
Or Oh, every day you have a discovery.
Okay. Every day. I guess that makes sense. And so typically we try to be out of the discovery window [00:16:00] about a third of the way through the event.
Okay. So if it's a, you know, a 45 day event, we try within the first 15 days, we wanna be like out of discovery. Like we've already opened everything. So the priority for a turnaround is to open everything and do your inspections. So the first thing you wanna do is I don't wanna put anything back together.
I don't wanna spend manpower fixing anything. If I've got manpower, I'm tearing this unit apart
because we wanna inspect it. Okay. So discovery phase and then we'll get to the next one. But quick question on that one. Let's say I am you know, the owner and you've got 15 days. Is it like, I've got a budget for this turnaround of 500 million, but you rip it all open and all of a sudden you figure out there's a huge problem here.
Do you have to go back to me to authorize another a hundred million? Like how do you budget? Yes, we is there this massive contingency budget that's already approved? 'cause some of these big [00:17:00] orgs, like you gotta go to the board of directors to get that approval. So you've got this facility ripped open, like how do you deal with the discovery phase?
I guess like bad discoveries in that phase and how do you get the money? A hundred percent. So turnarounds are so large that they do end up on quarterly and annual reports and the cost and all that is reported and audited at this scale. And so, turnaround performance is huge for shareholders.
Uh, any shareholders of, of these major companies they know that that's a heavily report thing. Just like Holly Frontier Sinclair Tim Go, you know, he literally just posted on LinkedIn about their quarterly performance and how excited he is that they're done with all their major turnarounds for the next couple years.
Right. Nice. They've done and it is, it's huge. You're, you're a hundred percent right and quite literally yes. We often have to go back to the board. I mean, I've been on events where we'll open up a major compressor. And, [00:18:00] and this was on one, the compressor got all installed and when ops went to go start up the unit, they lined it up wrong and they blew up the compressor and we had to spend another 30 days rebuilding the whole compressor again.
And so you're talking, they did declare force majeure. They had to like, so that is huge dollars. And uh, and that is exactly how it happens. So what we do is, is one, we have a ton of inspection data, historical inspection data. We're all getting very, very good at kind of like managing and predicting what we're, what, what you're gonna see.
And so typically what we do is a lot of places do just like a flat contingency. They'll take the turnaround budget, Hey it's, $90 million and they'll just throw 10% on top. Okay. So that's a pretty common amount that we have like a pre-authorization for. And then. But what, what we do also is we'll go through each piece of equipment and we'll put like a risk-based contingency, [00:19:00] and then when you open up, and then sometimes in our schedule we'll plan for a certain percentage of found.
So we'll say, okay, we might find this much refractory sping wall off, or we might find this much, corrosion or this many trays, outta commission, on the tower, whatever. And so, we'll sometimes pre-plan for some of it. And then if you open up the vessel and it looks great and everything's good, then you give the contingency back.
Okay. If you open oven as bad, well now you need to use the contingency and sometimes a little more.
And that's the discovery phase is when you usually wanna be using that contingency or That's right. Sort of getting a disposition on it or giving it back. That's,
yes, that's exactly right. And so we follow a, uh, a pretty rigorous change management process that is built for speed.
So we typically get change. Changes approved through the entire organization. Um, I mean, I've seen, a $20 million [00:20:00] change get approved in less than 24 hours. Wow. And so there, there's, there's quite a lot of interesting things that you find. 'cause you gotta remember these are refineries and you're, you're pulling things out of the ground.
So we found where we've opened exchangers and there'll be, you know, norm in it, not normally occurring radioactive material. And then we've got people out there and Geiger suits and
you know,
do you
like fly those people in on a private jet or something from Yeah, wherever. 'cause yes, again, you've, you've got no time.
Yeah. So we've, we've had to, uh, for certain types of material and equipment we've had to charter planes. We've, uh, we've had, uh, a big rotor one time on a truck. We'll always pay for a, tandem trucking where we have two truck drivers. So one can sleep while the other one drives to ship things from different facilities for repairs.
We have a facility called hot shots so people will pay, will pay. Drivers 800 bucks to drive three hours with a part. Wow. And I've seen the smallest of [00:21:00] parts shut down an entire startup. We needed a little light bulb. Uh, I was working in a No way. Yeah. I was working in a fuel gas recovery unit, turn around, and we needed a light bulb, but it was an indicator for something.
And if you didn't have that indicator, light functioning, they couldn't start up the unit. And we had to, are you kidding me? We had to, I'm not joking. And we had to petition. And this is, you know, you find this out five hours before you're supposed to start up. So now we had to do management of change, MOC, on procedure changes to make sure we could start up.
And so we had to have an operator place standing there watching the pump instead of the light. And like you had to get all that done within, you know, five hours
or you Right. So it's, it's interesting. Well, and, and going back to that, you said one time startup is what's important, right? Yeah. Alright, so getting back to the phases real quick. So you have discovery phase and is there like another phase after that or is it like out of discovery and then all the way till startup is your next pretty [00:22:00] much piece? Okay, so
I have quite a lot of phases that I track. I think there's like nine of them, but basically, you know, you have to.
Isolate the unit. So you do all your blinding and then, you know, you tear everything apart. You know, you do all your initial inspections, you know, you have to do some hydro blasting and cleaning along the way. And then when you get all of your initial inspections done, that's when you're outta your discovery phase.
You found everything you need to find. And then basically you do your scope of work. So then you do all the repairs, things you need to do, and then basically you start restoring the unit. You basically just start putting everything back together. And so, you know, there's usually quite a bit of piping work that we're doing that you track and manage has just more of a, um, you know, earned value methodology.
But the main equipment is really what drives the event. And, uh, and there's a lot of this equipment that's like not built for maintenance. So if you do a, a, a, a cat head off turnaround, I [00:23:00] mean, you are, you have to actually cut the vessel apart and lift it. And, um, and get into it that way. So some of these things you can just open a manway and get in, and other ones you have to tear the whole machine apart and Wow.
Weld it back together. So it's, uh, it's an interesting endeavor, that's for sure. All right.
The Mercenary Workforce
So then I guess like, one, one question I have on this is where do you find 4,000 people? I mean, these, these places, these refineries are not exactly in the middle of New York City where you can marshal crew. Like how does that work that you, how do you get the people Yeah.
How do you manage your contracting contractors? Yeah. And, and marshal your forces. So like
anything, there are basically people that, that's the market they're in, that's the industry they're in. And so we use specialty contract workforces that they specialize in having these people. And [00:24:00] when these people come and work, they typically will work like 12 to 13 hour days.
Okay. And they will work 13 days on and one day off. Okay. So when they come in, they, on average they will be about an 84 hour a week, like week. Okay. And so they're, they make a lot of overtime. They're basically making more overtime than strict time. Yeah. And a lot of the welding crews only work on double time.
And so it basically, there's a lot of money. Got it. A, a turnaround welder might only have to do two, maybe three turnarounds a year. They might only work actually six months outta the year, and they might be putting down three to $400,000 a year. Wow. Yeah. Okay. [00:25:00] And so we pay for it. Yeah. That's my answer.
And you're paying them to work hard, hard. So it's like, I'm a welder. I fly in and it's basically like you're a mercenary at that point. Yes. 80 hour weeks if not more. Whatever it takes to get the job done. That's right. Wow. And so,
so for example, when we were working with BP in Toledo, um. I, I showed up a little late to the party and, uh, my, my, a good friend of mine was the turnaround manager, and he was trying to tell me what was going on.
I don't, I'm from Texas, so I don't have a ton of experience with oh man. What's the word for it? Uh, unions. Okay. Yeah. I don't have a ton of experience with unions up in Toledo. It's all union work. Got it. But when I went up to the turnaround, there's a whole lot of Texas license plates in the parking lot.
Mm-hmm. And there were a lot of issues because the union hall literally [00:26:00] could not deliver the number of people required to do the event. Right. And so we had to call an audible and reach out to non-union. Labor force. And uh, it made it quite interesting. There were, you know, rats blowing up on the side of the street and a couple fist fights here and there, but but it was, but we, we all got it done.
But it was, um, getting the labor force is very challenging. And actually one of the things that we are starting to focus on right now is actually how we think about staffing it. Traditionally, like you, you brought up a great point, honestly it's very difficult. And so traditionally kinda what we do is we'll build a schedule and when you tighten a schedule to meet a window that's called crashing the schedule, uh, to anyone in the capital world that's crashing a schedule.
And so, you know, we'll set our milestones like, well, we gotta start here. We gotta have it done here. How much do we have to get done Right? And do all these [00:27:00] constructability reviews. And, and what ends up happening is you have a real big. Where you're showing man within the first day or two, I need hundreds of welders.
Mm-hmm. And can the facility even process them through the gate and badge them that fast? Do we even have enough parking? Usually there's not enough parking and we'll have to run a lot and bus people in. There's, the amount of city that you have to like municipality type stuff that you have to make is worth.
Yeah. And uh, and so one of the things that happens is our contractors will typically say, oh yeah, I can do that. I can do that. I can bring you 300 welders. Sure. Overnight. That's what I'd say. Yeah. And I'm like, counting the dollars. Yeah. I can, yeah, absolutely. Pay 3000. What? Yeah. And, and these are really great partners of ours, and so they're trying, but what they, what we're seeing in the industry right now is that we're actually missing the ramp in by maybe 20, 30% they [00:28:00] will get to peak.
But they're like on a three day lag on getting all the way there. And then we're seeing some slippage in the back end of our events where they just weren't able to get enough done because they missed so many man hours early because they didn't hit peak. And so what we've done traditionally in the industry is we've basically carved out a scope of work.
We've approved a resource plan from a contractor, and then we just kinda say from the owner side, we just say, okay, cool. That's the contractor's problem. Now they owe us 20 boilermakers on this day and 15 welders on that day, like we approved it. But what I'm moving to now is more of a thought process of those 200 welders.
Well, those are open roles that the owner owns that we earmarked for the contractor. To fill. And so what we're tracking now is we're saying, well, what's your fulfillment rate if [00:29:00] seven days out you don't have a name, you don't have a, who's gonna be here? We haven't pre-registered them in our badging system or whatever.
Like if you don't have name, date, you don't have an engagement, right? No ring, no na, no date, no engagement, right? And so we're starting to track our fulfillment so that we can have those conversations because when you're in the middle of ramp up and everybody's going crazy trying to do work, you can't talk to anybody about, Hey, you're missing five people, but those five people really matter or they wouldn't be on the event.
And so it's, it's, we're now shifting the conversation further forward and earlier. So we're starting to track what we're calling the, the fulfillment of the resource plant. Earlier. What we used to do is just basically, Hey, I'm expecting 200 boilermakers today. And we would just wait until they report that they've got 200 here.
And so, you're, you're a hundred percent right. It's something the industry has to focus on. And when you're talking about resources, [00:30:00] another thing that's absolute critical is there are only so many people that specialize in these events. And in a weird way, all the different refineries have to kind of know what they're doing.
So there's some reports that tell you when different refineries are doing different things. And if all the refineries say in the Corpus Christi Ship Channel want to do their turnarounds at the exact same time, oh yeah, that's not, could work. They won't be able to. And so what you'll literally, at senior level, you uh, the turnaround manager will call another turnaround manager at, from Citgo to Valero and say, Hey, I'm gonna pull mine up a week.
You move yours back a week. Okay. And then. Yeah, I've, uh, it's, it's
a very challenging situation, the labor. But what about like, globally, right? Because aren't we seeing like massive structures going up? Turnarounds, you know, can be impact. Like turnarounds aren't necessarily always refineries, right? Like say you have a [00:31:00] nuclear facility.
Yes. Those have probably have turnarounds. So like what, what happens when, I don't know, Argentina spins up a brand new facility? Do they start drawing people from from the us or even like China, for example, but it's like, sometimes. Yeah. How do you coordinate that globally? Or do, is it like regional, like every sort of.
Community handles it. Yeah, it's
regional. It's kind of regional. You, you actually find the kind of Texas Gulf coast region to be a big supplier of the labor, labor turnaround, labor workforce for pretty much everywhere in the country. Oh, that's interesting. Um, yeah. Most, uh, most other countries around the world, they, they start having a more globalized workforce.
So in you know, in the, in the Middle East, in those kind of areas, they tend to use a lot of workforce from India. And so it's, um, kind of a mixed bag on how you get your labor force across the globe. But yeah, it, it's definitely, definitely [00:32:00] fascinating and you have to remember that these guys are.
They are there to make money for their family. And so we've been on events where we'll get extorted where there a whole piping division will find out about a job somewhere else, paying an extra dollar an hour. Yeah. Yeah. And they'll show us the paperwork and they'll be like, yo, they want my whole crew and these guys wanna make money.
And we'll have to say, okay, we'll raise your rate and we'll pay everybody an extra $2 an hour to stay. Sure. And so there's, it's a, um, it's a tough one. And because it's a high pressure. Yeah. And because you don't have the time, this isn't a large scale capital job where you can say, ah, well screw you buddy.
I'll find somebody else. Yeah. You don't have the time. And so it's a, it's a tough
Mercury & the $80M Mismatch
one. Well, so story time grants can you give us sorta one example of maybe a really great. Turnaround operation, something that you've been a part of that you thought was really well executed and then maybe [00:33:00] one that didn't go well and kind of tease out what, what do you think the causes of that were?
Oh man. So for me, the absolute number one cause of turnarounds not going well. Now there's, I'm gonna preface this. There are unplanned things that happen that are unforeseen and the timeline so important that you just can't, one time I was on an event and we're in a crude unit, and you have to remember you are sucking this oil out of the ground.
It's got every element that the earth has in it. We opened a tower in AC crude unit and. The entire tower, the trays and everything was sparkling, shiny, silver. We go in and we're like, no. No one's ever seen anything like it. And it's dripping from the trays. It's filled with [00:34:00] mercury, no mercury dripping from the trays in 120 foot tall tower.
We had no what? Yeah. What do you do with that? What? How do you abate mercury from a tower? So we had to fly in a specialty company to train crews on how to do this. And I mean, we had to stop the whole turnaround because where else is it? Yeah, like this is mercury. I mean, there was like literally four inches of mercury at the bottom of the tower and no one's expecting it.
And so barring those kind of things. One of the biggest issues we have is a resource planning mismatch, and one of the biggest exercise that we do is we build a schedule. Mm-hmm. That is an extremely detailed estimate, so literally a schedule will have nearly every weld, every bolt up, every action, and be [00:35:00] estimated for the man hours required to do it by craft and discipline.
Now a turnaround is executed by 40 to 60 contractors, so there are anywhere between 40 and 60 actual companies doing those scopes of work. Each one of them is supplying resources on A TNM basis. So it's on the own. It's the owner's circus, and it's the owner's job to ask for, I need this many lion, this many tigers, this many bears.
I need this many ringleaders. I, it's their circus. Yeah. And so one of the number one things that lead to poor execution is not overlaying the committed resource plans with the schedule. Okay? 'cause 'cause you'll the last turnaround I was on that was a major one with BP that went sideways.
There [00:36:00] were, they ended up getting a, it, it ended up leveling out at around a 0.8. And so if you've got 5,000 people working and your productivity factor a 0.8, that's 20%. That means that a thousand of those people could have stayed home. They could have been playing with their kid. They could have been cracking a beer with their dad.
They could have been fixing every pothole in Toledo. They got a 0.8 because their resources didn't their re committed resource plan, they were not able to overlay it against the detailed estimate in P six. Mm-hmm. Until a week before the turnaround. This should be done months in advance. And so what you found, when we finally were able, we got called in about six weeks before the event, and it was the first thing I asked for.
I'm like, where's the resource plans? Where's the schedule? Yeah. And they had them, but they [00:37:00] weren't coded to be overlaid. Oh man. You could not overlay 'em, right? Yeah. And so when we finally got it overlaid, we found that the contractors were just staffing level. And there were periods where the schedule is demanding 80 plus welders and look good, and like you needed 80 plus welders, but they only had 20.
There were periods where the schedule only needed, you know, on the end where it only needed 20, but they had 80.
Mm-hmm.
It was like, absolutely, we had a plan and we had resources committed, but nobody ever looked at 'em together. They just looked at it at bulk. So what the contractors did is they said, oh, there's 20,000 man hours of piping, and they just made a leveled resource plan.
They didn't ramp in and ramp out. They didn't realize that we were replacing an entire heater. And so there was a period where we didn't [00:38:00] need people, so they should have been on another unit. There was like an absolute mismatch. So the number one thing that you need to do in my opinion, in any project.
Even if it's fixed cost, lump sum turnaround capital, the number one objective is I've got an estimate. I put it in P six. 'cause now I know my time scaled estimate when I'm gonna do it. And then I need to know is my re committed resource plan IE, what I'm cutting a purchase order for, is my committed resource plan aligned with my executable estimate and my buck?
And uh, and you're just, with so many contract resources, it is very difficult to get that to line out and get it coded correctly so it overloads. And so I, in my opinion, like literally when you, you can actually, what's crazy is that when we [00:39:00] finally looked at the resource plan and the estimate together, you literally can see.
Overages and underages, and I'm gonna use underages. I make up words. Mm-hmm. Um, I, and my favorite word is, uh, when I'm highlighting something and I ask someone to review my highlights, I say, Hey, did you, did you review my high? And so, anyway, and so, you know, when we looked at it, I literally, the minute I got together, I said, holy shit, they are gonna spend $80 million more than they have to.
Mm-hmm.
And, and that it was absolutely true. There were times that we were fighting for our life trying to get enough resources to do the work. And there were other times that we had people just sitting on their hands. And that, that is the biggest, that is the biggest one is not having a matching between your resources and the scope work you gotta go do.
It's that simple.
Well, and it's really interesting because I think the level of detail you're talking about in a P six schedule, [00:40:00] oftentimes. From what I've seen on like major projects that aren't turnaround the people doing the, like, manpower planning, they're not using P six a lot of times. And so, yeah, I just I know a lot of really good scheduling departments that are super, tuned in to P six, but if you're talking about actually trying to connect those two systems, yeah, I, I could see a situation easily where, a department's doing all their like resource planning.
Using a completely separate tool and trying to merge those two together can acquire someone like you instead global to probably get, get the programming going to sort that out real quick. Yeah. Which is, uh,
and that, that's exactly it. I mean, and we've been, these are problems that we've identified and that we've been building solutions for.
And you know, it's one of these things where it's like, we know what needs doing, but at scale you need the right tools and you just, you almost like, like sometimes I like to, when I talk about like, 'cause we build technology and mobile apps for the, industrial construction [00:41:00] industry. And I mean, I'm deep in my bag right now in piping estimates and how to do that for construction.
It's amazing. But it's huge that I'll talk to people about what we should do. Then they're, you know, and I don't talk about the applications and the database. I just say, you know what we need to do? We need to record everything. Buy crap, buy discipline by unit, by area, buy system, buy package, and, and have our estimate.
And then when we do our resource plan, these be like this. And people look at me, they're like, yeah, you might be able to do that on a $5 million project, but we've got a month to execute. Like, how are we gonna do that? These refineries weren't built without equipment. They weren't built without a crane.
Mm-hmm. They weren't built without tools. And so if you wanna do something, we are human. Like we are human. We build tools, we build capabilities, we do whatever we want. We'll build spaceships and go to Mars. Like we'll do whatever we want. We're human. And um, [00:42:00] and so you just gotta build it. You gotta commit to, I wanna go to Mars.
Then you start building the tools and the spaceships and the things you need to get there. And so we, we pretty much that's what we do is we say, this is what we need to execute. Well, let's go build it. And so, you know, we've got some tools that use the industries, that kind of common matrix based resource plans that they use, and we kind of can ingest that and lay that out.
We've got tools that compare, you know, the actual unit rate cost breakdowns that then tie that back to our schedule. And so we build a lot of cool stuff that helps us do all this stuff that I'm, I'm talking about. And it's some of, sometimes it scale's the only way you can do it and um, and where it gets really challenging is that a lot of people will see a detailed schedule like what I'm talking [00:43:00] about, and they'll say, is that really necessary?
And what you will find is that you build a schedule, you plan to progress. Whatever you plan, it has to be progressable. Okay? And so who is gonna progress it? The person that I want to tell me what got done is the guy that did it. Yeah. And they speak in drawing number iso. I got line number X, Y, Z. Sure.
I've got piping, isometric, drawing number. This sheet one of five. And I have marked, my weld has got a designator number on it. This is weld A 1 0 2, and I completed it. And qa, [00:44:00] QC has checked it and it's approved and done and ready for paint and at detail in the field. The contractors have to track all of that.
They have to because every weld gets put into a, a, a turnover package, a completions package. Every pipe support, everything that gets built, gets qcd a weld, uh, you know, a four inch butt weld might have 20 steps that gets signed off. Is the fit good? Right. So I've cut, I've beveled, I fit. Is the fit good?
Yes. You can proceed with the well root cap. Weld. There's different, you know, PT X-rays X, Y, Z, we've got a post weld heat treated. Did that happen? Yes. They are tracking and did it, is it a three, you know, a two, two part paint system? You know, two coat paint [00:45:00] system? Is the paint prep good? Is the first coat good?
Is the second coat good? Is it signed off? Is it, does it require insulation? Is it whatever it is? Every single step is tracked in the field. Okay. I can ask the piping company at any time. Where are you at with weld? A 1 0 1. And they can tell me what the field can't tell me. Is what percent complete? Are you done with piping package?
Two. Okay. Hey, I'm the owner. I wanna have all the piping done in this little area. How far along are you? It doesn't work that way. Yeah. There the welder's gonna look at you like, I don't know. But weld a 1 0 1 is done, bro. And so the field is in the details. They live in the details 'cause they have to do it.
So the easiest [00:46:00] and best thing to do is to build your schedule to match the field execution. Sometimes we'll roll the schedule up to a document number. Sometimes we'll roll the schedule up to a line number. But all of the estimate details by drawing, by weld, buy bolt up, all of those have to exist somewhere.
And so we use those as rules of credit to drive our schedule tasks, but got it in a short duration. So that might be fine for a capital job. But in a short duration window, you just need all those steps in the schedule because you're so focused on productivity that I need P six to move those little Tetris pieces exactly where they need to go so that I can tell these guys, you need to make this weld this day and that weld that day on a capital job, you can kind of just roll it up to the line number and just kind of let it eat.
But in a turnaround, I need to be able to [00:47:00] move them where they fit in like grains of sand.
The Three I's & the Tables Turning
And this is the, this is maybe, okay, let's zoom up way high. I'm vice president of Americas at bp, whatever, and, uh, some big company, I've got some turnarounds I need done Now as a big owner organization, I'm not in the weeds like that.
So I'm listening to this podcast right now and I'm like, okay, well I like what tip does Grant have for me as an executive to have successful turnarounds? 'cause I know my organization doesn't really. We're not really detail like that oftentimes. So like how does a vice president or some high level leader, what do they do that can make these projects go well?
Like what's your advice to them? Oh man. Ah,
so the number one thing that a leader that doesn't have a lot of detail needs is confidence. When you have leaders in a [00:48:00] position and we can talk about how to build confidence in an event. 'Cause I, I deal with this all the time. And so one of the biggest things that happens when you don't have confidence is all of a sudden they've gotta go into a meeting and they're calling everybody freaking out asking for information.
Mm-hmm. Hey, when are we gonna finish? What was that date again? And ba ba ba ba ba, you know? And then they go into their meeting and they come out and then they've got a whole nother set of objectives. And they're waffling all the time because they don't know what they need. And so for these guys, one, during the planning process, we track budget change over time.
And it has a very specific, I if you, if you've ever seen this chart, you should go look into capital projects, budget change over time. And what they do is the chart the, the, the curve always follows the same method. And so you'll always see budget starts at kind of a baseline historical, when the team engages, it goes up and [00:49:00] this and that.
And you're gonna see this profile where the cost goes up. Then they do constructability and the cost comes back down. Trends towards historical, and these are management inflection points. So typically I teach them to look at these charts to understand if the teams are actually performing and doing the planning they need to plan to get to the readiness of the event.
Number two is they need to focus on the three is. Okay. And the three is in order, are information, insights, and interventions. And guess who I got that from? Dotch. Dotch, yes. Our mutual friend. There you go. There you go. That's awesome. Dotch Phillips with, and he's with Cenovus now? Bigwig. Apparently he just moved up again.
He's like CFO of business unit. Wow. 'cause he's amazing dude. Yeah, he's, he's a legit And so. [00:50:00] When you're at that high level, you have to ask yourself, are people coming to me with interventions, options? Mm-hmm. Right? And then what they need to do is they need to look at the process of how these numbers are given to them.
And they need to ask where the information is, the data. And that's the number one thing they can challenge their team on is, okay, great. You told me it's 20. Yeah. Why? And, and so when you have information is your data, right? That's all of the, the estimates and the numbers and all the data insights is when you put that together and you roll that up, and now you can say, Hey, your performance factor is a 0.8.
That's an insight. Intervention is the only thing that matters. What do we need to do to fix it? [00:51:00] And so that's the only thing that makes money. I don't want, I don't care if you come to my site and make a whole bunch of data. I might not even need it. I don't care if you come to my site and you tell me it's a 0.8 or your number's a seven in your little bi chart, it's a seven.
Good for you. I don't even know what that means or where it's from. And now that you told me it's a 0.8 or whatever number you put on in front of me, is that good or bad? And now that I know it's bad, all of that's useless as a leader, unless you tell me what the options are, how do we fix it? And so everything for them needs to be proposed as an intervention.
So if you're a leader, if you're C-suite. Your people are coming to you with charts and graphs, and they're telling you, it's 90 million today and [00:52:00] it's a hundred million tomorrow, and it's, it's this or that. At the end of the day, they should be coming to you with an intervention methodology. It should be we were at 90 and now we're showing a hundred million, $10 million increase.
What is my decision making framework that they, that the leader should be given? Hey, if you are dead set at 90 million, we got two options here. We do nothing and we burn an extra 10 million. That's my option as a turnaround manager. Just gimme more money. Or if we want it to be 90 million, here's what we have to go do.
We gotta cut this scope. We have to change a resource methodology. We have to do something. And so I would say the number one thing that I would tell a, a leader over these events is that if the people that are reporting to them are not reporting in an intervention methodology, then they [00:53:00] need to start digging.
Because if their people aren't bringing them interventions, then their people don't know what they're talking about. So the next step is great. You're not telling me interventions. You're only giving me insights. You're telling me it's a pf of 0.8, right? So the first thing is to coach the people that are reporting to you to say, well, what do we need to do about it?
Are we gonna accept the unacceptable? Are we gonna accept a 0.8? Are we gonna accept these things? Or what are we gonna do something about it? And so as a leader, people should be coming to you with actionable insights. That's what actionable insights means. It means I've got a PF of 0.8, but I can dig into the data and I can get a deeper insight.
I can not have the project level. I can say it's this crew, it's this ninth crew.
Mm-hmm
That we need to fix. We need more welders in this unit. We need [00:54:00] these things. If we wanna accomplish these goals or we can shift our end data, or we can spend more money, we can increase our envelope. So if people aren't coming to you with interventions, that's the first thing you need to coach them on is the classic.
Don't bring me problems, bring me solutions. Yeah. And if they can't do that, you keep digging. And then the number one thing you gotta do is you have to say, great, you're showing me an insight 0.8. What information backs this up? If I pulled up, you know, if you were an owner or Ryan and you had a project you wanted do for a hundred million dollars, this is your money.
It's a hundred million dollars of your money. And I said, oh, we're gonna have to spend an extra $20 million. Yeah. You're gonna have to dig in. And so these leaders have to become active. They have to be activated. And the only way they're gonna have trust is if they dig in. And so if your [00:55:00] team just says, well, it's gonna be an extra $20 million and you can't figure out where is it an extra $20 million in pipe?
Is it an extra $20 million? 'cause we're adding a new unit. If you can't dig into the details, then you can't have confidence. So if, if I have a Power BI chart, you know, you're, you're familiar with this. If I've got a, if I've got a chart that says it's a 0.8 and the people reading the chart don't know how it became a 0.8, are they gonna have the confidence to make decisions?
Mm-hmm. Off of a 0.8? Yeah.
How do you, so with your business, you know, you do a ton of data work. Mm-hmm. And then Paramount is visualizing that data to, to tell that story. And often you're trying to incite action, right? You make something bright red and annoying looking because you want someone to be annoyed by it and go do something about it.
How, when you're building up that data and you're doing [00:56:00] that, how do you connect people to, Hey, the number I'm showing you is a 0.8. How do you connect them to the confidence in the data that's driving it?
That's a great question. So I think there are a couple things that blows confidence in the data, in my opinion.
So number one is if your data's late, an executive will always remember that because it really says that the data is not trustable. 'cause if you can't produce the data on time. On demand. On demand. It's like, where is this coming from? Something in the 'cause data is like flowing. It's supposed to be flowing, and then all of a sudden it's just not flowing.
Like there's some sort of like intermediary editing Excel file, something going on there. So I think having your data on time is super important. And then I for me I sort of believe that actually data goes through almost like matter transition states. So it's like if you try to roll up from the very, very, very bottom [00:57:00] sometimes to the executive there's like rounding errors and significant digit things that can like, get screwed up as the data bubbles up.
And then that can make an executive feel like it's, they're not getting true rollup data. So I think you have to like, you have to allow and accept like almost like a transition state in my mind. So it's like data coming from your contractors. There may, there's like a wiggle room. If you try to roll it all the way from like, because it's like, what is reality?
So if you're, if you're trying to do a sum up of all of the time sheets of your contractor to like bubble it all up to like some grand number, there's so many little deviations and stuff that happens like almost at like the subatomic level that like those start showing up when you're trying to do physics at, like motion of planets or whatever.
And so I feel like you have to allow that a little bit of phase transition. Or else you'll get lost trying to sum up everything. Yeah. [00:58:00] But that doesn't mean you can't roll up the data. So it's, and then it goes back to what you were saying, grant about discipline and culture. To me, the worst.
Data culture is the vacuum culture where it's like, oh, you know, I'm gonna, eat up everything. I got this Excel from this contractor, I got this from that guy, and then you like, throw it in your giant warehouse and you just start producing data. And I actually see this not picking on finance, but I, I feel like finance oftentimes spins up data teams and they go around and like grab stuff from operational groups that they don't fully understand and then they start producing numbers.
And so I think analytics without the context of the crew and without like strong discipline Yes. Um, kind of results in bad analytics. And once you lose trust, it's over. Which goes to one other thing, which is I kind of recommend having like a stamp. So like, if you're the ops team or [00:59:00] you're, you know, you're the turnaround team or whatever, and you're presenting reports, I always like to have some sort of like.
Officiating thing so that when executives see, they know like, this is my data. So it's the source, like it needs to be understood. They have to own that data.
Yeah. You
have to own the data and you have to represent it and then, yeah, exactly. So
yeah.
I guess those are my, like off the cuff Yeah.
Feelings on it. But yeah, it's, um, it's super challenging man. And, and you know, and you will see you're, you're a hundred percent right and that's kind of where, you know, we started databasing a lot of information and one of the biggest things I tell my guys is I'm like, I don't want to sit there and doing pivots and v lookups and looking at somebody's spreadsheet if I can't consume that spreadsheet every other day.
Because what'll happen is you'll see these guys in these ad hoc exercises where they're taking all [01:00:00] these spreadsheets and then they're spending a day or two trying to make them all turn into some report, and then they're sitting there and they do that every day, every day, every day. And so, you know, what we've built over time is we have this, um, a little bit of an ingestion tool that basically if we've got a spreadsheet from the same person or the same system, we can kind of import it so we can do change management.
So when you import it, you can actually see, oh, there's like a staging screen. Here's all the things that have changed, this and that. And so what we end up doing is, the way we build is we don't collect anything like autonomously. We have people provide us information.
Mm-hmm.
And, and then they are reviewing it.
Here's what's changed, here's what's added, here's that, how that affects information. And then they're committing and confirming that. And by not just reaching in and grabbing information, we found like that people trust the data a lot more because they can [01:01:00] literally see, oh, Leah so-and-so uploaded this on this day and confirmed it and we know it's real.
And, um, but yeah. That's, that's huge man. It's, it's interesting. And so, a little bit of a topic shift if you're down.
The Hard Shift: Docs, AI & Stead Global
Yeah. Let's go. Lemme give you one more question and then Okay. I was just real quick for students. Anybody else that wants to get into the turnaround industry? You got some advice for them about what they should do?
I'm, I'm, I'm a Caltech or whatever. I wanna I like turnarounds. What, what would
you I have absolutely no advice. The turnaround industry would love to have your mind and your body and, um, and there's so many different aspects and realms you can do. There's, when you do a turnaround, you impact pretty much every single organization at the facility.
We ramp up environmental specialists, [01:02:00] we ramp up industrial hygiene specialists for abatement efforts, and we ramp up safety professionals. Engineering gets ramped up and so that they're available 24 7 process engineers are critical for going through the vessels and making sure, you know. Bubble valves and, and all that are functional.
There's pretty much not a single realm in the kind of industrial market that doesn't get touched. And so if you are, from technology as well, you don't move $10 million a day in approvals without technology. And, um, and so I would just say that if somebody's actually listening to the podcast, they can hit me up on LinkedIn.
They can connect to Grant. Yeah, they can connect with me and I will put them to work.
You just [01:03:00] made a lot of new followers and friends. Um, alright, let's hard shift, let's go. Where are you gonna take us Grant? I'm, I'm, I'm super curious. Hey,
document contextualization. All right. Okay. We build what the engineers tell us to build.
And so the engineers produce a ton of documentation on what we need to go build. And that often defines our scope of work. Okay. And they can get as fancy as they want. They can make models and they can, I don't care if they got a Microsoft Hollow lens on their face mm-hmm. And whatever they wanna do. But I'm gonna tell you, the welders in the field, look at an isometric drawing.
That's what they do. And these guys will [01:04:00] spend weeks and months, depending on the size of the job, going through these drawings. From structural steel drawings to piping isometric drawings to tubing layouts, to you name it. And they're going through this system and they are estimating, they're identifying what all these tasks are.
They're mapping these components across the drawings and the tools to help them manage that are almost non-existent.
Mm-hmm.
And, and, and it's absolutely fascinating problem. And so, for example, you know, we're, uh, you know, we're helping this, loves project in Nebraska and there are 8,000 piping, isometric drawings, 8,000.[01:05:00]
Now each of these piping isometric drawings has, anywhere between 2, 4, 5 spools on it. So it's broken up into spools. Those spools are sent to a fabricator. So now we're having to track at the details we are tracking and maintaining, well, what drawings have been done, what spools have been done, which ones have been shipped here, what dates are they gonna get here from a procurement standpoint.
Then in the field, we have to look at that piping circuit and say, how are we gonna erect it? Yeah. So we're going through and say, okay, I have to do a field weld here. Okay, well that's a four inch carbon steel schedule. 80 butt weld is now I know my man hours. I know how long it takes, and I know the, the, um, weld procedure.
So now I know all the qa, qc, I'm looking at that asymmetric drawing. Does it need paint? Does it need insulation? What is the paint spec? What's the insulation spec? Okay. [01:06:00] These are the details that a contractor that has to go do the work has to prepare. Now, if that contractor goes through every single drawing, how is change management affected?
So now, do they have that information catalog tagged and bagged in a way that when engineering sends rev one, that everybody's actually notified that rev matters to them or not? And how is that actually disseminated? So this is a, so just a little perspective, this is a billion dollar greenfield project.
So, I mean, this isn't the biggest capital job in the world. It's pretty out. I think this is a, it's a pretty good size, but I would say this is like a [01:07:00] normal mega sized project. Mm-hmm. It's relatively simplistic. There's a few units, it's a renewable diesel facility, and, uh, it's pretty straightforward facility.
And this the document management system that they're using is Oracle, AEX, Aconex. Or whatever. Yeah. And, uh, they're, they're really not using like every feature of it. Everybody's in there. Cool. Just basically using it as a hub to download the docs. They're mislabeled misorganized, like it's horrendous.
And there are 64,000 documents, 64,000 documents that make up that billion dollar facility. Okay. Mm-hmm. Um. They had a contractor in there and no names will be named for months, basically getting [01:08:00] paid to build these detailed work packages. 'cause that's what you need to execute. And then they were gonna execute.
They're not on the project anymore. And when you are looking at the information they had and what they were able to compile via Excel and looking at PDFs, horrendous, the amount of oversight, minimal the mistakes, egregious and wild. And they spent millions of dollars doing that and uh, really to no value.
And, um. And so I have been just head down trying to fix this and you know, the issues you see are just wild. And so when you're in this system a X and you say, oh, I wanna see all the piping, isometric drawings for this unit and only the IFC ones. Those are all mislabeled. So you pull 'em and then you realize you're missing four drawings here, [01:09:00] you're missing this, you're missing that.
How do you know that? And so we've been building a tool that we call Doc Ultra to solve this problem. And what it does is it uses some AI components. So we built thought you might be going here. Yeah, this is cool. We built, I had to like, I literally had to. Because it was so frustrating to be working with this subpar data that was manually created.
And so when we did this, what you do now is you drop the documents in and we call it BlackLine ai, but we're using some Google Cloud vision and it takes the title block of the piping isometric, which has got all your paint spec, pipe, spec line numbers mm-hmm. Document numbers, rev numbers, all that.
And it parses that into a good table. And, uh, and then we also automate the material takeoff into another table. [01:10:00] And then the way the program's set up is that you can click on the drawing and it pulls it right up in a view that you can zoom in and look at right with the table data that it created. So you can validate the title block and the MTO boom.
These other tools, you have to download the PDF and look at it over here, and then somebody else has to do the, it's a nightmare. And what we're finding is fascinating. We're finding file names for one document number, but then, you know, so for example, you have a pipe symmetric that's got six sheets. Six pages.
Well, we found one where all six pages were the file names. But the document was exactly the same. It's the same document. It was just page one, six times. Right. And these are things that now it's like, whoa, where is it? We ended up doing some digging. Come to find out we, they literally were never delivered.
So every single one of those spools were never sent to Fab. Yeah. Were never estimated. [01:11:00] And everybody's saying, we've got all the piping we need for that circuit. If we would've waited, then what? Yeah. Right. And so what we're finding. Is we're finding that when well guided AI is able to produce better results than what humans can.
And so in this is got an interesting situation where we're leveraging this tooling, but so many other groups have came before us that we have all of their outputs. Mm-hmm. And so we're able to validate at scale, not always just one drawing at a time. So we were able to actually process 8,000 piping isometrics, spot, validate and validate what our numbers were different from every other source in less than a week.
Wow. For 8,000 piping isometrics, we ended up with a material takeoff database that had almost a hundred thousand records in it. Fully [01:12:00] validated. Their procurement team is like tripping out 'cause they're like, oh my God, I didn't know I was missing this spool. I didn't know I had this. And like, and they're validating what we gave them and they're just blown away.
And, uh, and it's absolutely fascinating. And so what is crazy in our industry is that the engineering group can't give us a full material takeoff by iso drawing number. And they said, well, the client didn't pay for that and we can't do it now. And what's really interesting is I've only been on one project that the engineering team was actually able to give us an Excel file that was the material takeoff by drawing number.
It just has to do with the way they export the isometric drawings from the model and this and that. It's basically a really manual exercise still. Yeah. And um, and so because things are executed at [01:13:00] isometric drawing, we have to put our mind there. We have to say, we can't just say, you know, throw the goat over the fence and pretend we're a controls organization and just say, oh, good luck contractor.
Give me your schedule and let me know how you're doing. When we hand them a pile of bullshit, somebody needs to go through and validate and organize the data and structure things so that it rolls up correctly. Well, and what, what this reminds me of is so interesting is like, schedule validator, you know, is a new startup tool, uh, that Todd is, uh, CEO now, I think Todd McKendrick, but, um, they're just, they're going gangbusters, right?
Like, everyone's like validating their schedule. Of course, Oracle had some tools in P six before for that, and I, I feel like what you're talking about is like a doc control validator, which is like never been done really, uh, that I can think of. 'cause AI is really the necessary piece to actually creating good [01:14:00] validation around this.
Yeah. And, and you know, and when we was that, and when we did it, what we found is that it's extremely clear that nobody's doing it. And because the file name doesn't matter what folder it's in, doesn't matter. The document says exactly what it's. And we have found so many interesting, like discrepancies, misses in the way it's managed.
And you know, we, like I'm saying, we have found, because we're parsing that title block ourself and we're not looking at a line list or a drawing list from the engineer. Mm-hmm. We are finding just, and this is a major engineer using real tools, when you go into Aex, they've got, you can see they've got registered automation bots in there.
Like they're trying their damnest mm-hmm. But they're clearly not pointing AI to look at the actual document and finding out what they can learn. And, you know, and when you, when you do this, you look at a drawing number and it's got all this crazy [01:15:00] stuff in it. It, it's got degree symbols and it'll be like 140 degree F and then it'll have, inches and 0.15.
And sometimes it's got that, sometimes it doesn't. And when we're comparing to like all the other. Entities that are building data, all the contractors, things like that. They start shorthanding things. They use different document numbers. They're, it's amazing to see how the data that's being built off of these drawings is not actually tied to the drawing.
And so it just
will never roll up and function correctly. And this is where like, I wanna plug Dead Global because you've got this awesome team of software developers, AI engineers, backing your abil. And you're a genius. You're making some really great connections here in this space. And you're building the tools you started building onto the frontier.
That's right. [01:16:00] That's really cool. And it's, um, I. Time cut you off. I'm sorry. No, I was just gonna, I was gonna plug you some more, I was gonna say that, uh, where, where can people find you to, to work with you and hire you because, uh, everybody should be working with your team and, uh, I, global products
literally as of now without, uh, you know, any self pompous, I would go out on a limb and say, if you are actively responsible for building industrial piping, whether you're an industrial piping company or you are an owner that's responsible for piping getting put up in the field you need to talk to us.
Like we are doing things right now that are going to completely change the industry. I've been at this hardcore since 2019. I started in 2013 and this is the first time. [01:17:00] That I've built something that doesn't play in the margins or Orion. I mean, this is everything we've built has a lot of impact and ROI and does little things here and there.
But this is the first thing that I think really changes the game in how we do work and what we do. The last person I talked to, uh, Danny, he runs a, um, an industrial piping company and they do 50, $60 million a year worth of piping work. And I'm talking to him and he's just blown away. He's like, dude, we do all of our, all of our work fixed costs.
You're literally just handing me millions of dollars. I'm so excited. Like he instantly went off and went and told his buddy, and this is something that we're like actively building over the last couple weeks. And it is it's wild. And so, the way we work. And the reason that we're able to build what we build is we eat our own dog food.
We [01:18:00] build things so that we can use them. Mm-hmm. And I am actively using them. So I am in the saddle using it. So the reason that I am only sleeping three hours a night for the last couple weeks is because I am like, Hey team, guess what? I just got 450 piping isometric drawings and I have to make a resource loaded schedule with non-labor resources.
Go. Why? How do I do this? I am like a blind boy with no tools. What do I do? And then you watch the team deliver. And the other key component here is that, yeah, we are a development house, but all of our people that work for us work full-time. We don't use an overseas development shop. There's no part-timers.
Yeah. I've got a kid in Canada that I hired from a buddy of mine. I've got a kid in [01:19:00] Poland that makes the best power BI stuff I've ever seen other than them. Everyone's local and they're all in the same time zone and we can operate quickly and it with speed and build really cool stuff and have a real team and so when we go out and say we wanna do something, there's no red tape, there's no roadblocks.
We just start building cool shit and we pressure each other and we're sometimes we're just coworking together, we're, I'll have a meeting open and I'm just on my box grinding and people are just popping in and out. Hey, I built this. Hey, I got this. Go test this new thing. Resync your app. And that's how we build stuff that works once it's functional and executable and works.
Then I start showing people, I get them in the tools, they become part of the team and we build more and what they need. And and so we have a very hardcore like service mindset when it [01:20:00] comes to building technology. And so when people are using our tech, they get our service. And so, like I told our new client, we're building this piping app.
I said, Hey, they were concerned about man, well we already did the estimate and so man, it's gonna take us all this time to like redo that in your tool. And I said, you're right. You already did it. You don't have to do it again. We got it. I'll do it. Yeah. And so it's like, I feel like there's a lot of promise of technology and people forget that you still can't be lazy.
You have to fucking grind. You have to get in the chair and do the work. Yeah. And so many people are not pragmatic enough to do that. And they're not gonna be successful in projects. They're gonna spend five years thinking about it and they need to get in the saddle and get it done. And so you've got all these technology [01:21:00] groups that had one guy that had a damn idea, and they don't focus on those operations.
And that's like, I love talking to you about your, that's why we, we work so well together is you had a background in the whole managed service space. Yeah. And it's like, you know, there are so many people in technology that they're like, they love touting themselves as the lazy developer. Right? I only develop because I'm lazy and it solves a need.
And it's fake news. Like you need to manage your work.
Like
speaking of that, when we had, we had a, a core value, our number one core value for that company, the managed service company, was, uh, was we work harder than our competitors. And every developer would be like, that's the dumbest I, that's so stupid because we're not gonna work harder.
We're gonna like. Smarter sneak around the work. Yeah. Like, technology's gonna gonna leapfrog us, so we're actually gonna be working less than our competitors. [01:22:00] And it was like, no, no. Get this they're just as smart as you about that. You actually just have to, you, that's the baseline.
Now we have to actually put in the work. That's right. And it was a real mindset shift for some people, and some people never got over it because it was like, look, if you're always looking to get outta work using technology, you're never gonna get anything done. And it's, it's tough
being a demanding partner.
And I am an extremely demanding partner. I, I don't know if you've ever worked like teachers have this, uh, if you've ever worked fast food, fast casual, when you clock in, you're on your feet and you are not stopping until you leave. And you know, I get developers that sit in the chair and I'm like, and they're like, whoa, everybody's just working.
Like, we don't like, and I'm like yeah, we're working. That's what we're doing and we are gonna absolutely kick the door down and get, [01:23:00] get it done. And uh, and there's a lot of layers to that and it's, um, it's
absolutely fascinating. It, well, and this is why everyone should be working with your team.
And what I like to say about a company like yours versus hiring like Accenture or someone like that, is you know exactly who you're gonna get. You can go cut a million dollar check to Accenture, or you can cut a $500,000 check to Grant. Yeah. But with Grant, you absolutely know the team already. Like you just heard about 'em, right?
Yeah. Whereas with Accenture, they're gonna spin up, you're gonna get who you get. They can give you a case study of an amazing project, 100 of 'em. But you, you're never gonna really know if you're gonna get those killers in on your team. And so that's the nice part about hiring someone like you, grant is like, you know what you're gonna get, you're hearing it.
I
understand. And, and that and that and the, the, the issues there are fascinating. We're, we're working with a major right now, and they, we went and met up with them and had a little in-person deal with them. [01:24:00] And they are literally suing their last technology provider for under delivery.
And I'm talking to the lead over it and he's asking me, well, how do you support your tool? What's your SLAs? How you this, how do you that? And I told him, I said, my CTO, the guy that runs everything, he lives 20 minutes from right here. You will have his address and you can go knock on his door.
Yeah. And I'm like, and I'm available. And he like, the sigh of relief that you like was audible. And he started telling me, he's like, yeah. 'cause it was like the team was capable, but we'd ask them to do something, but they like literally didn't have the keys to make the change in production. And we would ask for something and it'd be like five days before it would get done.
Not because they weren't doing it, but just [01:25:00] because of the whole company. Yeah. And, um, and it's just, you know, so we, we spend a lot of effort really partitioning our data and making sure that we've got good developer access control on who's doing what and what they're developing. And, uh, and we're pretty serious about that.
And we're serious about all of our documentation. And one of the, one of the main things that, that we. We have here is that we don't build a technology and then tell our clients to conform their process our way. We build our technology around their process, and so we really eliminate because technology's amazing right now.
Yeah. And so you don't have to buy a tool and enforce everybody to use it. We build to suit and we build fast. Mm-hmm. And so, you know, there's, there's the uptake and the like, oh, well, people are fighting the change. [01:26:00] No, because I just talked to your piping superintendent, showed him what we're working on, and when he said, Ooh, can it do this?
I said, yeah. And then the next morning I met with him again and I showed it to him. And now he feels like part of the process, part of the development team. He's gonna have no problem disseminating this app to the people that have to use it and being the champion because we built, he showed me his spreadsheet, he showed me the reports he makes, and then I went and built a tool that helps him do that.
Yeah. And, and focused on him and what he needs to do. And then in the background, since we understand controls, we know what the data needs to dovetail into. He doesn't need to work. And, and so it's just a whole different methodology, in my opinion, in how you execute technology. So I feel like we're, I think of us as kind of a rogue IT group, but I don't know.
Close
Yeah, I like that. The pirates of the industry, but you're, you're breaking ground because really managers I think face two choices. They're like, oh, we need to buy a [01:27:00] product 'cause defining all my process is too complicated, blah, blah, blah, blah, blah. I'm gonna buy this product and it's just gonna do it.
Yeah. And then it's like, why does that always fail? And it's like, it fails because you're putting the tool before the process and the tool is actually like handcuffing you. So if your process doesn't match the tool, you have no hope. And so then you end up with these weird sort of optionality pieces.
Like maybe you, you buy like SAP or Salesforce and it's like that is a platform that you now lock in some charge to and you gotta build on top of that, which is way better than just the product. And then you can also just call someone like Grant that can build you what you need. Using the platform, tools, technology that you need to build it.
I really am a pro development future. I think that people are like, no code, low code tool, tools should lead process are missing the point. It's like, yeah, you can build a company, but it's like gardening without a spade. Yeah. Why would you do that? Yeah. Like why would you hamstring yourself? I've always been a [01:28:00] fan of, uh, of the approach you're talking about.
Mostly 'cause I believe technology's getting easier. It's um, I think a lot of people got burned trying to develop something, then it's just not that way anymore. There's, there's a lot more options out there. But this has been incredible conversation, grant, and I gotta thank you definitely for sharing your expertise.
Absolutely. Like talking about turnarounds was really illuminating. I, I don't know anything about turnarounds. I feel like you gave me a master's degree and in an hour and, uh. Yeah, if people wanna connect with you, they can, uh, we'll put the information in the podcast, but it's gonna be Grant stead on LinkedIn.
That's right. I assume stead global.com. Is that right?
Yes, sir. And I appreciate, this was my first time doing a podcast and it was, it was a ton of fun, man. And yeah, I
heard you've turned
down many podcasts in the past, so I, I have, I have. Thank you. Get I, I get a lot of inquiries and, uh, you're, you're the only person that I actually [01:29:00] have met face-to-face that then at later said, let's do a podcast.
So this
is the Gold Edition, the, uh, the original grant podcast. Maybe it's the Floodgates, buts and then the name of the product you're building again.
So for the drawings, so we, we, we have a turnaround specific kind of offering that we call Stoic, STO iq. Mm-hmm. And the applications that we're building right now that serve the capital space is called Doc Ultra.
Doc Ultra and Doc Ultra. Yeah. My, I let my developers name things whatever they want. So we've got apps like the Cost Reaper and uh, and the Piper. The logo They die. Icon they came up for the app is like a little piper lane of flute.
Yeah.
And so yeah, doc Ultra in the Piper revolutionary changing the game.
Awesome.
Wow. I looking forward to that. And also you should we are, we're actually gonna have a doc control specialist on, so I'll hook you up Leanne Gyer she used to work at bp [01:30:00] so that'll be a cool episode. Uh oh heck yeah. To kind of dovetail with this one for sure. Heck yeah. Uh, we'll have to, we'll have to give her a demo of Doc
Ultra.
We will need to do that 'cause we are having some fun with that and it is a game changer.
Well again, grant, thanks for coming on today. Really enjoyed learning a lot about your business and what you got going on in the space and turnarounds in particular, so thanks for your time today, grant. Thank you man.
Have a good rest of your day. Thanks for listening to the Major Project podcast. Be sure to follow us wherever you get your podcasts and learn more at the major project podcast.com. Until next time, keep building big.