Mike Nosbisch:  

Okay, good morning, or good afternoon, or good evening, depending on wherever you are in the world. My name is Michael Nosbisch. I’m a longtime member, fellow, and past president of AACE International. I have the honor of introducing today’s webinar from Contruent, a firm that a lot of you might know from their previous name, ARES PRISM. Actually, before that, they were, I think, the software division of ARES Corporation, a firm that I know very well from the past. I actually almost worked for them several years ago. So I’ve stayed in pretty close touch with one of the presenters, Karl Vantine. Karl’s actually the Chief Customer Officer of Contruent. And he’s going to be presenting today along with Meghan Russell, who is the Chief Marketing Officer CMO of Contruent.

Karl, like I said, I’ve known him for a long time. I’ve always thought of Karl as a very polished sales guy. But he actually, if you look at his bio on LinkedIn, he’s actually started in Project Controls, just like a lot of us did as well. Meghan actually spent about 12 years with Google. And that’s the perspective she’s going to be talking about in today’s presentation. She’s going to talk about how Google and firms like them actually process data and use Contruent software to support that. Karl is then going to take it over from Meghan and talk really related to an engineer and include a mini case study from a current customer as well.

At this point, I’m going to turn it over to Meghan. For everyone out there, I hope to see you next month in Chicago. Go ahead, Meghan.

Meghan Russell:  

All right. Thank you, Mike, for that wonderful introduction. So, before we get into the heart of the presentation, today, I’m going to tell you a bit about us and what we do. Karl, you can go to the next slide.

So as Mike mentioned, many of you are aware of ARES PRISM, but you may not be aware of the fact that we rebranded earlier this year as Contruent. ARES PRISM was founded 26 years ago as actually an internal solution. It was built by our cost engineers for cost engineers while they were working on a project for Los Alamos and NASA. Earlier this year, we rebranded as Contruent and launched Contruent Enterprise. We were very intentional with our name selections, so you can see the derivation of Contruent is true visibility into construction. And it’s a reflection of our mission statement, which is to empower decision-makers to build our world’s infrastructure with precision and speed.

The rebranding and the launch of Contruent Enterprise mark an evolution of the same award-winning project control software that has been building mega projects for 26 years. But now, with significant investment and new ownership, we’ve innovated and brought it to the cloud to deliver world-class capital project management software. We are headquartered in Naperville, Illinois, which we fondly refer to as the Silicon Prairie. We’re a global team with five offices around the world and customers in 26 countries. We work with industry leaders in infrastructure, oil, gas, mining, rail, energy, utilities, and nuclear sectors. Recently, we’ve proven that our software drives significant ROI for our customers. A few months ago, we brought in a third-party highly regarded in the software industry called Hobson & Company. They interviewed our customers and actually quantified the business impact of using ARES PRISM. From there, they built an ROI model that can be customized using project data and customer data to show the revenue impact. As an aside, if any of you are interested in seeing this model for your own projects, we’re happy to work with you to build that custom report for you.

So, we like a little bit of levity here, and we’ll start with a cartoon. I’ll give you a second to read it. And it’s funny because it’s true. I think we all know that measuring and reporting data is critical to improve decision-making and to monitor project progress and control costs. But the difficulty lies in making sense of this data in a way that tells a story and helps us make informed business decisions.

Today, as Mike mentioned, I’m going to walk us through an example of how Google, the world’s thought leader in data science, harnesses the power of data to drive decision-making and, ultimately, impact revenue. And I know it might seem unconventional to talk about Google during a cost engineering webinar. But, there are quite a lot of similarities between capital project management and Google’s advertising platform as it relates to data. After that, I’ll hand it over to Karl. He’ll take us through the critical elements needed to get value out of your data. And you should walk away from this presentation with concrete strategies to apply in your everyday work lives.

I worked at Google for 12 years as a product marketing manager. My last three years there, I spent working on their recommendation engine. That was Google’s answer to how to use data to make decisions. And I’ll explain more about the recommendation engine in a little bit. Google maintains its position as one of the most valuable companies in the world by mining data and making it accessible to not only its users but advertisers. So, I think we’re all familiar with Google Search, Maps, and YouTube, which, until Amazon dethroned them, was the number two most popular website. They also own Gmail, which, though it seems very personal to us, contains quite a lot of useful data. Google mines across all of their platforms to gain insights into user behavior, preferences, and trends.

With this unfathomable amount of data, they’ve developed visualization tools and storytelling techniques using interactive dashboards. They leverage machine learning and predictive behavior models, all with the goal of personalizing the user experience each time someone interacts with one of their platforms. So, I’m going to take us through an example to show the similarities between Google ads and managing complex megaprojects.

Karl Vantine:

I knew I would screw up the clicker at some point, so I was a little unsure.

Meghan Russell:

You can click it again and give me Randy’s hardware store. So this is Randy’s hardware store. And it is just a local hardware store in Virginia, actually, I think it’s near Karl’s house. And in this analogy, we can think of this as a very simple 100-foot beam bridge. This is the simplest example of a Google ad in the wild. Randy’s hardware store just wants to make sure their customers can find them. And in order to do this, they filled out something called a Google business profile. All you have to do is put in your information, your location, your hours of business, and boom, you are now searchable. And then the output of what they get from this is also going to be really simplistic. It’ll be how many clicks did we get? How many people clicked on call? How many people clicked on directions? So, this is the simplest example of how an advertiser uses Google ads.

Now, on the other end of the spectrum, is this powerhouse retailer like Home Depot? So sticking with our analogy, this is now the Goldstar Bridge in Connecticut. It’s a 6000-foot-long, super complex bridge. And, Karl, I’m going to put you on the spot. How much money do you think the Home Depot spends on paid advertising every month, one month

Karl Vantine:

$50, $70,000?

Meghan Russell:

Close, they spend three or four times that. They spend $295,000 every month just on their paid ads. So in comparison, thinking about Randy’s hardware store, when we think about Home Depot. So in order to execute that type of spend on a paid ad, the team at Home Depot needs to input massive amounts of data, so we’re no longer talking about a Google business profile or even a simple Excel spreadsheet. This has now become a very sophisticated strategy that they’re employing to reach the largest audience possible by using targeted advertising, calculating their ROI, and constantly measuring their performance.

In order to execute a strategy at the scale, they need a well-defined plan. The same basic tenets of planning that applied to capital project management essentially developing their project scope, defining milestones, and applying budget. So, these are the different types of platforms that you can advertise on Google.

And if Karl, if you build this slide, you’ll see that really, if you drill a little deeper, each one of these advertising channels is comprised of unique ads that require unique datasets to be created across hundreds of campaigns. And this is just a very small smattering of the types of ads that are within each platform. This to something that they’d have to create, analyze, and optimize. Within each one that requires they get so far, audience targeting, demographic targeting, geo-targeting, the creation of the ad copy, and the call to action. So they can meet the user where they are in their buying journey. They have to set a budget across every single campaign and then allocate their funds. And it all needs to be done on a very large scale. Similarly, this would be like developing a budget that outlines your costs of labor, equipment, and materials.

So then it’s massive datasets. Everything has been input. And now, it’s how we use data to tell a story.

Meghan Russell:

And one of the most powerful ways to make complex information easier is through data visualization. So by presenting data in a visual format, it allows the user to quickly identify patterns and trends that might not be obvious in the raw data, and thus facilitates better decision-making and more effective communication. So just like a work breakdown structure, users can then drill down into specific data points and look at it from different perspectives to uncover hidden insights, patterns, and trends. 

This is an example of the Google Ads dashboard. And anyone that’s worked with a dashboard, it’s going to look similar to anything you’ve seen. So we’re looking at performance. Is it trending up or down? You see, you can see some red and some green, good and bad. Which campaigns are getting the most traction? Where are the users located? We can get into location performance. And then, as you continue to drill down, you can slice this data in a million different ways.

And so, if we move through this, it’s now how do you tell the story of it. And I think we’ve all been there. You’re in a dashboard, you’re clicking through, you’re slicing it, you’re absorbing it, and you think you’re starting to understand the trends. And then you pick your head up, and you say, What am I supposed to do with this now, I have all of these insights. And I’m not actually sure how I’m supposed to move this forward. And, this is the same exact problem that afflicts every major organization.

And Google’s response to this was their recommendation engine. And they launched it a few years ago. And it’s essentially like, how do you aggregate massive amounts of data and make it useful, and we had advertisers like Home Depot, not to pick on them specifically, but investing hundreds of thousands of dollars with agencies dedicated to it, analysts, marketers, and they could not make heads or tails of their performance across hundreds of campaigns with thousands of ways of slicing this data. So Google threw their engineering heavy hitters at it. And they applied machine learning principles to essentially scrape every account for performance opportunities.

As you can see here, it says the account score is 85%. And then, everything below that has a personal recommendation, that would get them on a path from 85 to 100%. And it was interesting because the engineers that built it, one of them actually built Gmail as his 20% project. So we’re talking about very, very brilliant engineers. They initially thought it should just auto-apply the recommendations that an advertiser wouldn’t want to dismiss them because they’re right. And I think this is where we’re at with AI. It is an incredibly powerful tool, but it still requires human judgment and discernment, based on years of experience, to make the decision to apply that recommendation or not. And in the end, they did allow it to be dismissed or applied. And there are many good reasons to dismiss it.

And so, I tell this story, one, because it’s just kind of interesting to see how other industries are handling their data, but also because it’s so kind of powerful to think about how this will someday be adapted in our industry and how we can use this to improve project controls, and namely, our software. And so, with that, I’m going to hand it over to Karl to find the alignment within Project controls.

Karl Vantine:

Thanks, Meghan. Thanks. Before I press ahead, I just want to thank Michael again for the great introduction. I want to thank the AACE community for joining the call, and for listening. As Michael said, and Meghan said, I have been with Contruent, formerly ARES PRISM, for close to 20 years now. I’ve been in the AACE community for just as many years, and ARES and now Contruent, we’ve been sponsors of this group for a long time. I’ve been to conferences around the world. I’ve worked with a lot of customers, and I’ve joined some of the local meetings over that time. So it’s just a great opportunity, and appreciate you having us here today.

So, I’m excited about where artificial intelligence will go in Project controls in the future. Interestingly enough, what’s on the screen right now is what comes up if you search project controls and how far it is, or artificial intelligence rather, and how far it is in Project controls in Chat GBT. So there are areas that it can help it can. It can obviously eliminate some of the planning biases where humans are overly optimistic, it can help with quality control and some estimation tasks.

But what it can’t do yet, and this is where you will come in, and there’s a lot of you’re on the call, there are decision-makers, there’s technical folks, there’s executives on the call from around the world. The system can’t make decisions yet, just like Meghan mentioned in the Google world, which by the way, I think is pretty cool stuff. And I appreciated everything I learned about Google from that. It can’t engage stakeholders. The AI can’t do client-affiliate conflict resolution yet. It can’t deal with customer and supply chain issues and how to resolve those yet. Someday, maybe. And we look forward to that. But we’re not quite there yet.

So what we’re going to talk about today, and what I’m going to do is try to share some experiences that I’ve had around the world and we’ve had around the world, in how do you get value out of your data in project control? So we’re going to start off with a little poll. And we’d like to ask you all on the phone, what’s your biggest challenge in utilizing data for effective project controls? There’s four answers on the screen: lack of data organization structure, integration of cost and schedule, keeping your arms around change, and reporting progress and making informed decisions. Now, while we wait for some answers to come in, we’ll give it 20 or 30 seconds. 

As Mike said, I have been in sales, but I have been a practitioner. I have worked on projects. I’ve implemented systems, I’ve done data and process consulting. And while it’s a big world, and there’s a lot of industries represented on this call, there’s a lot of the same problems, a lot of the same problems everywhere you go. So we’re just curious to see, see your results.

Oh, well, this is a good answer, very leading. So I think everyone can see what’s on the screen. But we had 42% of the attendees say that lack of data organization and structure is the biggest problem, with integrating cost and schedule being the second biggest problem. And then changing efficiency as third and fourth. A lot of my presentation from here forward talks about data organization and structure as well as integrating cost and schedule. So I’m glad we picked the right things that will be interesting to you.

So some of you may have seen some variation of this chart out there in the world. But I’ve been on projects, whether you’re dealing with a small project or a mega project, building a bridge, building an airport, building a new mine site, or a power station, there’s a lot of data, and data overload does no good for anybody. So I’m often asked, and when we do system implementations and engagements with customers, where do we start? How do we start dealing with all this information? And how do we get it into the right bucket so that it’s sorted and arranged so that it makes sense when it’s presented? And it comes with a good story? Well, the way to start, the answer is surprisingly simple. Think about the end and where you’re headed with your data, and then work your way backward.

So, what do you need in terms of reports? What kind of stakeholders need to see the reports? What kind of decisions do you need to make from your data? Then, you can begin to build what buckets you need to put it in and how you should arrange how you should present, whether you’re presenting detailed reports or high-level dashboards. So, it’s starting with the end in mind to avoid that data overload.

Now, in our world, one of the things that I wish I had created this sentence, but it’s come from a number of folks over the years that I’ve worked with, a project is a project. So again, whether we’re building an airport, or a power station, or a bridge, you got a lot of data to deal with. You need to train your resources, and a lot of the challenges are the same.

You break it down into parts, and you start building your plan the engineering lifecycle across projects. You could argue that there may be some variations on this theme. But for the most part, the engineering lifecycle is going to be the same around the world. So that’s one way to start to think of bucketing your data.

The integration needs around the world, project after project, after doing this for 20 years, are the same. You need to figure out how to get budgets and actuals, and procurement commitment information into the same environment. We need to figure out how to integrate schedule. We need to get progress. So data warehouses have become a big thing. Now the new term is common data environment. But it’s the same integration needs everywhere you go. And the goal is to avoid these same things. We want to avoid mistakes. We want to avoid rework. We cannot afford to have multiple versions of the truth, or you lose credibility and you lose control of the project. Can’t do it manually in Excel anymore. The projects are too complex, the stakeholders involved are too many, that we just can’t rely on old tools.

And most importantly, and this is something that I confess I didn’t appreciate fully earlier in my career, but I’ve begun to appreciate it more now in later years, is you have to connect functional value with business value if you’re going to be successful in controlling your project. I came up as a detailed functional guy. I liked thinking about things like schedule integration and earned value management, and cost management. But the executives that were involved—and you need to have executive stakeholder engagement to solve problems and to escalate issues—they don’t think about the functional value. They think about business value: how am I going to optimize project spend and performance, how is our business or our project going to optimize our spend to control costs or increase revenue, or get a project done? That’s the business focus. So you have to be thinking of both as you’re building the output.

Same situation with data warehouses and common data environments and reporting. That’s the functional lens, which is hugely important. And you have to get that right as you’re organizing the data and putting all those Legos in the right box. But you also have to be thinking about how you’re improving visibility to decision-makers and that the data that you’re putting in the right box also facilitates the decision-makers so that they’re getting the right information. They need to pick which direction to go on the project.

Need to manage contracts and contractor activity. Now there’s a lot of functional capability involved there. There’s contract management, there’s bid analysis, there’s RFP management, and you need a system or process. It’s not just systems, it’s processes and training for your teams, to control all those different elements, but at the high level, you have to connect it to what the decision-makers and the stakeholders are going to care about.

And the last bit, of course, is change on any project. The minute you have the perfect plan, and you start, it changes, so you have to have change control, you have to auditable change, you have to as you’re building these Legos into a story that creates that result, you have to connect the functional and business value to really, to really get what you need out of your information.

So let’s switch gears now. I’m going to talk about three things that need to be in place to get value out of your data. There’s four, there’s five, I’m sure you can come up with other examples. But in my experience, these are the top three things. And conveniently, they matched your poll results for the folks on the call.

You’ve got to get the data organized, and you got to get it connected. So there’s more to it than just putting it in the right buckets; there’s connecting it so that you have a single source of information. You have to have effective measurement and reporting. And you need some judgment. Not everybody who’s going to read the output that you produce with your data has the same level of understanding. There are some hardcore, heavy, highly capable people on this call who really understand cost details. And then there are others, maybe like me, who need a little bit more spoon-feeding on how to interpret the results. So you’re going to add that judgment, you’re going to add some context to your results. 

So sorting and organizing, let’s take sorting and organizing, again, you want to optimize your project spending performance. We’re going to talk about coding structures, cost schedule integration, and change management, three of the most important things. 

First, you have to arrange the data. One way that we approach this, and I think we’re using pretty common terms that most folks on the call should be familiar with, is you’re dealing with estimates and cost and schedule and contracts. So these are three different disciplines, and each set of stakeholders has its own data reporting needs, their own management needs, their own way of looking at the information so that they can do their job, have to get them connected, you have to identify a common integration point, and so we call that a control account. Think of it as a work package, if you will, but you want to get this data into a stream where you can understand by looking at a control account how it relates to the estimate, how it relates back to the base schedule, and how it relates to contracts. 

This is hard stuff to do. Okay, this is it looks easy on a slide with a couple of colorful bubbles. But it’s hard to do. It requires workshops where folks from these different disciplines are engaged in Workshop. It forces discussion. It breaks down walls in silos between functional groups. It forces stakeholders to get engaged. Why is that important because if you get these people in a room and you survive the workshops, and you come up with a good agreed control account structure, then you get better adoption of the results, you start to get the reports that people care about, you start to get people using the processes in the system that you put in place because they had a vote, and a voice in how you came to that final conclusion. 

So as you’re developing a control account and a structure, you want to insert some intelligence into that coding structure so that as you read those, the series of 16 or 20 digits, it’s fairly easy and fairly quick to understand. As an example, which project or which contract you’re dealing with? It could be which engineering group, which, which geography, which owner. There are a lot of different ways to configure control accounts. This isn’t prescriptive; it’s just an example. The idea is that if you get all the right folks in the room, you’ll get a result that people will buy into. And by the way, this will support your reporting output. So it means something to the folks who are reading it. 

So, one example of how you do this, and an example of showing that it’s hard. The second-highest challenge that folks listed in the survey was connecting cost and schedule, and it’s hard to do. A lot of folks try it in a scheduling system. Some folks try it in Excel. But it’s hard to do. I mean, a cost-loaded schedule, which is a good way to start a project, most projects I’ve worked on have started that way. But it’s hard when you have to get into multiple budgets. So if you have a joint venture, for example, or if you’re getting into managing a finance budget versus a project budget, it’s hard to track multiple budgets, it’s hard to deal with things like contingency, it’s hard to deal with differing levels of detail and maybe intangible costs. So it’s a good starting point. But it’s not the only way. So as you’re thinking about those Legos and you’re thinking about how to organize the data, you have to make sure you’re picking the right tool that solves the right need. And you have to have options; it’s not always going to be the same answer on every project.

So, an alternative, and this is one alternative, and there are others, is that you time-phase your budget instead. So instead of having the schedule, like Primavera or Microsoft Project, be loaded with all the costs that you try to keep up to date every month and every year on a 10-year program, maybe you get that setup once as a starting point. And then, you move into an easier system that focuses on cost and lets the schedule do the schedule. So if you want to deal with non-budgeted costs or changes, you want to deal with performance status, you can deal with that over here in one system. And you can let the schedule be the schedule system focused on performance and progress and percent complete and critical path and the things that it was designed to do. You have to have options and alternatives as you’re doing this data structure setup.

And then, life’s not fair, to quote my boss, no matter what you do, no matter how perfect your plan is, once you start, there’s change. And on some of these programs on Mega programs, I’ve been involved, luckily enough, in a lot of mega programs over the years, from the Dubai World Fair to the high-speed rail program going on to Toronto Transit. They have tons of changes, hundreds of changes per period, sometimes thousands of changes at a time, small changes, and large changes. You need a framework that accommodates different change categories so that things move through approval and review processes quickly and they don’t get blocked up and held up at the wrong level.

There need to be escalation rules so that folks know and understand when a change has to be approved at the project level or when it has to maybe go all the way to the board because it has a material impact on the budget. So you need a change program that accommodates these things. You can’t just have a list of changes. You got to have the right reviewers, the approvers, folks who understand what the impact of accepting a change or the impact of the change what it has on other areas of the program so back to connected data and why it’s important.

So that’s the sorting and organizing. Now, you have to measure and report. So assume you’ve now gone through those workshops, you’ve set up your information, you’ve got a good plan, and you’ve connected your schedule and costs. You want to make good and informed business decisions. You have to have quantifiable objective measures, and earned value management is one that is popular in the US. It’s frankly popular everywhere in the world, but it’s not always called Earned Value Management. It’s sometimes called performance management, but a rose by another name. So you have to have that quantifiable measure and the progress report and, even portfolio reporting so that you can see from, from the top view, what’s happening, what’s happening at the details.

So, let me just talk about what I mean by that. Progress needs to be objective and quantifiable. I’ve been on hundreds of projects where all the performance and all the progress measured every month come from a schedule. And you ask the scheduler how they got the information, and they were 75% this month and 76% the next month, and then as you get to the end, you start running out of things, and you become 89.5, 89.6. And it’s not necessarily measurable or objective. It becomes kind of subjective in a lot of cases. So the schedule is sometimes exactly the right place to get the measure from, but not always. So, if you’re dealing with engineering drawings, for example, you probably want to get it from some defined rules or from time sheets around the drawings.

If you’re dealing with the supply chain and you’re connecting your procurement and contract activities, you might get it from detailed contractor pay items and from a visual inspection of the site. See, it’s not just making sure that you have objective progress. It’s getting it from the best source so that it’s reliable. And you’re not just doing manual progress updates because for folks on the call and there, and again, I know there are a lot of technical folks on the call, the whole idea of earned value. If you get your progress wrong, it all falls over. So, the progress is the most important, important part of an earned value, or any, frankly, any performance measurement system. If you get the progress wrong, your forecasts are wrong, your reports are wrong, and you’re giving wrong information to your stakeholders. So this is one of the most critical factors to get right.

As you report to make it effective, you have to be able to roll up and down throughout that chain. So we talked about a coding structure, we talked about control accounts, work breakdown, structure, cost breakdown structure, you have to be able to move your reports, depending on who’s reading the report, so that they get the right kind of output that suits their specific need, has to be auditable. If it’s not auditable, and you can’t tell where the problem started or when it started, then it loses its value. So it has to be auditable period after period after period from the beginning of the program.

You have to stay on top of budgets, changes, and trends. One of the great things about keeping control of your change and having good performance measurement is that you can then start to track trends and forecast where things might be going. The sooner that you get that information, the sooner you get a forecast that says something might be going off track, the sooner you can make a decision to course correct. So forecasting future expenditures and financial trends is a big part of it. And all these things need to come in place. And I know that for some folks, you might think, well, this is kind of basic stuff. But I go to project after project, where the report just has an actual view. And that’s it. And it’s a rearview mirror that focuses on what happened yesterday, and it doesn’t forecast what’s going forward. Six months later, people wake up with a problem and don’t know how they got there. So you have to be thinking about how you’re going to do the forecasting and performance management.

And it’s helpful to have an enterprise view once you can have lessons learned, too, so you can compare project performance and see why one might be doing something right and others doing something wrong. So you can figure out better solutions and promote communication. But maybe most importantly, because the folks at the bottom half of this picture that are in the project world have a different set of needs and a different set of information that they need to do their job than the folks at the top, the folks at the top are focused on the business value. So they want to see a different type of report than the folks that are managing it the functional value, or lower down and down on a project and a different understanding of what they’re consuming.

So let me pause for a minute. Let’s do another poll now. So before we talk about how we turn that data into a story, how confident are you in your datasets right now and your ability to create a positive story so that your decision-makers can make good informed decisions? Very confident, kind of confident? Let’s hear your feedback on this one.

Okay, so we have a few folks who are very confident, which is I think, very, very encouraging—14%, almost half somewhat confident. And I’d be interested at some point in that during the end of the call when we’re taking questions or maybe down the road if you want to contact this to understand why somewhat confident, where the confidence comes from, and what challenges you’re dealing with. And then there’s about 20% of you that are lacking in confidence or not quite sure you’re getting the right output from your data.

So let’s move ahead and talk about what you do now. So you’ve got, you’ve got this information set up, you’ve organized all the Legos into the right box, you’ve figured out how to measure performance, you’ve got a change process in place. So how do we start to apply judgment to that data to make some informed business decisions?

Well, you got to have smart people on the project, who understand what they’re looking at who can interrogate the data at different levels, you got to give it context, and you got to tell the right story.

So there’s a theme that you’re seeing here, of course, that you got to, you got to make sure that people understand what the story is, and what the business objectives are. So interrogating detailed data. I picked these charts very deliberately, so SPI and CPI. These are metrics that I’m sure folks are very familiar with—scheduled performance and cost performance indices.

You want to be able to drill down, so if you start getting a report like this at the portfolio level, where for each project, SPI is all great, it’s green. CPI, I’ve got two greens, a couple of yellows, and a red. I want to understand what’s going on? Where did that red come from? Why, what’s driving that? 

So, you need to be able to drill into the information. This reinforces that need for connected information and a control structure that lets you drill down into the details. And you’re gonna go sometimes all the way down into individual control accounts, and the slides not big enough to show it. But often, you might even go further. You might go back to the base schedule and want to interrogate: did we schedule wrong? Did we have a bad estimate? Did we underestimate how long it would take to do work in a hazardous area as an example? So you need to have those links, not just to what’s on screen, but all the way back to that base estimate to that base schedule, even to the engineering design. Maybe the engineering design was wrong.

So, having that information connected and available to you in one place so that you can take it back to the source is a really important part of getting the story right. It’s not just that you have that point nine, five CPI. It’s why did the contractor show up late? Did something go wrong? Or did we plan it wrong?

You have to apply some judgment and experience. So what I want to convey on this slide, this is all kinds of motherhood stuff. This is textbook. CPI of .95. If we quizzed you, I’m sure everybody on the call would know that that’s not, that’s not great. That’s less than one, and you want to be one or above. Whereas an SPI of 1.23, you’d read that and say that’s brilliant. We’re ahead of schedule. Everything’s great. But is it necessarily bad that your CPI is down? You have to have some context of what’s going on there at a combined level. Is this a good or bad story? Are you in one period right now? Or are you looking at an entire project performance?

Maybe you’re in one period, and you deliberately spent a little bit more money to catch up on your schedule. So, you got exactly the result you planned for. Maybe the schedule is more important to you than the cost? And so, again, a very deliberate decision could have been made. So you have to have that context. And how do you get that context? You have to train your people. They have to understand the business goals, not just the individual slices of data and the functional component they’re working on. You have to have communication from the top down and throughout the team. So people understand what’s going on. And that’s hard to do.

You need to tell the right story, then with the results to the right audience. So again, and I know I’m repeating the theme here, but at the detailed level, you’re going to be reviewing very detailed reports with your project team. Whereas at the executive level, you’re looking at dashboards, and you got to add some context, maybe some descriptions or explanations of what some of those metrics mean to the cost engineers, but might not mean to my boss, I’d have to explain whether that’s good or bad or why. And maybe more important than that, you have to give options so that the decision-maker can make a decision and choose a path forward.

So, a lot of what I just talked about is textbook material, but let’s apply it to real life. Let’s talk about those three elements on a real-life project.

So as Mike mentioned, we do have a case study, we’re going to use one of my customers in the UK, National Highways. And guess what? They had the same challenges that a lot of you have, they couldn’t figure out how to integrate cost and schedule. They had a ton of spreadsheets, they had inconsistent data, lack of accountability and control. So it should sound familiar because it’s the same kind of problems we encounter pretty much everywhere we go.

Think of the scope of their problem. This customer’s data journey was enormous. They had 26 billion pounds worth of projects, 97 infrastructure projects ranging in size from relatively small, under 30 million to billions of pounds in size, and it’s roads and smart highways and bridges and tunnels. So there’s a lot of complexity, a big supply chain, a lot of interactions, five-year investment schemes. So we’re measuring information against government funding. We’ve got portfolio reporting needs, operational reporting, and weekly, monthly annual metrics. We got to integrate not just the schedule but the finance and the accounting information from Oracle and changes everywhere you go.

So how did we deal with this problem? And what did we do for them and help them with in order to go from all this data and this huge problem to a single story?

Well, a lot of what we talked about earlier is exactly what we did for this customer. So we started with, how are you going to do control accounts. And we got a lot of people in the room. And we brought all the stakeholders together. And we helped them build a control account with some information in that control account that makes sense to different stakeholders. And I’ll point out that the PCF stage, one of the reasons I picked this example, and it makes sense, is that it’s referencing a project controls framework. That’s a good thing to reference a standard, a worldwide standard, because then that helps people understand it, it helps with training your resources, it helps people agree on where something is and why it’s there.

Also, in this structure, we included things like the contract number, work elements, and even a cost breakdown structure with various levels, and the cost breakdown structure again, used some defined standards. So now we’ve got a lot of information across 26 billion pounds of scope in a system where we can understand which buckets each set of data fits in.

We had to get it connected with a schedule. So coming back to that theme, how did we take all this information back here and connect it to our schedule? Well, we took that control account, and we began to layer it into the Primavera schedule through activity codes. And if there are schedulers on the call and I’m sure there are a few, we layer these into the activity codes, and then you get an automatic sync where you can tie up the schedule information with the cost information, monthly, weekly, daily, and real-time even if you need to, might be excessive, but you can. You can do it in Microsoft Project, you can do it in any scheduling system, but it’s about connecting that information.

And then we made them. So we helped them make some decisions about which information from P6 should come across to the cost system. So in the blue box there, for the initial performance management baseline, we started with the schedule cost. Why not it was a good place to start. But as I said, too hard to maintain month in month out. So then, for the monthly reporting and for forecasts and percent completes, we pull it from the schedule, but we don’t pull the budget information or cost information. We just bring the progress across. That’s one example: we integrated a variety of systems. They’re not just P6. We did the Oracle Financials and other integrations as well. But this is one example just to show you how we tackle that problem of getting organized and getting the data across.

Then, we had to turn it into reports. So this is all the data is grayed out. There’s nothing proprietary in here. But this is an example of a very detailed report for the practitioners and for the folks at the functional level that need to understand, estimate at completion and cost and schedule variance and very detailed analysis to complete estimates, things like that. This is great for one audience.

But again, we got to take it, and we got to make it relevant for a more senior audience who doesn’t understand or need to understand the details in that performance report. So all of that same information flows up using that structure. All the control structures we talked about up to the portfolio level in Power BI dashboards, which is what’s on the right, and it’s a little bit of a dull report because the section to the right with all the fun data is whited out on purpose so that you can’t see the information, but it’s all run up into Power BI all from the same data warehouse, all single source of truth and all connected. And we helped them do that and it took months of workshops and lots of stakeholder engagement to get agreement. But then even that goes a step higher.

So all this information from that detail level in the end makes its way into the Department of Transport, reporting to the Office of Rail and Road. Now they don’t look at those detailed reports. They don’t look at the Power BI dashboards. They get an altogether different view in a written report with context and with some written explanations and things that then translate what we focus on in our cost engineering world to plain English. And it ends up in that report for DOT. And that’s an important part of the process is knowing the audience and knowing who’s going to be consuming your data. So you give them the right story.

Now, we’re winding down towards the end of the hour. So let me just say this, there’s a lot more that went on in this program. We’ve done this kind of work for hundreds of customers around the world. And it does always involve training people and engaging stakeholders, and getting their input. We do it well, and we’d love the opportunity to do it for you. We are faster to deliver because we’re an out-of-the-box system with 26 years of experience doing this. And for the folks on the call that are various levels, whether they’re practitioners or executives, we do make sure you connect your four business values of optimizing project spend, improving visibility, accounting for change, and managing contracts, we connect that to your detailed data so that you get the real business value and functional value out of your system.

So, we have the cost tracking, we have schedule integration, we can help you with that time-phased budget. We have dashboards built-in. We have a data warehouse and common data environment report writer to improve visibility. And we’ve got a full-blown change management capability with workflows and audit trails, that can improve that transparency and help you stay on top of all your project decisions including cost and contractor activity.

And with that, we’d like to thank you for your time. Point out again, thank you AACE for this opportunity. We are the premier sponsor in June in Chicago. That’s where we’re headquartered in that area. We’re at Booth 304. And we’d love to see you there. Thank you.

Meghan Russell:

Thanks, Karl. We have quite a few questions. And they’re really good. I’ll start with some of the tactical questions. And then we’ll work our way up to the philosophical questions around AI and its disruption. So tactically, one of the questions is do you load the resources and costs at the activity level or at a higher level, using a level of effort in P6?

Karl Vantine:

Well, I used to be a scheduler a long time ago. I won’t pretend to be a detailed scheduler anymore. But I would say whether you load it at a high level or low level is less important than that you do it consistently. If you’re gonna go high level, stay high level. If you’re gonna go detailed, go detailed, but you need to be able to compare apples to oranges. So you want to apply consistently. Level of effort activities and loading resources against level of effort is, frankly, a hard thing to do. In a scheduling system, it ends up as one line that becomes kind of a junkyard for costs. So not sure that’s necessarily even the right place to do it. But that could be a long, fun conversation.

Meghan Russell:

Okay. Oh, can you integrate data from previous projects to analyze forecasts in your current project?

Karl Vantine:

How do you integrate data from previous projects?

Meghan Russell:

And can you?

Karl Vantine:

We can, sure, I mean, we can absorb anything. So referring to our system, specifically, we can onboard all kinds of information and data, we’ve gotten an import-export engine, and as long as you have a flat file available, Excel is the most typical example. We can load previous projects, we can load their full performance history, and we could start running reports against them as though they were live projects. We just would be looking back in time. So it’s pretty easy to do.

Meghan Russell:

Great, this question is similarly related, but does Contruent also look into predictive analytics and using historical information to improve the performance of future projects?

Karl Vantine:

It does, I think. One of the things I’m most excited about with our rebrand and our new investor behind us is that we will make more headway into the engines and the artificial intelligence capabilities that will make that a little bit easier to do. We do some of that now; there’s a bit more manual work involved. We run the performance reports, and we can feed back benchmarking and lessons learned to the original estimate because we’ve got that connected data.

There are still a few manual steps in taking that feedback loop back to the original estimate. But we can do it, and we can compare projects, one project to another, to see why one performed and one didn’t, so you can adjust your starting assumptions for the next project. I’m excited about being able to do that faster and in a more automated fashion as we continue to move forward with our Contruent investment.

Meghan Russell:

So, going back to the dashboard example, there was a question on, can you advise which tool was used to create it? Was it Power BI or your own tool in Contruent?

Karl Vantine:

It’s Power BI. We do not have our own dashboard tool. The reason is very simple: Power BI is a force to be reckoned with; it’s taken over the world. We’re seeing it everywhere we go. So we made a strategic decision about a year ago to move in that direction. We’ve interfaced with Tableau and we’ve interfaced with other dashboard tools along the way. But Power BI seems to be the most prevalent tool out there, and a lot of folks have it with their Microsoft offering anyway. So what we’ve done is we’ve built all of our connections to enable Power BI to work easily and get access. And then what we do is we offer—I’ve lost count of how many we have now, but I think we probably have 16 or 17 pre-built dashboards that just run the minute you put data in the system, and then you can go create however many more you need.

Meghan Russell:

All right. Okay, this question is from Michael McNeil. It says, addressing data management, i.e., location and formatting, when you have project controls data within several enterprise systems, each managing various aspects and data, how do you combine and link and assess the data when it’s stored by various enterprise systems and departments?

Karl Vantine:

So that’s where the common data environment comes into play. I referred to it as both a common data environment and a data warehouse. I recognize that a common data environment is a much more robust version of a data warehouse, with more going on. But we have that capability. We can pull the information from other systems. We can pull it all into the same platform, and then you start to produce the results so that you can evaluate it, but you have to have that one central repository that other systems feed through APIs.

Meghan Russell:

Okay, from John Schyler. Are you estimating or scheduling with assessments as probability distributions?

Karl Vantine:

Estimating or scheduling with probability distributions? Well, if I wasn’t a great scheduler, I’m definitely not an estimator, that’s for sure. We do apply probability weightings to the scheduled activities when we bring them across into our cost field, but I’m not sure I could cover that one from an estimating perspective. I’m happy to follow up with whoever asked that question afterwards and connect them with the right person, but I don’t think I can answer that one.

Meghan Russell:

Sure. Okay, this one is, can you please elaborate a bit on training staff on project data as it relates to scheduling?

Karl Vantine:

That’s a broad question. So, when training project staff on scheduling, I would argue you’d have to train them on all the project controls. Folks need to understand how the schedule interfaces with other data components throughout the project, but you’d want to make sure that folks understand how to set up activities correctly, how to ensure that they are appropriately assigned durations, so that you have short, measurable, compact durations, rather than long, six-month tasks. 

You’d also want to ensure that they understand how to connect each of the schedule activities so that you get a good flow of information, good predecessor and successor relationships. There are some fundamental and basic scheduling principles that need to be applied to ensure a reliable critical path. Otherwise, your forecasts aren’t going to work. If something’s delayed and it doesn’t show the proper downstream impact, you’re not going to have a good forecast of anything, whether it’s cost changes or other impacts. So, there are probably some fundamentals around scheduling you’d need to train folks on, and then train them on the business connection and how their data drives other processes throughout the project controls framework.

We do these things, by the way—I don’t anymore, but we have folks who do, so we can help with it.

Meghan Russell:

Can you explain a little bit about the integration with P6 and MS Project? How is that integration completed?

Karl Vantine:

So, we have an API in the system. And it’s really quite simple in its execution; it’s an elegant design. I gave an example of how we did it for National Highways; we basically take the coding structure that’s set up on the cost side and load it into the schedule in P6, where it’s in activity codes, and in Microsoft Project, you can do it in text fields and numerical fields—it doesn’t matter so much where you pick. But you’d put the different attributes of your coding structure for cost into the schedule, and then you connect them at the database level, and they sync up the activity ID with the control account ID in our system. How often you do this are all decisions you make with your data. Do I want it once a month? Do I want it every week? Do I want it for some slice of data, whereas a schedule that’s in progress or being built still, maybe I don’t integrate that? But it’s pretty straightforward, the data connection there. I’m happy to show you how we did it; I had it on the slides, but we can show you in real life.

Meghan Russell:

Okay, and I know there are so many good questions here. We’ll try to answer them after the webinar offline. But the last one is, I love it, it is direct and to the point—EVM, can you do it?

Karl Vantine:

Absolutely, we do it. Sure, it is what we are.

Meghan Russell:

Can you elaborate? 

Karl Vantine:

Can I elaborate? Our in value management, I know it’s a lot of things. But it’s really tracking the performance and progress against your packages of work so that you can get good metrics on your actual performance and understand how well you’re performing against your original plan budget. So you need that good percent complete, if you don’t have that good progress or performance measure, then that earned value falls over, but we have all the things you would look for in the earned value world. So actual cost of work performed, budget cost of work performed, earned value metrics, TCPI, CPI SPI, all those items are standard. And as soon as you start putting budget, dates, and progress in the system, we can run you tons of Earned Value reports right away because once you put data in, you start getting the reports out. Now you have to check your data and make sure it’s good, and makes sense. But yeah, once you load budget, dates, and progress, you get all the earned value reports you’d ever need. 

Meghan Russell:

Okay, thank you, Karl. And thank you everyone for attending. I know AACE will be sending this out with a confirmation of the CEU credits tomorrow, so you can look out for that. So thank you, everyone, and thanks, Karl. 

Karl Vantine:

Yeah, thank you. Appreciate the time and the attendance and more than 500 people, so that was fantastic. Thanks.