Getting the Most Out of Your
Historical Project Data

Adam Palant :

Good afternoon and welcome to the webinar Getting the Most Out of Your Historical Project Data. This event is brought to you by Engineering News Record and sponsored by InEight. Hi, I’m Adam Palant, Manager of Special Sections and Projects at ENR and your moderator for today’s webinar. Thank you for joining us.

As many organizations pursue their digital transformation journey, it’s important to consider the big picture relaying the leveraging the data that is collected. The data we harness from past projects can be used to build more accurate estimates and schedules and avoid many project surprises. Through benchmarking, estimators and planners can validate assumptions for new work based on historical data. Today we’ll be discussing different techniques and getting the most value out of historic data, along with a practical lessons learned approach to making benchmarking a cornerstone of your preconstruction process.

Now let’s get to know our presenters. From InEight, please welcome Rick Deans, Executive Vice President of Industry Engagement, Aaron Cohen, Estimate Product Director, and Jordan Brooks, Product Manager for Planning, Scheduling, and Risk. Joining them, we’ll also hear today from Mike Albani, Vice President of Estimating at Aecon.

Since 1998, Rick Deans has worked with InEight customers in more than 35 countries to help identify innovative solutions that address their biggest project management pain points. As Executive Vice President of Industry Engagement, he leads InEight’s efforts to engage with its most strategic customers through the industry advisory group. Aaron Cohen is the Estimate Product Director at InEight and is responsible for defining product requirements and overseeing development of best in class estimating software solutions for today’s market needs. He has over 15 years of experience in the business as a project manager and estimator for various infrastructure and utility construction projects.

Jordan Brooks is the Product Manager for Planning, Scheduling, and Risk at InEight. His real-world experience and knowledge of project controls practices is prevalent in the technological advances his team delivers in construction planning software. Prior to InEight, he gained invaluable project knowledge working on programs for large capital projects ranging in size from 5 million to more than 10 billion. Our fourth presenter is Micheal Albani. In close to 30 years with Canada’s largest publicly traded infrastructure development company, Aecon, Micheal has experienced tendering bids for projects across Ontario. His duties have also included the generation of project control estimates, preparation of budgets directly from estimates, cost and productivity formal tracking and analysis, tenure reviews, and tenure selections for bidding.

Now, I’ll rejoin the presenters at the end to answer your questions that come in throughout the webinar, so don’t forget to submit them in your Q&A section of the webinar console. And now I’ll hand things over to Rick Deans from InEight to kick things off. Over to you, Dean.

 

Rick Deans:

Great, Adam. Thanks so much and thanks to all of you for joining us today. We realize that everyone has a lot of competition for their time and we’re… bit through some polling questions here in just a moment. So we do want this to be interactive, but I thought we’d start off with a quick discussion of the origin of the term benchmarking. Many of you might know this, but the term originated in the mid 1800s and it was a way of validating the compatibility of interchangeable parts during manufacturing processes. So as certain manufacturing processes evolved from hand fitting each component, the whole concept of interchangeable parts and pieces came about and bench tests were performed, which created marks basically generating data points. So there’s your origin of the term bench marks. This data was used to predict outcomes in the future and in today’s business world, benchmarking strives to provide predictable outcomes in the future based on what we’ve measured and studied in the past.

So as mentioned, we do want to do some polling with you folks. And so what we’re going to do is we’re going to ask you to describe your organization. We’ve given you several options here. One might be a better fit than the others. If you do choose other, by all means throw the answer in your chat there and this’ll give us a good understanding of what our audience looks like in terms of composition.

And to avoid any kind of awkward silence during the polling, I wanted to share with you all some fun facts. I had the opportunity of looking at the lunar eclipse early this morning, late last night, and one of our fun facts deals with astronomy. Did you know that Venus is the only planet in our solar system to spin clockwise? It travels around the sun once every 225 Earth days, but it rotates clockwise once every 243 days. So in earth talk, its years are longer than its days. Did you know that? Okay, so it looks like we’ve got a good representation here between owners, engineers, architects, contractors. Beautiful, beautiful. If you did put other, I want to see it in the chat. So let us know what that looks like. So we’ve got manufacturing, perfect. We’ve got building survey and laser scanning. Excellent. Wonderful, wonderful.

Now what I would like to do is understand which best describes your role in your organization. And again, we’ve given you some options there. If you do choose other, by all means, throw that in the chat and we’ll hit another fun fact here. Let’s talk a little bit about continuing the theme of astronomy, but let’s relate it to some geography here on earth. Did you know that Australia, the continent of Australia is wider than the moon? The moon sits at about 3,400 kilometers in diameter. For those of you in North America where we don’t use the metric system, it’s about 2,112 miles, while Australia’s distance from east to west is almost 4,000 kilometers or 2,500 miles, to round up. So interesting that Australia, the east to west distance on Australia is larger than the diameter of the moon. So if Australia were in space, we’d be able to see it then.

So we got a lot of project managers, we got some leadership management folks. Beautiful, beautiful. Well thank you so much for taking the time to fill out the pole and let’s see if we got any others. Scheduling engineer. Very good. Okay, thank you. It looks like we got some consultants too.

So let’s dig into our presentation. In terms of continuing the theme of benchmarking as it relates to construction, what many of our customers like to do is take the guesswork out of future work by relying on their historic data. Here we can see we’ve got some benchmark data represented visually in sort of scatter graph approach. The idea being I’ve got an average somewhere in the middle. I’ve got past data points that help me understand what I’ve estimated and what it actually took to build some of this work. And I’ve got maybe the estimate that I’m working on is represented by a white diamond and you can see it’s maybe a little bit outside of the range of where I would feel comfortable as determined by that green section in the middle of the chart.

So I’m going to pick on Aaron Cohen. Aaron is our Director of Estimate Products here at InEight. Aaron, help us understand some of the thought that went into building the estimating benchmarking and give us an idea of some of the anecdotes or real-world applications of how you’ve seen it used in the real world.

 

Aaron Cohen:

Yeah, thanks Rick. So if anybody’s ever tried to use their past costs to price their work for stuff that they’re going to be doing in the future, you probably learned pretty quickly that that might not be the best way of establishing a price for your work. Every job in our industry tends to be different. And as soon as you think you’ve figured everything out, you find out the hard way a lot of those things that you didn’t know all of a sudden come to the surface in the next job that you got.

So that being said, there’s a tremendous amount of institutional knowledge that we capture in our information systems on a daily basis. And the benchmarking feature as we’ve built it and as we see it used throughout the industry is really intended to help unlock and leverage some of that information in a meaningful way so that as you’re establishing a price for your work, you can go back and take a look at other similar scopes of work, get a sense for maybe not necessarily the average price is this, so that’s what I’m going to use, but understand where your price for your work based on the assumptions that have gone into it has really been established relative to some of those other things.

If you’re using, let’s say that you’re going to be placing some cast in place concrete for some footings, maybe you’ve gone through and you’ve established that we do this type of work all the time and we can see that the price for this work should be anywhere from X to Y dollars per cubic yard or per cubic meter. And then for this particular project, all of a sudden you’ve got a huge price escalation in the cost of concrete material shortage or something like that. And all those past costs kind of go out the window and you’re kind of stuck needing to understand, where am I? Well, we can also flip from looking at costs to looking at man hours and productivity units. So, maybe instead of looking at the cost per unit, you’re looking at the amount of time per unit to do an installation.

So a lot of different ways to slice and dice the data to take a look and to really try and validate the estimate that you have in a current project and see if it stacks up reasonably against past jobs and what winds up happening is when you see things that are out of variants, you see things that are outside of that range of the normal, it gives you an opportunity to take your estimate and to dig deeper into those specific areas.

Sometimes that information can be hard to really see at its surface when you’ve got a large estimate that’s got a lot of different moving parts and pieces and there’s a lot of information. You kind of get lost in spreadsheet land, but when you can see this stuff put together on a scatter plot and you can start seeing things falling outside of a range of normal or acceptable, then a lens for the estimator to be able to take some further action and try to investigate further as to why that price, maybe it’s right, maybe it’s not, but it really helps to manage some of the risk that goes into an estimate.

 

Rick Deans:

No, that’s excellent, Aaron. And Mike, from the voice of the customer, help us understand how Aecon uses benchmarking and from your role as a leader of the group, I’m sure there’s no shortage of data. How do you sift through some of the data to really hone in on the stuff that might be more applicable?

 

Micheal Albani:

Yeah, that’s a good question, Rick. I agree with Aaron too. I think the most relevant parts of the benchmarking are related to the productivity factors, not necessarily the dollar side. I think the whole industry is seeing this everywhere, especially across Canada and the rest of North America. Escalation is really affecting a lot of the pricing that goes in today, whether it’s fuel or union labor agreements or the price of steel. All of these things can trip up benchmarking if you only do it by dollar. So I think the productivity is the key thing that can be reflected on in the tables.

And to your question, Rick, I think the difficulty becomes when we have so much data, it gets almost impossible for an estimator to look through the data and then pick, well my job looks like this one of 50 jobs. That gets kind of difficult, especially in today’s environment where we have a lot of mobility between workers and employees going from company to company. And they may not know the history per se of each project specifically. They may be aware of it, but not the intimate details of the project.

So, one way that we found to drive benchmarking with so many jobs being bid in inside the estimating office, it gets to be a little bit overwhelming at times trying to review them all to make sure they’re accurate. And one of the ways that we invoked was the benchmarking. So we could actually see the results of these graphs on the front CBS screen, on the CBS register itself. So what we had was a summary of data on all the productivity histories and we analyze that information and then fed that back in through Excel basically into the InEight estimate system. And then we used that data to reference, and it kind of looks like what’s on the screen right now being presented, so you get a curve and you get it for every individual type of work that you have benchmarked, whether it’s supply and place concrete to a footing or placing structural steel at a very specific type of item level, we can do that.

 

Rick Deans:

Talked about what if scenarios, what if we had A team out here? What if we had the B team out here? Do you use benchmarks to, you mentioned earlier that with the mobility of the workforce, that you might end up with really skilled, seasoned folks that understand the work. You might end up with a crew that maybe this sort of work is new for them. Do you do any modeling or do you do any what if scenarios based on what you know about the availability of the resources that is going to perform the work?

 

Micheal Albani:

Yeah, we do. We tend to take a look at it on kind of a global perspective. So we’re not trying to nail it down and say, this job is exactly like this past job. We’re not trying to get that finite. We tend to look at it from the perspective that we have a workforce and there’s some people that are maybe more skilled and experienced within that workforce than others, some may be newer to the field and they’re in charge of different types of work. So if we’re placing concrete to a footing for example, there may be some crews that are really fast at that and others that are not so good at that, they may be better at doing form work or better at doing the deck concrete for example.

So what we tend to do is look at the data that we have on the past productivities and we’ll tend to break that down into probability of occurrence. So we might say the median productivity, that’s the 50 percentile, is this productivity for a concrete place to a footing. But then if we say, “Well, if we want the 25% chance of that productivity being achieved, we’ll be a lot more aggressive in the production.” And that’s only reserved for the top performing based crews. So we know then, if we have a new project and we know that a certain crew is available or a certain quality of crew is available, we can then bid with those higher performing metrics.

So it makes it easier, I think, to look at that versus trying to find a specific job. And I’m not denouncing that approach, but when you have lots of jobs in history like we do at times, it’s overwhelming and the estimator can almost never pick a production that’s wrong because it’s such a wide range on all these jobs. That’s where the statistics comes in.

 

Rick Deans:

Yeah, and I guess that is helpful. As one of the leaders of the team that’s helping make some of those final decisions, you do have the ability to, as you point out, to determine really how aggressive you want to be based on the data and based on what the likelihood of… And I’m going to come around and Jordan’s going to talk about this, I think, in a little bit, but as we talk about the likelihood of reaching a certain performance metric, maybe we can do some what if analysis around that. No, that’s excellent, Mike. I appreciate that.

And we talked about the data being presented graphically within user-defined variance thresholds. The data is also available in what we would call more tabular format. And one of the things that I’ve noticed just as I go around and I talk to people in the industry, there’s no shortage of this data. It exists in organizations. It exists in ERP systems. It exists in three-ring binders which contain printouts of old cost reports. The trick is, can I leverage that data? Can I get to that data? Can I make decisions based on that data within a relatively short period of time? We’ve certainly, in this discussion, we’ve been looking at this from a contractor’s point of view. I did notice there were a lot of owners and engineers as well. And maybe we’re not estimating a specific project or a specific scope of work within a particular project, maybe we’re doing some preliminary capital planning and we want to come up with a capital budget, but we want that budget to be based on elements of work that ring true, right?

So as an owner, I’ve seen a lot of owners do benchmarking, we’ll talk about this a little later in the presentation, to really help them come up with those early stage budgets, those early stage funding approvals that they can continue to refine as they learn more information about the project. So data is out there. I think one of our value propositions is that we organize the data, we normalize the data, and it’s there in the context of planning new work. We don’t have to go hunting around for it, I don’t have to… And I know Aaron and Jordan are going to get back to me when I leave them a voicemail or when I send them an email, but it takes a lot of that tribal knowledge out of it and it helps me look across the organization as we’re planning our work.

The next thing I wanted to talk about, usually when we talk about benchmarking, as the panel has already surfaced, we talk about cost performance, we talk about productivity performance, but Jordan, as our schedule expert, I’m going to pick on you a little bit, help us understand some of the things that InEight Schedule does that helps us understand historic durations and how the schedule might be impacted on this upcoming project based on what we’ve seen in the past.

 

Jordan Brooks:

Absolutely. Thanks, Rick. I’m going to keep with the theme of data and relate this back to what our approach is for this benchmarking concept when it comes to scheduling and planning and even risk. Back when I was a schedule engineer, and I’m sure the schedule engineers or those with have experience with scheduling on this call will know that in your day-to-day business process, the amount of data that you produce from the scheduling, creating monthly updates, doing what if schedules, doing risk analysis is massive. You have massive amounts of data just sitting there. And I think it’s become more of a push within the industry to capture this data and use it in a way that’s meaningful for not only contractors but owners with your subcontractors, vendors, all of that.

So what Schedule does is it works a way to not only capture that as-built data or possibly past project data or even productivity rate data that you’re just producing throughout that business process, it’s giving a way to verify that data into a knowledge base. And we capture multiple types of data within this knowledge base from entire past projects to, like I just mentioned, activity, productivity rates, resources that may be used on a project or schedule, some subcontractors that you may use typically over and over again. And then I know later on we talk about this, but even risk register events that you want to capture not only issues but opportunities that you may see on jobs. Those are all captured in what we call a knowledge base and users can verify this data and then once that information is verified, what that allows these users to do is go in and quickly build projects, quickly compare projects to say, am I within the threshold that it should be?

I relate a lot of this benchmarking work to subcontracting work. When you’re on an active project, let’s say, and you have a subcontractor who’s doing work and they’re going to leave the job and then come back later and do more work, a lot of the times you’re making a guess about how long it’s going to take for that sub to get that work done. It’s just they give you the amount of time they need, it’s a guess, they don’t have much backup to it, but maybe they’re coming back later on the job and they destroy these durations, they do a lot better than they said they were going to do initially. You can go to that work later on on that project, adjust your schedule durations to match what their productivity rate was for that previous activity, and forecast out where those durations are going to end up. All of this goes into being able to forecast where this project is going to end, where that work’s going to be done.

You can then look at things like indirect cost. It has obviously big impacts on cost when you’re talking about where that duration is going to end. And another thing that we begin to utilize with this knowledge base is what we call augmented intelligence, which is a hot button throughout the industry today. This allows users to go to like projects that they may be at the estimate phase or early on and they say, “What should I expect from a job? Have we done it before in our organization?” If it’s captured in that knowledge base, that gives those users an ability to go in there, mine that data, Schedule will make suggestions based off of whatever parameters a user’s given within Schedule and say, “Is this what you’re looking for? In the past, this is what it took you to get this job done.”

So we’re utilizing a lot of this past data and even as-built or even productivity rates that we’re seeing just in those day-to-day business processes for scheduling, capturing that, putting it in a knowledge base, allowing users to mine it, and then use it for future work on their jobs.

 

Rick Deans:

So keeping with the theme of lots of data, I’m glad that you mentioned the InEight Schedule knowledge library where a lot of this data is stored. And then as those schedules are built out and people apply metadata tags, knowledge tags to this data, this inference engine can then help put together new schedules. So again, I realize we’ve got a mixed group here in the audience. We’ve probably got some contractors that are given some pretty well defined scope or maybe some variable scope and it’s their responsibility to come up with a cost and a duration to do that, but we’ve also got owners and engineers, owners’ reps who might be looking at a future project and just like they want to put together an early stage capital budget, maybe they want to have some early stage scheduled durations and maybe they’re trying to get a plant built by a specific point in time.

Can you elaborate a little bit, Jordan, in terms of how maybe an owner might be able to leverage this data to throw together a pretty high level, maybe it doesn’t go down into lower detail, maybe it does, but to put together a high level schedule that might be appropriate for an early stage look at a duration for a project?

 

Jordan Brooks:

Yeah, absolutely. When you get into the owner realms, I’m going to tie this into the program level scheduling here just to get an all-encompassing view of what they deal with. On the owner end, you’re going to have projects, let’s say you’re an owner and you do a lot of light rail work, you’re going to have projects from past contractors all over the scope saying, “Here’s how long it took this contractor. This is how much light rail mileage was in there.” It’s all captured in that knowledge base.

These owners can take that type of work and maybe you have a couple different types of projects that go across this program. You can go into your knowledge base and say, “Okay, in the past when we’ve done this type of project, how long has it taken us at a high level look?” And you can adjust that, you can right size it for maybe this job has a couple less miles of light rail. Okay, well let’s adjust this duration overall. We know this light rail work may impact some road work that we’re doing that’s involved in this program, so we can say, “Okay, this light rail work’s going to get done here. When can we start this next road project that’s dependent on that light rail work at a program level?”

So the owner can go in and at very summary high level, or even down to detail if they wish to go to, but keeping it for this discussion at a high level look and do what we call planning packages of these projects and say, “Okay, this is when we think this job’s going to get done based on past experience and our knowledge base. We can start this roadwork.” And then you can kind of plan out what budget you need for each one of these projects, look at that overall cash flow at a program level, and start making decisions based on, do we need to get extra funding for this time period that we don’t currently have? How can we do that?

So a lot of business decision making goes into when you start looking at those past projects, when does timing line up? What do we need as far as cash goes? All of those decisions could start being made. And then you can look at, again, we’ll talk about this later, but then you bring into the risk factor those past register events that may impact those types of jobs, those certain types of jobs, and then you can have some type of certainty that, yeah, we’ve had this contingency covered for in case this risk hits again on this type of project.

 

Rick Deans:

Well that’s excellent, that’s great. That sure beats having to start with a fresh sheet of paper every time, doesn’t it? An empty canvas. Mike, from your perspective, do you ever have, from a contractor’s perspective, do you ever have any of your clients bringing you on early on in the thought process of putting a project together and work with you to help them come up with preliminary budgets and preliminary durations for certain projects?

 

Micheal Albani:

We do. Yeah, for sure. It’s a style of contract that seems to be happening more in Canada now. The owners are getting more engaged than they used to be and these are large public infrastructure projects we see this on.

 

Rick Deans:

And what sort of value add does a firm like Aecon bring to the table in terms of those types of projects, Mike?

 

Micheal Albani:

Well, we’ve got a lot of history in different sectors. So Aecon, as far as heavy civil type work, that’s roads, bridges, highway construction, power generation, the nuclear field, utilities, gas, oil, fiber optic, you name it. So we’ve got a long history in a lot of these areas and we’re a self performed company so we can draw upon that for our work that we’re going to perform and use that for productivities and durations and schedule kind of know how.

And then the rest of the work that we subcontract out, we’ve been working in many areas with the same subcontractors for usually quite a number of years. So there’s a relationship there often that takes place. So it’s good to draw on that.

 

Rick Deans:

Yeah. And you hit on something that’s really important too, and I’ve sat in a lot of meetings with both owners and contractors where maybe the contractor has come to the table with a plan that’s maybe not the cheapest, but it’s predictable and it’s something that they’ve done before and they bring a lot of credibility to that discussion when they say, “We might not be the cheapest to do this work, but have you considered X? Have you considered Y? And for our Canadian friends, have you considered Z?” And it really helps to get people thinking in terms of that value engineering in terms of, wow, if we did it this way, it might cost a little bit more, but these are some things that we could avoid as a result of doing it this way.

 

Micheal Albani:

And that’s very true. There’s a lot of nuances on the contracting side that others maybe don’t see as design engineers. They’re not working in that space as the contractor. And same with the owners. They don’t see all the day-to-day potholes sometimes that we have to go through and overcome to go build a job. So contractors in general want to build a job and do a good job on it, but they want to get in and they want to get out as efficiently as possible. They’re not trying to delay the jobs at all. They want to construct them and hand them over as quick as possible.

 

Rick Deans:

Sure. No, that’s excellent. Great stuff. With the project management triangle that we’ve all learned about since our days in school; scope, time, and cost, I kind of think of it as a five-sided triangle, which I realize isn’t a triangle anymore, but it’s a catchy name. When we layer on things like quality and safety, InEight has a set of tools that measure these things, compliance, for instance. And as we’re embarking on a new project, certainly we want to look at things like cost per unit, man hours per unit, durations for specific activities. Which activities are we going to likely need for this project as we’re looking at it for the first time? But there’s this other element of quality and safety and how much rework can we expect?

And if we’re going about our normal everyday jobs and we’re tracking this information and it’s not in a point solution, but it’s in an inclusive end-to-end, integrated project management platform where it can be leveraged, many of our customers are saying that they can build in time hold points for QA and inspections and things like that. Again, based on the way work has happened in the past. And Aaron, you mentioned outside of your experience within InEight that you know did work in the contracting industry for some time. Any anecdotal feedback from you in terms of quality or compliance or safety issues that you’ve given consideration to when you’re looking at a new project?

 

Aaron Cohen:

I mean yeah, absolutely. I’ve always felt that when you’re estimating work, the devil is in the details. And I think some of the best estimators out there are the people that have actually put their boots on and gone out and built the work. I think it’s that level of detail that you’re referring to there, Rick, that really is getting down to the details that matter.

So, as far as the compliance issues, as far as the safety issues, as far as understanding, there’s a very large jump. I did a lot of underground utility work, very, very large swings in your productivity just by getting into different types of grounds and the types of shoring that needs to be considered and required and the safety implications to keep everybody safe. So those are really, really hard to predict in many ways and I’ve found that throughout the years, there’s a lot of folks that just have this institutional knowledge that they understand, they can go out and look at a job, kick the dirt, sniff the air, and have a really good sense for, boy this is going to be a bad job. Not a shovel’s been stuck in the ground yet or nothing’s happened yet to really start saying, “Hey, we’re going to be losing money.” But the guys that have been doing this for a long time have a good sense and a good feel and it’s why they’ve been doing it for a long time, because they’re good at it.

They might not know exactly what they know, but they just know something, right? That intuition. I don’t think it’s any surprise to anybody that our industry is very, very challenged with a lack of skilled people. I also work with university and trying to fill the pipeline of graduates that are coming into this industry, there’s just not enough graduates, there’s not enough trade people coming into our business. So I think the way that we’re going to start doing more with less, we’re going to have to start relying more on our information systems. We’re going to have to start relying more on the data that we’ve collected over the years to start looking for some of these patterns, the things that you see when you start looking through, what happens if I layer on compliance and safety type issues with some of my good benchmark data that’s talking about cost and schedule is you start adding more layers to bring to the surface and to the awareness of the people that are using this things that they need to think about that maybe they hadn’t thought about before.

So I think you’re going to see a heavier reliance on the data and it’s really important that all that data is available and can be used by the people that don’t have the experience, they don’t have the 30 years, have been there, done that, and they don’t have the intuition. They’re going to need to rely on the data, they’re going to need to rely on some analytics, some patterns. And the folks that we do have that still have that intuition, they need to help augment that process and put some context to the things that they’re seeing when they find those patterns in the data.

 

Rick Deans:

No, that’s excellent, Aaron. I appreciate the insight there. We did talk about owners, engineers being present online. One of our tools, InEight Design, allows for EPC projects to really monitor and keep track of that quantity growth as projects are going from the 30% to 60% to 90% plan stage. So again, we can all sit around and say, “Well, we know we’ve seen quantity growth in the past, but we shouldn’t expect any this time because this time is different.” Well that’s certainly one approach. Another approach would be a little bit more pragmatic in my opinion, and that would be, “Hey, for this type of project in the past, what sort of quantity growth have we seen, for instance, with our engineered commodities? And maybe we should be looking for or trying to include some of that in our real world budgeting and cost estimating.”

The next thing I wanted to talk about was, at InEight we’ve got a number of products, but when we go to market and we talk to our prospective customers and our customers, we really want the discussion to be product neutral and really focus on business processes. And we’ve identified 16 different business processes that are tools in some way, shape, or form support, and sometimes it’s through a singular tool. Sometimes it’s through an integration between tools. So for instance, I’m taking off a set of plans using some 2D takeoff and I want to be able to put a cost on the price against that with estimating and pricing. So, some of these are very well intertwined and our message to the market is when you look at these business processes, a lot of these are managed in the real world by point solutions. We’ve got a point solution that does this, we’ve got a point solution that does that, and that’s great. And in many cases, those are best of breed applications.

But one of the things that I think our value proposition really lends itself very well toward these days is that just by doing our jobs, just by going out and performing our work and recording what’s happening and making sure we’re doing timekeeping so that people can get paid, making sure we’re measuring progress quantities so that we can get paid or we know what to pay our contractors or our subcontractors, that data is being collected. And rather than a bunch of point solutions where we’re doing the sneaker net or we’re moving data around on shared drives, it all lives in a holistic environment where it can be tapped. And through visualization reporting, we natively use Microsoft’s Power BI within our applications to surface that data. It’s very easy to identify trends and, again, leverage that data to help you understand what potential outcomes you’re going to see in the future, mitigate risks, mitigate surprises, et cetera.

And I wanted to have a discussion. There’s five or six bullet points here I wanted to explore for the next five or 10 minutes, and that is what I’ve seen, I’ve seen the evolution of benchmarking go from some of the examples we talked about earlier, we’re going to pour a concrete footing and we want to know what sort of production rates, what sort of costs we’ve seen in the past, but I’m seeing owners, for instance, one owner, they’re a customer of ours, they’re a pharmaceutical giant, they build plants all over the world. They were interested in understanding what their risk and their contingency was as a percentage of their direct cost. And sure, they’ve always loaded up budgets with a certain percentage of contingency money set aside to handle known unknowns.

But one of the things they were very keen on doing was to actually monitor that draw down of contingency. And Jordan, we hinted about risk a little earlier. I’m wondering if you could talk us through some of the capabilities of InEight Schedule as it comes to risk and measuring what might be an acceptable appetite for risk and being able to see, okay, if we want to get really aggressive, here’s one simulation, if we want to be conservative, here’s another simulation. Can you help us understand how the tools might help there, Jordan?

 

Jordan Brooks:

Absolutely, yes. So as you mentioned risk, or Rick, risk is something that a lot of our clients ask about constantly. And one way that Schedule has approached the concept of risk and contingency as a percentage of that direct cost is applying that to a schedule model, producing a bunch of simulations, and saying, “Okay, with the input of your uncertainty,” which may be you’re unsure of your productivity that you’re going to hit, or some type of outside uncertainty that may cause a duration like quantity growth to increase those durations, with a combination of that and the risks that may have been identified, whether that be on this specific project or even early on, you’re saying in the past on these types of projects, we have a knowledge base of our risk events that have occurred on these and we want to incorporate these into a simulation model that’s going to give us some type of confidence level that this is a duration we can rely upon to apply that contingency cost to.

So within InEight Schedule, you’re bringing in those register events, those uncertainties. You’re running multiple simulations quickly and it’s giving you an output that’s going to say, based on your inputs and your confidence level of, let’s say a 70%, this is where we see you hitting as far as duration goes. And what I’ve seen users do is not only, we have a cost side simulation and a schedule side simulation, but users can take either model and apply that to their estimate or even to their contingency draw down that they’re trying to incorporate into their iterative processes and say, “Okay, based on this model, this is how much we can plan for having to draw down if these risks hit.” Or, “In our estimate, we need to plan for this much contingency to be 70% sure that we’re going to hit the duration that we’re showing in our schedule model.” So I’ve seen that from a lot of our clients being applied on an iterative basis and that’s kind of the way we’ve approached it with InEight Schedule specifically.

 

Rick Deans:

Excellent, excellent. The next bullet point I’ve got up here is engineering as a percentage of procurement and construction costs. So I’ve got a handle on the P and the C, maybe early stage I want to have that E value, right? EPC, maybe I want to have that engineering cost as a derived value just based on history. And Mike is a large contractor, I’m sure you do a lot of EPC contracts. Do you ever look at the metrics, or maybe it’s procurement that’s a derived cost based on the level of effort in the engineering and the construction. Do you ever look at the variability between those different aspects of an EPC project, the engineering, the procurement, and the construction?

 

Micheal Albani:

We do. We try to look at that as a percentage, engineering as a percentage as an example. It’s difficult. It probably goes a little deeper than that, but that is a good benchmark. That’s how we start actually. And probably one of the last numbers we get when we’re closing a large job is actually the engineering component because they’re usually working right to the very end. So we always need placeholders because every contractor on major projects is going to take the number to maybe the president or the Board of Directors for approval and you can’t have a blank, you have to have a placeholder for that. And these meetings occur weeks prior to the closing. We’re talking three, four weeks when the engineers are still working away trying to capture all the scope and make sure they’ve got it all covered in their pricing. So it’s a very handy and very good tool to use.

 

Rick Deans:

So it’s a starting point, right? Like you say, every project’s going to have its own level of detail, but at least you’ve got that placeholder so when you have these meetings, it’s not a big surprise at the end when the value of the contract jumps suddenly because we’ve now included the engineering costs?

 

Micheal Albani:

That’s right. And it’s a bit of a fallback with all these things too. Nobody expects the guesses to be perfect. So in cases where we’re wrong and we’re using benchmark values, you can go back and say, “Well listen, the number maybe came outside of that range, but we expected this range because of X and Y and-”

 

Rick Deans:

Here’s the justification for it.

 

Micheal Albani:

Yeah. So it’s estimating with some intelligence behind it.

 

Rick Deans:

Sure. No, that’s excellent. We talked about the third bullet point a little bit earlier, and that as an owner, maybe I’m not ready to go to market and get competing bids for this project, but maybe there is a project that aligns with our strategic initiatives and we want to put together an early stage budget just for internal purposes. So as we’re going and trying to get some funding for this project, just to Mike’s previous point, we’ve got a placeholder. We understand what we think it should be, but the market will tell us for sure when we’re ready to do that, but this will allow us as an owner, as an owner’s rep, as an engineer to come up with a rough idea of what we’re expecting it to be. And then it can also serve as a sanity check and a place of validation.

So if you’re a lot higher than that, maybe there’s some good reasons we can justify that. If you’re not quite there, if you’re a lot lower than that, do we understand the scope fully? Are there things that we might be missing? So, just a way of validating or coming up with early stage estimates for capital planning purposes from an owner’s perspective. One thing that we’re seeing in the market and that I’ve seen a lot of energy around this just in the last six, 12 months, and that’s subcontractor management and performance. Mike mentioned this earlier, they’ve got a lot of history with a lot of different subcontractors. If you’re an owner, maybe you’ve got a lot of history with certain contractors that you’re hiring to perform the work to be able to measure their performance, look for areas where there might be some shared value in increasing that performance.

As Mike said earlier, most contractors, they want to get in, they want to do the job, they want to get out, and along with that get paid. And if there are ways that we can look at past performance and maybe identify some opportunities, certainly historic data can help us with that. We’ve talked also about, during the design process, allowing users to track changes during the design stage of a project. Quantity growth was the example we brought up on the screen. And again, it seems like we might be jumping all over the place, right? We’re doing early stage capital planning, we’re doing safety, compliance issues, we’re talking about direct costs, we’re talking about scheduled durations. But I think one of the unique value points of InEight is because all of these solutions live in an integrated project controls environment, the environment itself, the platform supports and enforces that consistency. If I’m going to be referencing a cost item over here, guess what? Along with that cost item becomes a quantity and a unit of measure and other metadata values that just enrich that data throughout the entire process.

So I wanted to thank our panelists. We’re not quite done with you, we might need you for some of our questions, but Adam, I’d like to turn it back to you to see if there’s any Q&A from the audience at large that we’d be happy to stick around for another few minutes and address any questions that have come up through the chat.

 

Adam Palant :

Yeah. Thank you all. I mean such great information you’ve all shared with us today. Now before we have our presenters address any questions, I’d like to remind you that we’d love your feedback, so please take a few moments to complete our webinar survey, which you will see on your screen now, or you’ll be directed there at the end of the presentation. So now onto our first question, there you go. So you have it right there, so you can fill that out while we’re doing the first question. Now, concrete pouring productivity varies depending on the type of form work or the density of rebar being utilized. How can I be sure I’m comparing my upcoming work against the most applicable history?

 

Rick Deans:

I’ll take a shot at this, a swing at it, and then Aaron, maybe I’ll hand it over to you for your expertise. But one of the things that our estimate application allows when we’re benchmarking is that it allows for really two quantities, a primary quantity and a secondary quantity. So in this specific example where I think the question is going has to do with, I want to compare apples to apples and maybe that rebar density, maybe kilograms per cubic meter of concrete or pounds per cubic yard of concrete, maybe that’s a ratio I’m interested in.

Obviously, if we’re going to be pouring a slab on grade for a parking lot, we’re going to have a different reinforcing strategy than if we’re going to build a foundation that’s going to support a really, really heavy turbine or some power-generating equipment. But that would be something I would be looking at is that ratio of rebar embeds to concrete. Other customers use the ratios of concrete volume to form work area as an example to help them understand like items in the past. Aaron, our civil engineer, our resident civil engineer, would you have any additional questions on that?

 

Aaron Cohen:

Yeah, no, I think you hit the nail on the head there, Rick. I would also suggest that probably for every company, they’re going to have different things that make them successful, different KPIs that they’re going to look at and key in on to really make sure that they’re controlling their work and they’ve got years of experience that in their experience, these are the things that matter, right? And I don’t think that’s consistent across companies and I think that’s one of the benefits of our solution. We’ve got a lot of different ways of letting you tag your information to do just that, to identify and segregate the work that’s going to be most important. So just as the question had said, these are the things that matter for concrete placement. We give you the ability to build that in to your benchmarkable data and segregate it in that way as well.

 

Rick Deans:

Excellent. Well, I appreciate you chiming in on that.

 

Adam Palant :

All right, another question. You mentioned a coding structure can help compare apples to apples, but do you have any customers that are using something other than a code to refer it to like work in the past?

 

Rick Deans:

Well, I think we might have one with us. Mike, what have been some of your experiences around developing codes, maintaining codes, maybe ditching codes all together, whatever you’re comfortable sharing with us?

 

Micheal Albani:

Yeah. I’ve seen a lot of variety, I guess, in the system. So some companies I’m familiar with years ago, there was one part of our organization that had 10,000 codes, for example, and it didn’t last because they’re trying to insert coding structures and the numeric formats change and computer system change and it doesn’t accommodate the old way.

So another way of doing it, which works fine in InEight benchmarking is to use descriptions which we’ve found very helpful. So to use the footing example, if the description is placed concrete to footing with a certain embed ratio of rebar beside it in the description, as one example, you can match that in your benchmarking library. And if you are consistent enough to keep bidding your jobs with the same descriptions, such as you copy from a library or a master CBS structure, all that matching is done automatically for you using descriptions.

The one nice thing I’ll add too, Rick, is sometimes estimators will need to make new CBS activities and they copy and paste, and then the first thing they might do is go change the description because they want the same crew, but for a different type of work. If you use account codes, a lot of estimators may tend to forget to change the account code that it’s referencing for a benchmark, but if you do change the description, it’ll break the link to the benchmark because the description is different. You don’t get an incorrect benchmark reference that way.

So, that’s worked for us on the descriptions and it alleviates a bit of management of more codes. Estimators already have a lot of numbers to deal with, I find, so the less count of numbers, probably the better.

 

Rick Deans:

Sure, that makes sense. And Jordan, I’m thinking attributes, I’m thinking metadata, I’m thinking knowledge tags, inference engines. Are those things required to take advantage of some of the augmented intelligence within Schedule, or are there other ways of leveraging the-

 

Jordan Brooks:

Yeah, Rick, I think you read my mind here with this one. The way Schedule approached this early on, and I think InEight as a whole continues to look at this, is we implement what we call fuzzy matching, which basically means you don’t need direct one to one matches of codes or descriptions or anything like that. There’s so much data and metadata that is captured within our business processes that by implementing a fuzzy matching instead of needing a one to one match, you can give users matches and suggestions and then you can present them with, how much does this actually match the strength of that match? And then a user can go in there and say, “Okay, this is what I’m looking for,” or, “This is what meets my requirements,” and they don’t need that exact one to one match or those detailed account codes or whatever types of codes you’re using. It’s really just all of that metadata is being utilized to present the user with the best case scenario that they’re looking for.

 

Rick Deans:

No, that’s excellent. Thank you guys for chiming in on that one.

 

Adam Palant :

Well, it looks like we got one more. Now we’ve tried doing this before and we seem to get stuck when trying to load all of our historic data. Do you have any suggestions on the best way to load our history?

 

Rick Deans:

It’s like the old adage, the best time to plant a shade tree is 15 years ago or today. So you don’t have to pull all that historic data forward. I think the most important thing, and I’m interested in our panelists’ thoughts on this as well, but what I’ve seen work and my experience working with customers out in the field is, let’s get started. Let’s identify a path. Guess what, if we identify a path today, in six months you’re going to have six months of really well normalized data that’s available. In 18 months, you’re going to have that much more. In 24 months, you’re going to be on your way. Does it make sense to go back and strategically pick a project or two because, yeah, we don’t want to forget all that organizational history, but maybe that’s the approach. If we focus so much on converting 100% of our history, we are going to get stuck there. We’re not going to make the improvements that we need to make to capture data in a normalized, clean way today and going forward.

So that’s where I emphasize customers start. Let’s put something together that’s going to make sure whatever data is coming in the door today is data that you can harness and leverage, and then let’s strategically pick some projects. That can always be an initiative is to go back and pick a handful of projects, make sure that data is normalized as well, and bring it into the tool that way.

Aaron, any further thoughts on that approach from what you’ve seen both in your life as a contractor as well as a software vendor?

 

Aaron Cohen:

Yeah, absolutely, Rick. I think it’s really important that you keep in mind your 80/20 rule, right? And I would say that generally speaking, what 80% of your work in an estimate is going to be in 20% of the line items? So the objective shouldn’t be, every single job that you’ve ever done, to migrate that history and create this massive database, but start building it over time and be strategic about it. Identify the things that you want to track, the things that add more variability, the things that add more risk, the things that you’ve lost money on before, target those and start to look at those.

The other thing is, as far as pulling all of your jobs that you’ve ever done, think about your business and think about the way the business has changed over the years. If you’ve been doing work for 30, 40 years, is the way that you build the work today the same as you did 10 years ago? The way that you track it, the way that you account for it, if it’s not, then that might be a good logical break point as well as far as how far back do you go. But certainly trying to make it manageable and get some small wins and get some good information early to get some momentum going in a positive direction rather than the, let’s move everything wholesale and try to get out of day one a comprehensive database. I would say that you definitely want to be strategic in your implementation.

 

Rick Deans:

Thank you.

 

Adam Palant :

Well unfortunately, that’s all the time we have for questions today. Please join me once again in thanking Rick Deans, Aaron Cohen, Jordan Brooks, Micheal Albani for this discussion, as well as our sponsor InEight. If you have any additional questions or comments, please don’t hesitate to click the email us button on your console and we’ll share them with our presenters so they can respond directly to you. If you didn’t have a chance to fill it out earlier, you will be redirected to our post-event survey at the end of the webinar and we look forward to hearing how to make our programs work better for you. Please visit enr.com/webinars for the archive of this presentation to share with your colleagues as well as information about our upcoming events. Make sure to tune back in for our next webinar, Defining Digital Twins: What They Are, Who Can Benefit, and When To Start Building airing on November 15th at 2:00 PM Eastern Time. Thank you again for trusting us with your time today and have a great day.