There is a large divide between how construction should work and how it actually works. This is frequently misunderstood by tech vendors aiming to 'disrupt' the industry, sometimes with disastrous results, also for themselves. Katerra, anyone?
Misunderstanding this rift between "is" and "ought" in construction is at the root of the slow tech adoption holding the industry back. Everybody agrees the sector is ripe for a technological revolution, but nobody's really sure where to begin or how to get there.
Large and complex projects are still built in pretty much the same way as they were in the past. We've barely left the age of pen and paper behind. Excel remains thoroughly cemented as the industry's killer app.
But, but, but, I hear professionals sputter, we also have Asta (in Europe) and P6 (in the US). And what about BIM? There's BIM too!
Okay. Let's talk about these applications/approaches.
Asta Powerproject and Primavera P6 are scheduling applications. And I am going to let you in on a little industry secret. Of course everybody uses these apps. Well, them, or Microsoft Project. How else can you schedule a large project with many interdependencies?
Nobody builds without a schedule (or programme, in the UK.) But the 'little secret' is that the schedule is often not primarily created to plan the build, but to win the bid. So it often does not match the project reality, even from the very beginning, and cannot be relied upon once the building gets underway.
This is not an unreasonable state of affairs. When you want to build something, you have to present a plan that sounds logical. It is impossible to foresee the intricacies of a project that hasn't even started: there are just too many unknowns. So the General Contractor does their best with the schedule but by necessity operates on a lot of assumptions that may or may not prove valid.
And even if the schedule is created to actually prepare the work, reality soon impinges upon it, since in construction many things tend to quickly deviate from the original plan.
Drawings change, technical difficulties require things to be built differently, material and labor shortages necessitate other changes, and on-site issues often require creative solutions that impact the project in unexpected ways. So construction is constantly in flux.
And since most projects (in the US anyway, it's different in the UK) do not have a dedicated scheduler attached to them like a fly on the wall, nobody is updating the schedule with the many changes that are constantly being inserted into the construction process. Result: the schedule is soon no longer up to date.
To put it somewhat irreverently, and in a way that’s sure to irk schedulers and planners everywhere: the schedule goes out the window before the first window is even put in place.
Then BIM. The internet is very enthusiastic. Many contractors have invested heavily in it. And I am going to be the last person to rain on its parade. It's a potentially wonderful technology.
But let's be real. It's so expensive that it's still only used on a tiny fraction of projects, even though it's been around since the 1970s. And when it IS used, even the VDC specialists themselves fail to keep the model up to date because it's just so friggin' hard and tedious.
On a fundamental level, both BIM and scheduling applications address the 'ought' part of construction I mentioned at the beginning. They are premised on the idea that we build with a plan — a great idea — but are ultimately unresponsive to the unchanging nature of construction, which is that it's always changing.
Those solutions help you get a great start in building but fall short to keep you going.
Now I wouldn't be saying this if I did not believe that our company, Disperse, provides a real solution to keep your project moving forward. However, I should hasten to say that our solution not only presents an alternative but also a complementary path, since we integrate heavily with both Asta/P6 and BIM for clients that need us to.
Instead of basing our solution on how construction should work, we have looked at how it actually works and how we can improve that admittedly imperfect situation.
Here's how we approach it:
When we are deployed on a large and complex project, we take all the building information like the architectural and shop drawings, the floor plans, the schedule, and the BIM model (if present), and we ingest it into our systems. This way, we have a clear picture about what 'ought' to happen.
But then we bring a dedicated photographer on-site every week, say on Friday afternoon. He or she walks a dedicated route taking 360º photographs of the site in all the same places. This enables us to see what actually happens.
We feed that data to our AI, and we have a dedicated team of engineers and architects teaching the AI what is happening in the photos.
We have been doing this for several years. For longer than anyone in the industry, in fact.
So we understand both what the intention was, but also what the current status of the project is. And often, the discrepancies between the two are so significant — because there are so many changes that are not recorded — that it simply makes no sense to run a comparison between what was planned and what is actual.
It serves no one. It is better to look at what is in the photos and assess what is really the case. And we can do a remarkably good job at figuring out what is going on, even without the plan.
And that's because of the simple fact that although each building may be different, the build order is often the same. And within the same project, similar patterns tend to occur.
It's 'easy' for a well-trained AI to spot if things have been installed wrong or out of sequence. And it enables us to single out issues that might impede progress with great reliability.
In fact, one client tells us that for every issue their project teams pick up, our system picks up another one.
This implies two things. The first is that with our system, more issues are noticed that do not need to get worse before they are dealt with; something we have written about at length elsewhere.
The issue is detected. Therefore it can be acted upon. Therefore the project team is not blindsided by its ramifications down the line.
The second is that since many more issues are being detected than would be picked up manually, you end up with a better quality asset. Things do go wrong in construction sometimes, and not everything is installed to the highest level of integrity all of the time. We all know this, and there is no reason to beat around the bush about it.
But with the extra set of super-eyes and transparency that the Disperse system brings, project teams can act on things early, keep it real with their people, and make better decisions because they can base them on objective data.
The power of our system is in the data analysis of the site. That’s very important to understand because we are sometimes incorrectly pigeonholed as a reality capture solution. We are not that.
Reliable reality capture is a requirement for us to deliver actionable quality outputs. The better the quality of data coming in, the better the actionable insights we provide as outputs to the project team.
Reality capture and processing are the foundation upon which a lot of the reporting we do is built. While we are primarily a platform that helps project teams with execution by augmenting their decision-making with timely insights, automated reporting is an excellent time-saver for them as well.
Reports need to be compiled anyway, for the owner or other stakeholders, and sometimes simply because the contract says so.
At the push of a button, Disperse makes reports available that normally take a project team many hours per month to compile.
Inflation, supply chain issues, labor shortages and rising labor costs make ours a viable solution for that reason alone.
Our reports enable project teams and other stakeholders to keep a pulse on how the project is doing at all times. In which areas has work started or not started? Is it nearly complete or complete? Where are my trades falling behind? Which issues are occurring on-site?
Our reports include everything from high-level executive summaries that can be sent to the owner and other stakeholders, down to detailed progress, cost reconciliation and labor overviews.
And sometimes, on rare occasions, we do come across projects where 'is' and 'ought' actually do match on the build site. These are genuinely unicorns, rarely spotted in the construction landscape. But they do occasionally occur.
I am referring, of course, to projects where the schedule is kept up-to-date. Combined with an accurate schedule, the schedule integration with Disperse affords project executives a truly unprecedented level of control. It means playtime is over.
We can also accurately track Earned Value then, comparing the planned value with the project's cost and the actual percentage complete. This is notoriously difficult to do in construction because it's so hard to maintain good underlying data.
But with an up-to-date schedule and the accurate progress tracking that Disperse provides, it is possible to do this for our clients reliably. This should be music to the ears of any forward-thinking project exec, and the most complex projects being built today often require it.
Everything is eventually a question of cost. If the cost of inputting a system does not greatly outweigh its output, it is not benefitting the project either in terms of time or resource reduction. Many tech solutions fail precisely because they fail to get on the good side of this equation; they offer some benefit, but their input requirements are so heavy that people get tired of them and eventually move on.
Many current solutions constantly leave builders a few steps behind the reality of their project and do little to help people in the field drive projects forward. But those people still need to find ways to minimize risk and maximize output. The competitive construction landscape increasingly leaves them no choice.
Disperse has developed a system that makes that state of affairs much easier to navigate. And it's based in the way you are actually building, not in how some software engineer in Silicon Valley thinks you should be doing it. The result is a solution that helps your project be completed sooner, for less money, and to a higher quality standard to boot.
These are not hypotheticals. We have been doing this in the UK for several years. We are still relatively unknown in the US market. But this is rapidly changing and we are already working with some of the biggest and most forward-thinking contractors here.
We have also developed an exciting new iteration of our solution that still contains all of the helpful reporting I spoke about before, but has now become much more useful for boots-on-the-ground project people — our intended audience.
Superintendents (or Construction Managers in the UK) are visually oriented people working in a high pressure environment. They have to make decisions on the fly without needing to do data analysis or number crunching.
Instead we present them with a visual, familiar, easy to understand interface and put the insights about issues that are missing, at-risk or wrongly installed, front and center in our application. This means that project people can be more secure in their day-to-day execution, because they now have a trusted system that will hand them the issues they need to care about right on a silver platter. Well maybe not a platter, but an iPad. A silver iPad, perhaps.
Are you a project executive on large and complex projects, and are you curious to see if we can deliver everything we say we can? Try us out and let us prove our value to you.
We could start by showing you how all this works in an online demo. Book one here.
Thanks for reading,
Brian Clarke
President Disperse Americas
