readme: Here’s a mock interview between myself and a simulated vc looking into construction tech. all questions are prompted by chatgpt 4. all responses are unassisted and unedited by ai. you can read the full conversation here.
jump to: impact, integration, roles, shifts, opportunities, public institutions
Q1
VC: Given your experience in architecture and construction, how do you see technology currently impacting the construction industry, and what are the most significant changes you’ve observed in recent years?
ME: The short answer is I don’t see technology impacting the industry in a meaningful way. More specifically, I don’t see much innovation in the field, where value is created. Take a look at the data on cost, schedule, and productivity. They are still as bad as they were a decade ago if not worse.
These are the only metrics that matter.
If you walk a job site — apart from maybe the most sophisticated projects — it looks the same and operates the same as it did before we had BIM, iPads, and fancy dashboards. There’s still a set of paper plans with a brick on them setup on a makeshift table. Everyone relies heavily on phone calls and text messages, because that is the one standard that can be used across companies.
A small amendment to my first statement, software has not had a meaningful impact. Tools and equipment have. Things like a cordless framing nailer.
This highlights the difference between innovating for the office or for the field. If we actually want to move the needle, solutions need to help those doing the work. It’s a people enablement problem. A training problem. A standardization problem.
But these problems aren’t as sexy as ‘an estimating tool with live cost data connected everywhere’. You can spin that yarn and get $20MM in VC but a nice estimate doesn’t lead to cheaper and faster projects, yet there is an obsession with improving efficiencies with overheads. It’s all an illusion. And please, no more modular home startups. They don’t work.
I’m not a complete cynic, though, I promise. I just don’t believe in vanity.
To the second part of your question, what significant changes are happening. I’m excited by the opportunities with AI and Robotics, particularly any innovation that is merging the physical and digital worlds. We’re super early on the hype curve, but my sense that what was once fantasy, is becoming more realistic.
Take a company like Caterpillar or Boston Dynamics. Autonomous or remote equipment, robots that can follow you on site carrying 1000 pounds of stuff, remote inspection — these ideas move the needle. They make field teams more productive and safer.
AI presents a significant opportunity to streamline the construction industry by digitizing the General Contractor role and simplifying the ‘construction stack.’ Many roles in construction, when examined from first principles, do not add real value and only complicate processes unnecessarily.
Q2
VC: That’s a very insightful perspective. You mentioned the potential of AI and robotics in enhancing productivity and safety on construction sites. How do you envision these technologies being integrated into the day-to-day operations of construction projects, and what challenges do you anticipate in their adoption?
ME: Let’s start with robotics, that’s probably the most tangible.
Robotics can be used to replace or augment the worker. Similar to equipment and machinery.
Replacing skilled labor that has a high-level of workmanship and nuance will be impractical. Trim carpentry for one.
I would expect robotics to solve more of the universal challenges with all trades. Help with carrying heavy things, moving tools and materials around, holding material overhead or setting something plumb. Activities that would usually require 2-4 people may only require 1. That’s the dynamic use case.
There’s also practical application for repetitive tasks such as laying bricks, masonry units, or what we see with 3D printing with articulated arms.
Another application would be monitoring and inspection. Anything where it’s typically dangerous for a human to go would be a good use case. Confined spaces, heights, hot/cold environments. These could be relatively simple semi-autonomous robots equipped with sensors and cameras mixed with computer vision.
The other approach is augmentation. Can the worker gain a new limb? Use multiple tools at once? Supercharged versions of themselves.
We did some prototyping when I was at Bechtel around exoskeletons. These carbon fiber vests that would connect down to your elbows. Not precisely robotics, but they helped reduce fatigue when doing overhead work. A real problem. If you’ve ever installed a ceiling fan you know what I’m talking about!
AI is a bit fuzzier.
A very hard problem to solve for mortals is scheduling. Extremely dynamic information, usually data poor, and reactive. By the time you make adjustments something else has changed and it’s too late. Your schedule is now a report, not a tool.
I’m a firm believer if you can solve the scheduling problem, everything else falls into place. It’s like an orchestra with no conductor and jumbled sheet music. You can stumble through the piece, but it won’t sound great.
AI can help not only solve and optimize a complex schedule, it can easily summarize what changed, send useful messaging, and start to direct the work. It could be proactive, real-time course correction. Then you connect it into all of your other systems. Ordering, dispatch, subcontracting…the foundation for diminishing the general contractor role in a healthy way.
Content creation and reasoning have a million applications. Interrogate a BIM model with natural language. Jurisdictions wouldn’t need to review plans. That’s half a year back on your schedule right there haha. What a world we could live in!
AI would be able to replace all the rote repetitive document creation tasks that plague jobs as well. Inspections, submittals, transmittals, execution plans. All of those things become much easier.
Communication. It’s quite common for construction management personnel and the workforce to speak different languages and be of different cultures. This is true for every project I’ve been a part of at least, and it’s true globally. Imagine all of those barriers suddenly go away and communication is frictionless.
This is available today, it just needs to be packaged in a simple piece of hardware with a speaker and mic or a cheap wearable and painted with a dead simple UI. Pick your two languages and go.
Haha maybe I’ll go work on this.
All of this is nice in theory.
There are the obvious technical blockers.
Look at how long it’s taking for autonomous driving to scale. Waymo is here in Phoenix and a few places now in California. Its first autonomous ride was in 2015, almost a decade ago.
Not only does it take a ton of resources and ingenuity to get the tech to even work, but getting it to work in a new environment is equally as challenging. Every city is almost a brand new problem.
That’s construction. Every job is different, and every job is constantly changing, making robotics, autonomous vehicles, and even AI much harder to leverage fully.
We can’t try and boil the ocean and need to stack the deck in your favor to avoid as much of that complexity as you can. Develop robots that are meant for interiors perhaps, or finishing which comes at the end of a job when there is more stability in the environment and the scope of the product is much narrower. Less variables to contend with.
As for adoption, we have behavioral and financial blockers to contend with.
Behavioral is easier. If you’re solving real problems for those that have hard jobs — those in the field — and making their lives easier, then the people side of adoption becomes easy. Bricklayers don’t want to haul around tons of bricks. It sucks. Give them a robot to do that part of the job, and they will be forever grateful and will deliver higher quality work. They could earn more money and have more time back with their families on top of it. This is why software is not exactly easy to implement and scale with subcontractors and craft. It’s perceived as an overhead, a tax. And it usually is. It wasn’t designed to help them.
The other reality is that construction is suffering from a severe labor shortage. It’s been getting worse for over a decade. The glass half full view on that, is robots won’t be stealing anyone’s job. The job is vacant and needs to get done. There should be more openness here compared to other industries perhaps.
Cost, on the other hand, is a real consideration. It’s unrealistic to expect a residential masonry contractor to purchase and carry around a robot to lay brick pavers. There will need to be consolidation or a sharing economy solution to make distribution possible.
Q3
VC: Considering the technical and environmental challenges of implementing these technologies in the ever-changing landscape of construction sites, how do you see the role of startups and established companies evolving to drive innovation and adoption in this space? What kind of partnerships or business models do you think would be most effective in overcoming the barriers to adoption you’ve mentioned?
ME: That’s a very apt question.
First, we can’t talk about startups without talking about VC and funding in general.
Many will disagree with this, and don’t take it the wrong way, but I believe there is a fundamental misalignment of incentives between the industry and tech. It’s not a function of ethics, but economic forces.
To truly innovate, the construction industry needs to boost field productivity. Field teams don’t need project management software, at least not nearly as much as they need something like autonomous cranes. A mobile app can’t lift a 2000 ton steel truss.
It’s not the best example, but it’s like Amazon over-indexing into their app UI and not fulfillment, or Moderna creating software to manage patents and not mRNA vaccines. It’s not bad, but completely misses the core purpose of their existence. These are the areas to innovate in last, not first.
The driver here, is it’s a lot cheaper to build software and distribute it. The construction software thesis works for VCs. There is opportunity for high-growth and fairly quick returns. People who don’t really understand construction start these companies, come in, and get something to work, at least on paper. Enterprises opt-in because what’s the alternative? One is greater than zero.
These startups may even get acquired by Autodesk, but is that a sign of progress from the perspective of the industry? I would say not even close. Money is not virtuous on its own.
If we look at the data that matters, we haven’t made any progress. Every project is over budget and blows its schedule and we keep trending in the wrong direction.
So what’s happening?
There’s a catch 22. Heavy R&D for construction-specific use cases won’t have high enough potential to secure traditional or VC funding, but the industry needs these to move the needle. I’m talking about leaps forward in operations-focused systems and hardware.
We are then left with this illusion of progress. Lots of capital being deployed, returns coming back, lots of excitement, yet nothing actually useful.
We need to approach innovation differently because construction is unique.
Startups need to be bold and demand to work on the right problems. Don’t formulate your thesis to secure VC. Formulate it to make operations more productive. Be OK not having some double-digit multiple or finding an exit. Define success differently.
Startups should still adopt the same hungry mindset and approach as they do now. Be agile, customer-centric, all that good stuff, with the tweak that you need to literally be getting your product into the user’s hands. Think about distribution models early on, particularly with robotics and hardware. You may just specialize in distribution and not the tech itself. That’s a solid idea.
Enterprises need to speak up. Be vocal about systemic problems and what is truly needed. Work with VCs, be LPs even. Push back on industry norms that we know are silly, particularly with public institutions. Step in and provide the capital and space for long-cycle R&D to happen safely, or at least to a point where VC or debt can jump in.
Enterprises should be partnering with startups to merge talent and methodology. Above all though, enterprises can accelerate the learning cycle for startups by giving them access to deep expertise and jobsites so they build something that is actually good. We pioneered this approach when I was at Bechtel and it was quite effective. Everyone in the ENR top 250 should have some version of this program.
The labor piece is tricky, I’m not entirely sure what the solution looks like, but it’s a shared responsibility across all the actors. Public, private, big, small, local, federal….if we can’t rebrand the industry and attract exceptional talent, technology won’t save us.
Q4
VC: You’ve articulated a compelling argument for a shift in focus towards solving the core operational challenges of the construction industry, rather than simply pursuing technological solutions that may not address the fundamental needs of field teams. Your emphasis on the importance of aligning incentives and fostering collaboration between startups, enterprises, and investors to drive meaningful innovation is thought-provoking.
Given this perspective, how do you envision the future of the construction technology landscape in the next 5 to 10 years? What kinds of breakthroughs or shifts do you hope to see, and what role do you see yourself playing in this evolving ecosystem?
ME: The pace of change is historically slow, but let’s assume we’re at an inflection point and we start to take off. The art of the possible. What should construction technology look like in a decade?
Good construction starts with designing for construction.
Generative design — or whatever term you want to insert here — should be huge.
Not just algorithmic generation of things like massing models, but fully detailed, optimized systems. AI-driven construction documentation working from a shared central repository of systems data.
Instead of physically drawing out a detail, or researching a new material, Architects and Engineers can use commands, images, references and descriptors to add fidelity to a model.
Even if design documentation is 100% dialed-in, coordinated, and optimized, it doesn’t mean anything if contractors don’t use it or follow it. This is the reality we live in now.
This could and should lead to a different way for contractors to consume and interact with design and engineering information. Even contracts. They would be able to interrogate an entire project, and also produce custom work packages or documentation tailored to their scope. Framing specific shop drawings automatically generated for the framers.
The most exciting opportunity, however, would be for Contractors to participate in the design process itself by encoding their expertise. Expertise that designers can’t normally access. Take the framing example again. During the design stage, the layout of walls and openings could already be optimized for ease of install for that exact location.
We have the opportunity to greatly compress the feedback loop between field and office.
The second big area is field productivity. Technology can be used to make field teams physically or mentally more productive. Those are the levers.
Physically more productive will be a function of better tools, equipment, material innovation, and robotics. The challenge here is in the distribution. How can small contractors — like those in residential — still benefit from these innovations? We can’t rely on Home Depot.
Mental productivity will be a function of less chaos. We need to use technology to remove the mental burden of planning, coordinating, and managing operational work.
I go back to the idea of an AI choreographer. Proactive scheduling, automated material ordering, dispatching, inspections, and most importantly communications. A digital superintendent of sorts. As an individual going to a jobsite to perform work, you wake up, get clear communication on what the scope is for that day, and your materials, tools, and equipment are at the workface — along with a coffee or red bull — ready to go.
Eliminate all the friction that slows down the deployment of unique skill.
The pace of technology-led productivity gains, however, can’t keep up with the drag created by bureaucracy and declining competency. Not just skilled labor, but overall competency to do the full job. This has become particularly acute in residential construction, creating a negative feedback loop exacerbating the problem.
Residential construction is in a truly desperate state.
Low barriers to entry, low competency, no economies of scale, and eroding integrity are at the root of the problem. The symptoms we see are in behavior and pricing. Here’s an example:
I saw three bids for the same 500 square foot casita come in at $375 a foot. This is more than the price per square foot of new build homes which include the land and new utilities.
I shared this news with one of the contractor’s we work with, and his response was “Hmm, is that a lot?”.
There isn’t a monopoly on building ADUs, I can promise you that.
One of the antidotes could be consolidation. Imagine and architect-led design-build company that also does a large amount of self-perform. It would be a long acronym, but a highly efficient delivery model.
The biggest obstacle here would be in training and specialization. People don’t go to school to learn how to build homes, they go to be an Architect or and Electrician. There should be a fork in the road very early on.
My point is that technology alone can’t have much of an impact on this side of the market. It’s very hard to compete with economic forces that push all the best contractors up market. It will always be better to work on one big project than 10 small ones, so you need to change the game. Turn every small project into a big one — take on more scope. Easier said than done.
A shift in the status-quo innovation model would also help kick things in gear.
The prototypical venture-backed SaaS startup isn’t working. As we’ve discussed, the outcome has been a bunch of point solutions that target overhead efficiency. They don’t move the needle.
SMBs suffer from by fragmentation. They need to stitch together a bunch of different programs because nothing solves the whole problem. Excel proliferates.
The other flavor of innovation is the homegrown solution within large enterprises. Enterprises aren’t usually great at building technology products. When they are successful, they don’t want to share it.
The solution seems obvious. Combine these two models. Enterprises need to become the VC. Startups need to be in the enterprise then spun out.
The risk is underserving the small side of the market. TAM is a decent enough incentive. Or it’s not. Maybe public institutions need to put some skin in the game.
As far as my role in all of this…persist and resist.
I hope to have the strength to keep trying. Share what I’ve learned with more people and inspire bold change. That’s what we need.
Q5
VC: Your vision for the future of construction technology is both ambitious and grounded in a deep understanding of the industry’s current challenges. The emphasis on generative design, field productivity, and a more integrated approach between design and construction speaks to a holistic view of where improvements can be made.
The idea of an architect-led design-build company that takes on more scope is particularly intriguing as a way to address the fragmentation and inefficiency in residential construction. Your call for a shift in the innovation model, with enterprises playing a more active role in fostering and scaling technology solutions, is a compelling approach to overcoming the limitations of the current startup ecosystem in construction tech.
Your commitment to persisting and resisting, and to inspiring bold change, is admirable. It’s clear that your experience and insights could play a significant role in shaping the future of this industry. Thank you for sharing your thoughts and aspirations with me.
ME: As investors, where do you all see the most opportunity? How are you thinking about the space?
VC: As a venture capital associate, I would see several opportunities in the construction technology space, particularly in areas that align with the trends and needs you’ve outlined:
- Field Productivity Solutions: There is a growing demand for technologies that enhance the physical and mental productivity of field teams. This includes robotics for carrying heavy loads, augmented reality for better visualization and planning, and exoskeletons for reducing physical strain. Solutions that make fieldwork more efficient and safer are likely to attract investment.
- AI and Machine Learning: AI has the potential to revolutionize many aspects of construction, from scheduling and project management to design and safety monitoring. Startups that leverage AI to provide actionable insights, predict project outcomes, or automate complex processes could be highly attractive to investors.
- Digital Twins and BIM: The use of digital twins and Building Information Modeling (BIM) is becoming more widespread. Technologies that enable better integration, collaboration, and utilization of these digital models can help reduce errors, save time, and improve overall project outcomes.
- Sustainable Construction Technologies: With a growing focus on sustainability, technologies that support green building practices, energy efficiency, and the use of sustainable materials are likely to see increased interest from investors.
- Construction Marketplaces: Platforms that connect contractors, subcontractors, suppliers, and clients can streamline the procurement process and improve supply chain efficiency. Marketplaces that offer unique value propositions, such as specialized materials or services, could be particularly appealing.
- Training and Workforce Development: Given the labor shortage in the construction industry, technologies that enhance workforce training and development, such as virtual reality simulations or online learning platforms, could be valuable.
From an investment perspective, we would look for startups that not only have innovative solutions but also demonstrate a clear understanding of the construction industry’s complexities and a viable path to market adoption. Collaboration with industry partners and a focus on solving real-world problems would be key factors in evaluating potential investment opportunities.
Q6
ME: What would you like to see from enterprises and public institutions?
VC: As a venture capital associate, I would like to see the following actions from enterprises and public institutions to foster innovation and investment in construction technology:
- Open Collaboration: Encourage collaboration between established construction companies and technology startups. This can provide startups with valuable industry insights and access to real-world testing environments, while enterprises can benefit from innovative solutions that address their specific challenges.
- Strategic Partnerships: Form strategic partnerships with technology companies to co-develop solutions that are tailored to the construction industry’s needs. This can accelerate the development and adoption of new technologies.
- Investment in R&D: Increase investment in research and development, particularly in areas that have the potential to significantly impact productivity and sustainability, such as robotics, AI, and green building technologies.
- Adoption of Innovative Solutions: Be more open to adopting innovative technologies and solutions, even if they require initial investments or changes to existing processes. This can help drive demand for new technologies and encourage further innovation.
- Support for Standards and Interoperability: Work towards the development and adoption of industry-wide standards for data exchange and interoperability. This can facilitate the integration of different technologies and systems, making it easier for startups to develop solutions that work across the industry.
- Education and Training: Invest in education and training programs to equip the workforce with the skills needed to effectively use new technologies. This can help address the labor shortage and ensure that the benefits of technological advancements are fully realized.
- Funding and Incentives: Provide funding and incentives for startups and research institutions working on construction technology innovations. This can include grants, tax incentives, and access to public projects as testbeds.
- Regulatory Support: Streamline regulatory processes to make it easier for new technologies to be tested and implemented. This can include providing guidance on compliance and offering fast-track approval processes for innovative solutions.
By taking these actions, enterprises and public institutions can play a crucial role in accelerating the development and adoption of construction technology, ultimately leading to a more efficient, sustainable, and innovative industry.
Leave a Reply