Every day for the next long while, I'm going to tear down a new public software company and highlight the AI risks/opportunities around it- products launched to date, top startups, key quotes from earnings calls, etc.
Day eleven, Procore,
$PCOR, which is up 10% today on earnings:
Peak share price: $103.28 (July 30th, 2021)
Share price today: $52.94 (-49%)
EV today: $7.8bn
ARR today: $1.4bn ( 16% Y/y)
NRR: 106% (95% gross retention)
EV/ARR: 5.6x
GAAP Operating Margin: -12%
EV/Run-rate GAAP EBIT: N/A
Headcount: 2,068 ( 4% Y/y)
What it does: Procore is the leading vertical SaaS company for the construction industry, focused on commercial/non-resi construction. The company has an incredibly broad platform, stitching together GCs, subcontractors, architects, owners and others into one unified platform for managing projects from bid to timetracking to construction management.
AI bear case: Procore is software, and as such is exposed to the bear case that as the price of creating software collapses, it may face pressure from independent ISVs as well as homegrown solutions spun up by nimble internal teams.
AI bull case: In almost any world of AI abundance, it feels likely that construction activity increases as the physical world becomes a bottleneck, providing a tailwind to Procore's business. The software has an intrinsic network effect as users from various constituencies use it to collaborate, making straight-up disruption harder. Construction is a famously conservative market- it seems deeply unlikely that customers will vibe-code their own solutions and if anything Procure could add valuable incremental AI revenue streams to its cross-sell motion.
AI traction:
Disclosed 700 customers and 66,000 active users on Procore AI. (Vs. 3m active users on Procore)
Adjacent AI-native startup summary:
Procore is overwhelmingly compared to/competes with Autodesk, Oracle, etc. but there are some AI-powered startups, especially with machine vision as a wedge:
BuildOps (Adjacent, targets subcontractors vs. GCs but colliding)- 625 employees, 59% Y/y
Kahua (Direct competitor)- 345 employees, 46% Y/y
Openspace (Machine vision for construction)- 310 employees, 9% Y/y
Doxel (Machine vision for construction)- 109 employees, 42% Y/y
Various others.
Management Quotes:
"We believe that the combination of Procore Helix and data grid will generate notable product synergies due to our highly complementary capabilities and road maps bringing premier advanced reasoning and broad third-party integration capabilities into Procore. We call this combined offering Procore AI."
"As construction's mission-critical system of record with nearly 3 million active users, Procore has a massive proprietary dynamic data set. And the value is not only in the volume of data, it's in the depth of its context. We map the complex dynamic interactions between people, workflows and the physical job site, capturing and continuously updating every document allocation and day-to-day change. This dynamic relevance is exactly what's needed to power high-stakes agentic AI."
"So let me just answer your question about monetization of AI and then I'll turn it over to you to Howard. Look, as in any business and new business opportunity, the first thing you've got to do is to establish a compelling ROI and we believe that we're doing that. We know that we're doing that. Our customers are seeing benefit and value from the technology, as I described in the example, where they're saving time and are able to do things that they wouldn't have otherwise been able to do given the shortage of labor and given the limited amount of hours in the day that they have."
Commentary:
Procore seems well positioned in an AI world, with its natural network effect and dominance perhaps exceeding even ServiceTitan's and potential LT tailwinds to construction over time. That said, progress on AI products still feels nascent and raises questions about the bull case mid-term- it seems unlikely that Procore sees any material AI revenue for at least the next year, if not longer.
Still, Procore typifies vertical SaaS- defensible, durable, etc. and given that its customer base will likely see AI disruption last, it arguably has a longer runway than most to adapt to AI disruption.