In construction procurement, coverage is the product

Supplier coverage, not software capability, is the binding constraint in construction procurement — and it is the reason a funded category still has room in it. Annija Eizenarma, a founder, investor and adviser to technology companies selling into legacy industries, told Artificial Reality, the AI buyer-evidence podcast hosted by Galina Fendikevich, that "there's still a lot of space" in price comparison, on one condition: be narrow enough to focus on a specific industry.

Why is there still room in a well-funded category?

Because every product in it is bought by one side of a two-sided workflow. Kojo has raised approximately $84M in total, SubBase $7M in a June 2026 Series A, ProcurePro $11M in a May 2026 Series A+, and Parspec $20M in a July 2025 Series A. The problem they attack is as manual as software problems get — Eizenarma's description of the baseline is that "in order to get a price, people used to call around 5, 10 vendors to just see what they can get."

When a category with that much funding and that obvious a value proposition still leaves space, the constraint is usually not the product. Here it is the supply side. The contractor buys; the vendors who have to respond bought nothing and are under no obligation to participate.

What does the coverage asymmetry actually cap?

Geography, and therefore usefulness. A platform can be excellent and still leave a contractor in a mid-size market making exactly the same phone calls, because the three suppliers who matter locally are not on it. Coverage is what the buyer is actually purchasing, and no vendor in this set markets it that way.

That has a direct product consequence. Vendor-side effort is the variable to engineer against: a supplier who must learn a portal to reply will keep replying by email. Artificial Reality reads this as an inference from Eizenarma's broader account of legacy adoption rather than a claim about any specific product — but it follows from her general finding that a reinvented process is the losing bet:

"The ones that are trying to reinvent the process of how the legacy industries are working are often less successful, because we do have to find a common language with the people that have been in a business for decades to gain that trust."

Is collecting quotes the same as comparing them?

No, and conflating the two is the most common gap in this category. The manual cost the operator described is the calling. The hidden cost is the normalization: five vendors respond in five formats — a PDF, an email body, a phone quote scribbled on a pad, a spreadsheet, a text message — and comparing them requires re-keying every one.

Trade structure makes that harder than it looks. A quote for electrical rough-in is not a line item. It is a scope with assumptions about what is included, who supplies what, and which submittals are required. A generic comparison view flattens exactly the differences that decide the award.

What should a builder in this category work on?

Requirements one to four below come from the operator's account. Five to seven are Artificial Reality's extension:

• Multiple vendor or subcontractor prices for the same scope, collected without calling each one — the explicit replacement for the call-around.
• A side-by-side comparison that shows where the better deal is, not just a pile of received quotes.
• Trade or industry specificity, which is the stated condition for where opportunity remains.
• A workflow that fits how the trade already works rather than a reinvented process vendors must be trained into.
• Vendor participation with minimal vendor effort (inference). Accept the email and the PDF; parse them.
• Quote normalization, not just collection (inference). The comparison is only real if the scopes are comparable.
• Historical pricing retained (inference). Knowing what the same scope cost six months ago is the second-order value of digitizing this at all.

What is the asset nobody in this set is selling?

The pricing record. A contractor who runs quoting through a platform for two years accumulates something the industry has never had: a structured account of what specific scopes actually cost, from specific vendors, over time. Today that is a byproduct. It is a pricing intelligence asset.

Eizenarma's view of where legacy industries are heading points at what would unlock it. She describes the future as collaborative — "how to establish strategic partnerships, how to collaborate with other independents, other construction companies... to ensure bigger, better projects, you can do more together with partners than just sitting alone in your corner." Pooled, anonymized pricing benchmarks across independent contractors are that logic applied to procurement. The trust required to build it is considerable. So is the moat.

What should you change this quarter?

Stop reporting platform capability and start reporting coverage, by metro and by trade, as the number a buyer evaluates. Make the non-participating vendor a first-class path rather than a support ticket. And name the category the way the buyer does: legacy industry buyers really do not want AI for AI, so sell four quotes by Thursday rather than sourcing intelligence.

Then ask the question this category has avoided: if coverage is what customers are buying, why is it the one number nobody publishes?

Where this comes from

S1E1: They do $20M a year on sticky notes — the full interview with Annija Eizenarma. Listen or watch: YouTube, Spotify or Apple Podcasts.