ConstructionBevy routes subcontractor bids through AI-run auctions – turning email-based chaos into a competitive, structured market.
ENTRY ANGLES
AI-powered auction platforms for procurement · Industry-specialized auction systems with pre-populated supplier directories · AI-driven market rate intelligence and bidding filters
VERTICALS
CAPABILITIES
AI/NLP to interpret industry-specific terminology and exceptions, Supplier directory and ratings database management, Market data and pricing intelligence
ConstructionBevy is a platform where construction companies source subcontractor services. The core mechanism is an auction – and a bilateral one at that: the platform runs auctions among subcontractors, and subcontractors filter and compete among companies.
The intake flow works like this:
Bid requests typically arrive by email, as they always have. To route them into ConstructionBevy's AI, contractors forward them – automatically or manually – to a dedicated platform address.
Once ingested, each request gets an AI-generated relevance tag: worth pursuing or not, based on the contractor's work types, current capacity, and configured preferences.
But raw bid requests are often incomplete. Missing start dates, unclear contract structures, or projects that turn out to be market surveys rather than real opportunities. The AI automatically identifies these gaps and sends clarifying questions to the project owner on the contractor's behalf.
Bids that pass the initial filters get automatically assigned to the right internal team members and relevant deadlines – like bid submission dates – sync to their calendars.
One notable feature ConstructionBevy claims no competitor offers: a shared board for routing unsuitable bids to partner contractors. If a request falls outside a contractor's scope or capacity, they can push it to the board where partner firms can pick it up through the same intake process. It's a lightweight referral network baked directly into the workflow.
For each vetted project, the AI generates a structured summary covering all key details and linking directly to relevant attachments like site plans.
Project owners can optionally display a competition level indicator – high, medium, or low – based on how many contractors received the same request. That's a small but genuinely useful signal for contractors deciding whether to invest time in a bid.
An AI assistant for freeform questions about any loaded project is also in development.
Pricing starts with a free limited tier. The paid plan is currently $49.99 per month under a launch promotion.
The startup started building in 2024 under the name HyperWater with somewhat different functionality before arriving at the current product, which just raised $1 million.
Auctions are theoretically the most efficient mechanism for a buyer to find the best-fit supplier at the best price. In practice, they're a genuine hassle for everyone involved.
For the buyer to get the best offer, they need to reach a lot of contractors. But more contractors means more responses to review – manually. The labor cost of running a thorough procurement process can easily eat into any savings the auction was supposed to generate.
On the other side, contractors in this industry receive an enormous volume of inbound requests. Small firms – which dominate the services market – often lack the bandwidth to properly evaluate even a fraction of what lands in their inbox. When most requests are irrelevant (because buyers maximize coverage by blasting everyone), contractors eventually stop engaging. That kills the auction market from the supply side.
Auctions are good in theory. Broken in practice – because they generate too much manual work for both sides.
Crown ([related review](/review/uspej-sdelat-to-chto-i-tak-proizojdjot)) raised €2 million at the start of last year to automate procurement auctions from the buyer side. But fixing one side isn't enough. The ideal end state is a full AI-auction: buyer agents broadcast requests, supplier agents evaluate them, the two sides negotiate details autonomously, and humans step in only to finalize decisions the AI has already prepared.
ConstructionBevy is an early approximation of that vision.
The partner referral board also reflects a broader trend worth noting: small firms actively collaborating rather than just competing. Collective OS ([related review](/review/luchshe-ostatsja-malenkoj-kompaniej)) built an entire platform for lead routing between boutique agencies and raised $2.5 million in January. Their thesis is that the future belongs to small specialized firms working together on shared clients – each handling their piece of the engagement. Collective OS wants to be the place they find each other.
Auctions are a genuinely powerful procurement mechanism that sees limited use in practice precisely because of the manual overhead involved. AI fixes that.
The direction worth pursuing: build AI-auction platforms for specific industries. Specialization matters for three reasons:
- It lets the AI correctly interpret the nuances, terminology, and exceptions specific to that market – which drives accuracy.
- It keeps the AI current on market rates and standard conditions, so it can serve as a useful reference point and filter for both sides.
- And it lets the platform pre-populate relevant supplier directories with specialization profiles, track records, and ratings – so the market is liquid from day one.
What industry could you build this for?