Pre-Construction Intelligence Becomes ConTech's New Battleground

Trend Spotlight

Two startups on different continents, backed by different investors, are attacking the same bottleneck: construction's least-digitised function is not on the jobsite, it's in the pursuit and tender room. Last year's ConTech wave brought jobsite tools. This year's is bringing revenue intelligence. Cascade, a New York startup founded by ex-Google engineers, raised $3.5 million from a16z Speedrun to predict construction projects before RFPs are filed. Kuadra, a Cairo startup backed by Edafa Venture, is automating tender document review for Egyptian contractors, cutting hundreds of pages of bid specs down to hours of analysis. Different markets, different regulatory environments, identical thesis: the bid phase is where construction wastes the most money on the least technology.

This matters because it signals where capital and talent are heading. The Class of 2026 ConTech founders are not building another field-management app or drone-mapping tool. They are building tools that help contractors find and win work — a category with a clear revenue line. Contractors will pay for something that helps them win a $6 million project. They are more reluctant to pay for something that helps them track a $6 million project they already won.

The pattern strengthens against the wider data. Mastt's 2026 survey of construction project managers finds AI agent usage at 9% today but planned by 46% within twelve months. The biggest barrier is no longer cost. It is trust in accuracy. That is a maturation signal: the market has moved past "is AI possible?" and landed on "can I rely on this specific output?" Construction software is projected to grow from $12.91 billion to $32.89 billion by 2035 at 9.8% CAGR, and the agent sub-segment within that is where the acceleration lives.

This Week's Headlines

🟢 Cascade Raises $3.5M to Find Construction Projects Before the Bid Exists

Cascade, a New York-based startup founded by ex-Google engineers, has raised $3.5 million in seed funding led by Andreessen Horowitz's Speedrun programme to build an AI platform that identifies construction projects months before any RFP is filed. The platform tracks bond filings, permit applications, property transactions and planning meeting minutes to surface opportunities that haven't yet entered formal procurement. TechCrunch reports Cascade has already surfaced more than $10 billion in opportunity pipeline for its early clients, a figure confirmed by HackerNoon and Sovereign Magazine [1][2][3].

Cascade's wedge is "pursuit" — the business-development function that remains the least digitised part of construction. Project management, estimating and field tools have all seen waves of software investment, but finding and qualifying projects to bid on still runs on spreadsheets, relationships and cold calls. One client reported winning a $6 million project found exclusively through the platform, a concrete ROI claim rather than a percentage improvement on an existing workflow. The startup's challenge will be data moats: if the opportunity-sourcing signals are public, competitors can replicate the pipeline, and the defensibility likely lives in the proprietary scoring models that predict which early signals convert to real projects.

Why it matters: Pre-construction intelligence platforms could become the CRM equivalent for AEC firms — a system of record sitting upstream of project management and estimating tools, at the revenue-generation layer where willingness to pay tends to be higher than for operational software.

🟢 Kuadra Wins Edafa Backing to Build MENA's First AI Construction Platform

Kuadra, a Cairo-based construction technology startup, has closed a funding round led by Edafa Venture Capital to launch what it calls the first unified AI-powered platform for construction contractors in the Middle East and North Africa. The platform automates tender document review, extracting specifications from hundreds of pages of bid documents and generating structured summaries that contractors use to make go/no-go decisions in hours rather than days. iAfrica reports Kuadra is in advanced talks with one of Egypt's top-five contractors and plans to expand into oil and gas project management, confirmed by Daily News Egypt [4][5].

Kuadra's emergence reflects a broader pattern in emerging-market ConTech: pragmatic AI applied to administrative bottlenecks rather than frontier applications like robotics or generative design. Egyptian contractors deal with complex, multi-language tender documents that take teams of engineers days to parse manually, so the ROI of automating that process is immediate and measurable. El Sewedy Electric, one of Egypt's largest industrial conglomerates, has reported an 84% reduction in process time across 30 AI projects — a self-reported figure that should be treated as a claim rather than verified data, but one that signals large regional players are seeing results worth publicising [4].

Why it matters: Kuadra is evidence the ConTech category is expanding beyond North America and Europe. The MENA construction market is large, under-served by software, and increasingly receptive to AI tools that solve concrete operational problems.

🟢 AI Agents: 9% Today, 46% Planned. The Adoption Cliff Is Coming.

A 2026 survey of construction project managers by Mastt finds AI agent usage sits at 9% today but is planned by 46% of respondents within twelve months. The survey also reports that roughly 50% use AI tools daily, 72% weekly, and 67.6% express favourable sentiment toward AI in project management. The AI-in-PM market is projected to grow at 21.77% CAGR, reaching $14.14 billion by 2030 [8].

The numbers carry an important caveat: Mastt is a ConTech vendor selling AI-powered project management tools, and the survey's methodology and sample size aren't documented in the source material — treat the specific figures as directional rather than precise. Even discounting for that sourcing bias, the signal is strong. The 9-to-46% jump is the most forward-looking adoption metric of the week, and it suggests construction is approaching an inflection point where AI agents move from experimental to operational. The survey also finds the primary barrier to adoption has shifted from cost to trust in accuracy — a maturation signal, since the conversation has moved from "can we afford this?" to "can we rely on this specific output?"

Why it matters: The next twelve months should see strong demand for AI agents that integrate with existing PM workflows rather than replace them. The 46% who plan to adopt aren't looking for new platforms — they're looking for capabilities that slot into the tools they already use.

🟢 Construction Software Market: $12.91B Today, $32.89B by 2035

SNS Insider projects the global construction software market will grow from $12.91 billion in 2025 to $32.89 billion by 2035, a compound annual growth rate of 9.8%. Cloud-based solutions hold 71% market share today. The estimating sub-market is growing faster, at 11.61% CAGR, while construction management software is projected to grow from $9.94 billion to $24.12 billion over the same period [9].

The figures come from an SNS Insider press release rather than a full report with documented methodology, and there's no Tier 1 research-firm data (Gartner, Forrester, McKinsey) available to cross-check the projection. The directional trend is consistent across other analyst firms even where exact numbers vary: the market is growing, cloud share is dominant, and estimating is the fastest-growing sub-segment.

Why it matters: A 2.5x market expansion over a decade means the addressable opportunity for ConTech builders is widening substantially, and the interesting growth is in sub-markets like estimating, where AI-native tools can take share from legacy on-premise systems that never modernised. For incumbents such as Procore, Autodesk and Trimble, growth is increasingly about depth in specific workflows rather than breadth across the project lifecycle.

🟡 APEC Adopts AI Initiative, Creating a Policy Tailwind for ConTech Across APAC

APEC ministers adopted the Chengdu Statement in July 2026, formalising a regional commitment to responsible AI adoption in existing industries, SME enablement with locally relevant AI tools, and trusted cross-border data flows. The initiative builds on the Gyeongju AI Framework endorsed in 2025 and launches a 2026-2030 implementation timeline; South Korea has already established an Asia-Pacific AI Center under the framework [6][7].

For construction technology, the initiative matters because it reduces regulatory friction for scaling ConTech products across Asia-Pacific markets. The emphasis on SME enablement is particularly relevant, since construction is dominated by small and medium firms in every APEC economy, and the initiative explicitly prioritises "locally relevant AI tools" for them — language that aligns with what Kuadra is doing in Egypt and what pursuit-intelligence platforms could do in Southeast Asian markets where public project data is abundant but fragmented.

Why it matters: The framework is policy, not regulation — it creates direction without enforcement. But the 2026-2030 timeline means implementation guidance and sector-specific frameworks will follow, and construction is a natural candidate for early sectoral guidance given its economic weight across the region.

🟡 Beam AI BIM CoPilot: Attentive.ai Pivots From Takeoffs to Full Lifecycle

Attentive.ai launched a fully managed BIM coordination service in June 2026 under its Beam AI brand. Contractors provide 2D PDFs and the Beam AI team handles 3D modelling, clash detection and bill-of-quantities generation. The service is backed by Attentive.ai's $30.5 million Series B, raised in December 2025 from Insight Partners [10][11].

The pivot is from pre-construction takeoffs to full project-lifecycle BIM management. Attentive.ai built its business on AI-powered quantity takeoffs for highway and infrastructure contractors; BIM CoPilot extends that into building construction, where the complexity of 3D coordination creates a service opportunity that pure software tools haven't fully addressed. The "fully managed" framing matters: Beam AI isn't selling a self-serve BIM tool, it's selling a service where its team does the modelling work using its AI platform — positioning it as a labour-arbitrage play rather than a software-licensing one. Attentive.ai's own site cites 1,200+ customers, 90% accuracy and a 75% time reduction, but those are unverified vendor figures; the independently confirmable facts are the product launch and the Series B behind it.

Why it matters: The gap between AI-powered takeoffs and AI-powered BIM coordination is closing, and the companies that bridge it will capture contractors who want BIM-level coordination without maintaining an in-house BIM team.

🟡 Generative Search Emerges as AEC's New Gatekeeper

OFA Group reported 161 ChatGPT-driven discovery events for its QikBIM platform in the first two weeks after launch — the first quantified evidence of generative AI search functioning as a customer-acquisition channel for AEC technology. Users arrived via ChatGPT across six countries, registering on the platform after conversational search queries led them there [12][13].

The product metrics themselves come from a GlobeNewswire placement and should be treated as promotional. The real story is the generative-search signal: if construction professionals are using ChatGPT to find AEC tools, the SEO and digital-marketing playbooks ConTech vendors have relied on for the last decade need rewriting. Generative search doesn't return ten blue links — it recommends one or two tools based on query context, which makes visibility winner-take-most.

Why it matters: The AEC software field is already crowded — Motif, Snaptrude, Arcol, Qonic and Autodesk Revit 2027 are all competing for mindshare among architects and engineers. If generative search becomes the primary discovery channel, vendors that invest in being "answerable" — structured data, clear use-case definitions, authoritative content — will capture disproportionate share. It's an early signal, not a confirmed trend, but the kind that tends to matter more in retrospect than at the time.

Data Point of the Week

AI agent usage in construction: 9% today, 46% planned within 12 months. Source: Mastt's 2026 survey of construction project managers [8]. 🟢 The jump marks a shift from copilots that assist people to agents that execute tasks — and the firms that adopt first will build proprietary workflow data that late adopters cannot easily replicate.

The Longer View

Pre-construction intelligence as a category

Two startups, Cascade and Kuadra, are building what amounts to a new software category: pre-construction intelligence. Cascade mines public signals to predict projects before RFPs are filed; Kuadra automates tender document analysis so contractors can bid faster. Both are attacking the pursuit function, the least-digitised part of construction since the industry went digital. The open question is whether this becomes a standalone category or a feature that incumbents such as Procore, Autodesk or Trimble fold into their existing platforms — and that depends on whether the underlying data (public records, bond filings, planning minutes) needs specialised infrastructure the incumbents don't have.

Generative search as a distribution channel

The QikBIM data point — 161 ChatGPT-driven discovery events in two weeks — is small but suggestive. If generative AI search becomes how construction professionals discover new tools, the competitive dynamics in ConTech shift: SEO optimised for page-rank rewards content volume and backlinks, while generative search rewards structured data, clear use-case definitions and authoritative answers. Those are different playbooks, and it isn't yet clear whether early visibility compounds into a lasting advantage or simply rotates as more vendors enter the space.

The trust barrier

Mastt's survey finds the primary barrier to AI adoption in construction has shifted from cost to trust in accuracy. That's a real change in the buyer's mindset: when cost is the barrier, vendors compete on price; when trust is the barrier, they compete on reliability and proof, which is harder to manufacture and slower to establish. The vendors that solve it — through transparent methodology, independent validation or demonstrated outcomes — stand to capture the 46% of project managers who say they plan to adopt agents within a year.

Sources

[1] TechCrunch — "Cascade Raises $3.5M to Predict Construction Projects With AI", https://techcrunch.com — ~July 2026. 🟢

[2] Sovereign Magazine — "Cascade Construction Tech Startup Raises $3.5M Seed", https://sovereignmag.com — ~July 2026. 🟢

[3] HackerNoon — "How Cascade and a16z Are Betting on Pre-RFP Construction Intelligence", https://hackernoon.com — ~July 2026. 🟢

[4] iAfrica — "Egypt's Kuadra Wins Edafa Venture Backing to Launch AI Project Platform for Construction Contractors", https://iafrica.com/egypts-kuadra-wins-edafa-venture-backing-to-launch-ai-project-platform-for-construction-contractors/ — ~July 2026. 🟢

[5] Daily News Egypt — "Kuadra Launches AI Platform for Construction Sector", https://dailynewsegypt.com — ~July 2026. 🟢

[6] Intelligent CIO APAC — "APEC Positions AI and Digital Transformation at the Centre of Asia-Pacific Growth", https://www.intelligentcio.com/apac/2026/07/24/apec-positions-ai-and-digital-transformation-at-the-centre-of-asia-pacific-growth/ — 24 July 2026. 🟡

[7] Digital Watch — "APEC Adopts Chengdu Statement, Launches 2026-2030 AI Initiative", https://dig.watch — ~July 2026. 🟡

[8] Mastt — "State of AI in Construction Project Management 2026", https://www.mastt.com/research/ai-in-construction-project-management-2026 — ~July 2026. 🟢

[9] SNS Insider — "Construction Software Market Worth $32.89 Billion by 2035", https://www.globenewswire.com/news-release/2026/07/20/3329806/0/en/Construction-Software-Market-Worth-32-89-Billion-by-2035-as-Digital-Transformation-Reshapes-Project-Management-Report-by-SNS-Insider.html — 20 July 2026. 🟢

[10] PR Newswire — "Attentive.ai Launches Beam AI BIM CoPilot for Construction Firms", https://www.prnewswire.com — ~June 2026. 🟡

[11] WebWire — "Attentive.ai Announces Fully Managed BIM Coordination Service", https://www.webwire.com — ~June 2026. 🟡

[12] GlobeNewswire — "OFA Group Announces Strong Early Commercial Validation for AI Platform QikBIM", https://www.globenewswire.com/news-release/2026/07/14/3326788/0/en/OFA-Group-Announces-Strong-Early-Commercial-Validation-for-AI-Platform-QikBIM-as-Global-User-Adoption-Accelerates.html — 14 July 2026. 🟡

[13] Highways Today — "Generative Search Emerges as the New Gatekeeper for AI BIM Automation", https://highways.today/2026/07/24/generative-search-ai-bim/ — 24 July 2026. 🟡