Wedge

The beachhead targets commercial cinder block wall takeoffs for mid-sized masonry subcontractors. CMU walls follow strict structural grids, making the computer vision extraction highly deterministic and the initial proof of value immediate. From there, the product expands into architectural brick veneer, stone cladding, and finally integrates directly with materials distributors for automated procurement.

Timing

Multimodal LLMs and computer vision models now reliably extract complex schedules, symbols, and spatial dimensions from unstructured architectural PDFs. Simultaneously, a severe shortage of skilled estimators in the construction trades forces subcontractors to adopt automation to bid on enough projects to maintain pipelines.

Why This ICP

Masonry subcontractors operate with tight material margins and high labor costs, making accurate estimates critical to profitability. They also handle a highly repetitive, rule-based counting process for standardized units that is perfectly suited for computer vision extraction.

Size Of Prize

There are roughly 25,000 masonry subcontractors in the US who dedicate at least $40,000 annually to estimator labor and takeoff software. This yields an addressable market of approximately $1B for an automated estimating solution.

Gap Narrative

Masonry estimators perform manual block, brick, and mortar takeoffs from 2D PDF blueprints, a slow process prone to material calculation errors. Existing takeoff software requires manual tracing and lacks specific material yield logic for masonry, such as mortar waste factors and rebar spacing. An AI-native solution parses structural plans to automatically count units and calculate mortar volumes based on wall area and coursing schedules.

Defensibility

Defensibility compounds through workflow lock-in and a proprietary data moat of regional material pricing and labor productivity rates. As the system processes more successful bids, it fine-tunes its waste factors and cost databases, making its estimates increasingly accurate over time. Switching to a competitor means losing the custom-calibrated takeoff logic that matches the specific crew's historical performance.

Why This Thesis

A Service-as-Software approach fits perfectly because subcontractors want a completed takeoff and bill of materials, not another SaaS tool they have to train staff to use. By delivering the final estimate as the output, the solution bypasses software adoption friction and directly replaces outsourced estimating labor.

Hardest Part

The single hardest part is reliably extracting multi-layered wall assemblies like brick veneer with CMU backup from flat 2D PDFs and accurately deducting structural openings without hallucinating quantities.

Min Viable Scope

Focus exclusively on straight commercial CMU block and standard brick veneer wall extraction. Leave out complex decorative stonework, landscaping retaining walls, and non-masonry structural steel components.

Cold Start Problem

The vision models require thousands of matched architectural blueprints and verified material takeoffs to learn standard notations and spatial relationships. Break this by partnering with mid-sized masonry subcontractors, offering them free historical bid digitization in exchange for their past plan-to-takeoff data.

Time To First Value

Immediate upon uploading the first drawing set, followed by roughly one hour of human verification against the generated material list.

Data Moat Available

true

Technical Difficulty

High

SAM

~$100M-150M commercial masonry subcontractors performing complex multi-wall bids

SOM

~$10M-15M

TAM

~30,000 North American masonry and concrete contractors × ~$10,000-12,000/yr ≈ ~$300M-360M

Growth Rate

~10-14%/yr, driven by an acute shortage of skilled construction estimators and margin compression requiring tighter bid accuracy

Paid Comparable Spend

~$70,000-95,000/yr per human estimator salary, plus ~$1,500-3,500/yr per seat for generic legacy digital takeoff tools

Kill Thresholds

• Manual correction rate > 15% of total material counts after 30 days
• Willingness to pay < $400/month during initial sales calls
• D30 retention < 40% for active estimators
• Customer acquisition cost > $3000 within the first 90 days

Leading Metrics

• Time-to-first-value from PDF upload to editable takeoff sheet
• Manual correction rate on block and brick quantities
• Number of wall elevations accurately detected per blueprint
• User-reported bid output volume per week

What Proves Right

Commercial masonry subcontractors upload multi-wall PDF blueprints and accept the extracted block, brick, and mortar counts with under five percent manual adjustment. Customers convert to an $800 monthly subscription after completing three viable bids using the tool. Week-four retention exceeds 80 percent because the platform condenses the takeoff process from four days to under four hours.

What Proves Wrong

Estimators abandon the platform for custom Excel workbooks because the extraction engine hallucinates elevation changes or miscalculates mortar yield ratios on structural block. Users spend more time auditing the generated takeoff than they would manually counting units in PlanSwift. Customers refuse to pay beyond the $200 monthly rate of generic takeoff tools, restricting the ACV below viable margins.