The agent successfully assembled a comprehensive explanation of measure.so's value proposition, target customers, pricing, and differentiation by fetching the homepage, pricing pages, customer-specific landing pages, and comparison explainers. The site is moderately navigable for an AI agent—pricing and use cases are clearly published and discoverable from the homepage, but core product mechanics (algorithm, data sources, measurement refusal criteria) are intentionally opaque, forcing the agent to rely on inference and product marketing language rather than technical documentation.

• ›Steps [1, 3, 5] returned complete pricing tiers (Per Report, Solo, Pro, Portfolio, Enterprise, Lead Packs) with explicit per-property pricing. This core business model was immediately available and well-structured in fetchable HTML.
• ›Steps [9, 10, 11, 14] reveal a sophisticated multi-funnel routing architecture: the site explicitly segments customers by use case (contractors, developers, property managers, homeowners, institutional) with dedicated landing pages. This segmentation was discoverable but required following internal navigation links rather than being centralized on a single 'who is it for' page.
• ›Steps [15, 16] returned detailed comparison pages (vs. site visits, vs. surveys, vs. EagleView) that explicitly positioned Measure as complementary to competitors, not a replacement. These were well-researched and accurate but required the agent to proactively search for them rather than being linked from the main positioning.
• ›The agent noted a critical gap: ~30% measurement refusal rate is disclosed on /plans but the homepage provides no explanation of what makes addresses unmeasurable or geographic variation. The site is transparent about limitations but doesn't surface the reasoning.
• ›Technical architecture (algorithm, LiDAR/imagery sources, data refresh cadence, API capabilities) is absent from public pages. The agent had to acknowledge 'intentionally a black box' rather than finding documentation—the site does not publish this for agent consumption.
• ›Step [6] web search yielded unrelated results about generic product metrics, indicating Measure.so's specificity makes it hard for search engines to surface the site in general queries. The agent relied on prior knowledge and direct URL construction (steps [7, 15]) to reach content.