Custom-embedded 3-year TCO converges on $300K-$630K (multiple, mostly vendor sources) with 20-30% annual maintenance; hidden costs are multi-tenancy/RLS, performance at scale, perpetual maintenance

Claim: Multiple (mostly vendor) sources converge on a 3-year TCO for building production-grade, multi-tenant embedded analytics in the $300K-$630K range (some cite far higher), with ~20-30% annual maintenance, vs. buying at a few thousand dollars/month. The recurring honest point even from vendors: building costs ~3× more than teams expect, and hidden costs are multi-tenancy / row-level security, performance at scale, and the perpetual maintenance tail.

Source: Synthesis from embedded-analytics vendor literature.

Confidence: Industry-consensus — but all major sources are embedded-analytics vendors with incentive to favor "buy."

Caveat: Vendor-modeled, not audited. Skews toward buy. Still: even discounted for incentive, custom-embedded at $100K-300K + 20-30%/yr is the realistic floor.

Why this matters for Candid: Custom-embedded should be reserved for analytics-as-core-differentiation + ≥1-2 engineers to own it permanently. For any other case, the embedded-analytics platforms (Embedded-analytics platforms (2026): Luzmo $995-3,100/mo embedded plans; Explo $795/Growth, $2,195/Pro; Cube Core open-source + Cube Cloud consumption; Embeddable still in beta) win.