Snowflake
Sorbet currently classifies Snowflake as the leader in its Data SoR bucket. The detailed selection note is not yet available.
No recent material read
Based on 0 material signals in the last 30 days: 0 positive, 0 negative.
Evidence summary, not a buy / sell rating
Latest approved decision: hold · 2026-08-29
The role and pillars explain what job this holding has inside the System-of-Record basket—not whether it suits an individual investor.
Snowflake is the portfolio's Data SoR leader because it owns the queryable, time-travel-enabled layer that every other system of record eventually replicates into — making it, in the thesis's own framing, the substrate's substrate. Its core durable records are the structured tables, schemas, and immutable query histories that accumulate as downstream copies of CRM, HRIS, ERP, ITSM, and source-control data land in a single governed platform. This directly embodies P4 — the platform exposes named, composable SQL and Snowpark verbs that agents can invoke deterministically — and P5, because Snowflake's Time Travel and zero-copy cloning give agents a stable, versioned API into every historical state transition across the enterprise data estate. As the bucket leader, Snowflake carries the heaviest allocation weight within Data SoR, reflecting the conviction that the platform commanding the broadest cross-SoR query surface accrues the most durable moat as agent workloads shift from human-driven dashboards to continuous, machine-issued analytical queries at scale.
The bucket's primary paper position: the company Sorbet currently treats as the clearest expression of this workflow category.
The thesis properties this holding is selected to test. These labels are portfolio classifications, not standalone proof.
Structural verbs
Queryable history
The thesis for Snowflake breaks on structural grounds, not price action. The clearest falsifier is hyperscaler bundling: if Microsoft Fabric, Google BigQuery, or Amazon Redshift begin offering equivalent cross-cloud queryable history as zero-marginal-cost line items inside existing enterprise agreements — and Snowflake's net revenue retention falls below 115% for two consecutive quarters as a direct consequence — that signals pricing-power erosion at the substrate layer, not a temporary growth pause. A second falsifier is failure to monetise the agent SKU: Snowflake's Cortex and data-sharing primitives are the P4/P5 surface agents depend on; if Cortex AI consumption ARR stalls or is de-emphasised in favour of generic compute revenue, the platform's claim to be the queryable-history substrate weakens materially. Third, open-source columnar warehouses (DuckDB, Apache Iceberg-native stacks, self-hosted Databricks alternatives) gaining enterprise deployment at scale — evidenced by a sustained decline in Snowflake's share of new data-warehouse workloads in third-party surveys — would erode the long-tail pricing power the thesis requires. An observer should watch for any two of these signals converging: NRR compression, Cortex ARR miss, and accelerating workload migration to open formats, because that combination indicates the substrate is commoditising faster than Snowflake can re-monetise through the agent layer.
Daily closes are shown for context. Reference levels are deterministic outputs of Sorbet's published ATR method, not analyst targets or advice.
2025-07-28 → 2026-08-31
$331.43
Short-term levels use 14d ATR and the 200d average. The 3y projection compounds the current revenue-growth rate with no multiple expansion. Missing inputs stay blank.
Revenue growth + FCF yield, not the conventional FCF-margin formulation. Visible for judgment, but not wired into selection.
Price is more than 50% above the 200d average; the model treats new adds cautiously.
| Quarter | Revenue YoY | Gross margin | FCF margin | Operating margin |
|---|---|---|---|---|
| 2026-01-31 | 25.97% | 66.80% | 59.45% | -26.48% |
| 2025-10-31 | 25.76% | 67.77% | 9.37% | -27.21% |
| 2025-07-31 | 31.33% | 67.53% | 4.97% | -29.72% |
| 2025-04-30 | 26.10% | 66.53% | 17.60% | -45.46% |
| 2025-01-31 | 27.26% | 66.23% | 42.10% | -39.19% |
| 2024-10-31 | 27.38% | 65.94% | 8.30% | -39.34% |
Snowflake's trajectory in the agent economy runs through its ability to deepen the role it already plays: the queryable, governed archive that every upstream system of record eventually replicates into. To compound as agent-economy substrate, it must execute on three fronts simultaneously — maturing Cortex AI so that agents can reason over stored records without data leaving the platform, expanding its MCP and API surface so that orchestration layers can issue structural verbs against Snowflake state rather than just running ad-hoc SQL, and hardening its governance primitives (row-level security, fine-grained access, audit trails) to the point where a customer can safely grant an agent read-write credentials without human review of every query. What remains unproven is whether Snowflake can defend the data-layer moat as Databricks, BigQuery, and Redshift all make equivalent agent-readiness claims: the pillar-five advantage (queryable history via a stable API) is real but not yet exclusive, and consumption-based pricing means that revenue scales with agent workload only if Snowflake wins the query, not just the storage contract. The next 18 months are therefore an execution test on Cortex adoption rates and on whether the Snowflake Marketplace becomes the preferred exchange for agent-accessible, governed record sets — not merely a data-sharing novelty.
No material signals for SNOW in this period.
Earnings in 1 day(s). Defer trim/add proposals through the print.
Earnings in 0 day(s). Defer trim/add proposals through the print.