Canonical thesis · Sorbet

Boring tools.

AI agents need structured, durable state — records, ownership, status, history, permissions — to do real work. The 'boring' platforms that already model the world this way are quietly becoming the agent-economy substrate. We bet a basket of them compounds 25–30% IRR May 2026 → May 2029.

Nate B Jones · Anthropic might buy Atlassian for $40B

Why now

The shift is already in motion.

Five concrete reasons the next three years compound into the records-owners, not the wrappers.

  • OpenAI Symphony · Linear

    The Symphony spec uses Linear boards as the canonical task surface for autonomous coding workflows. Tools with rich records are already the agent control plane.

  • Atlassian MCP

    Atlassian's MCP server made its installed base of work state machine-readable and machine-writable. Acquisition rumours are strategically plausible — owning the SoR is owning the substrate.

  • Incumbent agent SKUs

    Salesforce, ServiceNow, Workday, Oracle, SAP have shipped agent SKUs in the last 18 months that monetise records they already own — not new content they have to build.

  • UX → data quality

    Linear's UX-driven cleanliness becomes a data advantage. Tools people enjoy generate cleaner state, which makes them better substrates than over-customised Jira instances people game.

  • Greenfield repositioning

    Greenfield 'agent platforms' are repositioning as thin wrappers over these substrates because they cannot replicate the records, permissions, or workflow engines.

Five-question diagnostic

The pillars the signal classifier scores against.

A tool is likely to become agent infrastructure when it answers yes to all five. Pillar-positive news drives add; two-or-more pillar negatives drive exit.

P1

Records, not content

Does the platform centre on durable records (tickets, accounts, employees) — not unstructured documents or chat?

P2

State machine, not labels

Are state transitions explicit, named, and enforced — not just tags on free text?

P3

Ownership, not implication

Is 'who is responsible' a typed field with permissions — not guessed from the message?

P4

Structural verbs, not conversational actions

Are domain operations modelled as named API verbs (close, reassign, escalate) — not buried in prose?

P5

Queryable history, not visibility

Can an agent read every state transition, attachment, and decision back through time via a stable API or MCP?

10 substrates

The buckets, in order.

Each bucket holds a Market Leader, a Challenger, and a Dark Horse at a 40 / 35 / 25 tilt. Each row opens the comparative breakdown.

  1. 01Dev workIssue trackers became the agent control plane for autonomous coding. Tickets are state machines; assignees, dependencies, and audit trails are the substrate.TEAM · MNDY · ASAN
  2. 02Source controlCode is the most durable record of system change; PRs are the canonical state machine. Whoever owns the source repo sits on every agent's commit history.MSFT · GTLB · FROG
  3. 03CRMThe original system of record. Accounts, opportunities, lifecycle stages — every agent revenue motion routes through one of these graphs.CRM · HUBS · FRSH
  4. 04ITSM & workflowIncidents, change requests, approval chains. The fattest agent-readiness narrative in enterprise software.NOW · PEGA · APPN
  5. 05ERPCash, inventory, fixed assets, the GL. The hardest, most regulated records in the enterprise — and the most defensible.SAP · ORCL · MANH
  6. 06HRIS & payrollPeople records, ownership, approvals, history. Compliance moat plus the most-touched record in any company.WDAY · PAYC · ADP
  7. 07SMB financeAP, AR, books — durable, compliant, queryable. Every SMB agent will be writing through one of these spines.INTU · BILL · PAYO
  8. 08Identity & permissionsPermission state IS the agent control plane. Without it, no agent can take a real action.OKTA · PANW · SAIL
  9. 09Data SoRQueryable history at scale. The substrate's substrate — every other SoR eventually replicates here.SNOW · MDB · ESTC
  10. 10Regulated & federalCompliance + records + auditability. Where agent trust gets earned and the moat is regulatory by construction.PLTR · VEEV · TYL
Structural claim

Four shifts that compound the bet.

  1. 01

    Records-rich platforms (Jira, Linear, Salesforce, ServiceNow, ERPs, source control, HRIS, ITSM, identity, data warehouses) become the backbone of agent systems.

  2. 02

    Content/chat-rich tools (email, Slack, generic docs) become secondary context sources that need expensive wrappers to feed records-rich systems.

  3. 03

    The boring tools win. CRMs, service desks, ERPs, calendars, source control, HRIS, finance — this is where meaningful agentic work actually happens.

  4. 04

    Incumbent moat compounds. Salesforce, ServiceNow, Atlassian, Microsoft, SAP, Oracle, Workday already own the records. Greenfield platforms either acquire one or wrap one.

The current gap · ADR 0016

Why we're down and still holding.

The market is reading the agent shift as software is dying — multiples have compressed across the SoR basket. Our read is the opposite. The user-facing interface tier is what commoditizes; agents will generate UIs on the go. The structured-data backbone underneath — records, ownership, history, permissions, the schema agents call into — becomes more valuable, not less. The 12-month price gap below is the cost of being early.

This frame is held under the four kill-criteria on the right. None has fired. The full reasoning is in ADR 0016.

Trailing window · 2025-05-05 → 2026-05-05

SoR equal-weight basket-25.99%
SPY+27.42%
QQQ+38.47%

Live from /backtests. The basket holds 30 SoR tickers equal-weighted; the engine layer on top of it is measured separately.

Kill-criteria · what would change our minds

  • 01

    Customer revenue actually declines at the SoR layer.

    Not multiple compression — actual top-line shrinkage at SoR holdings, sustained 4+ quarters, multi-name (5+ holdings simultaneously). Not explained by macro / FX / one-time churn.

    Watching · no SoR holding has shown sustained YoY revenue decline.

  • 02

    Agents observably acting without an SoR.

    Sustained, at-scale agent workflows that bypass the structured-data layer entirely (raw LLM memory + ad-hoc retrieval displacing real CRM/ITSM/ERP workloads). Anecdotes don't count; aggregate platform usage shifts do.

    Watching · no public evidence of platform-scale displacement.

  • 03

    An agent-native upstart materially eats one of the buckets.

    A single agent-first competitor takes meaningful share from SoR holdings within a bucket (Dev work, ITSM, CRM, etc.); the displaced incumbent doesn't recover. Bucket-specific, not portfolio-wide — triggers a bucket re-pick, not a thesis exit.

    Watching · no bucket has yet shown an upstart taking durable share.

  • 04

    5y rolling underperformance vs SPY exceeds −20pp annualized after 2027.

    A 12mo gap is noise on a 3y thesis. A 5y rolling gap of that magnitude is structural and forces the question: thesis right but basket wrong, or thesis wrong?

    Not yet evaluable · 5y rolling window opens 2027-05-04.

Forward-looking risks

The watch-list.

If two or more invalidators trigger, the daily scan generates exit proposals across the affected buckets and a portfolio-level rebalance toward leaders + the hyperscaler hedge.

Not

What this portfolio is not.

  • — A short-term trading strategy. Daily scans collect signal, not turnover.
  • — A diversified market portfolio. It is concentrated in SoR + adjacent infra by design.
  • — A robo-advisor. The chat declines personalised investment advice.
  • — A backtested fund. The clock starts on the actual seed date — no historical alpha is claimed.