Strategic Brief, Decision-Layer Architecture for Agentic AI

The Trust Layer: Built to Reason

A strategic brief defining Aphrodite's product strategy, market footprint, go-to-market motion, and delivery model across the Data Platform, Data Services, and Agentic Activation.

Project Aphrodite
Status Draft, In Progress
Products Data Platform · Data Services
Agentic Activation
Last Updated June 2026

01
Core Thesis

The future belongs to those who can turn an operational data moat into something agents can reason on, decide from, and act through. The model layer commoditizes; the trust layer doesn't. That turn requires semantic infrastructure, delivered through a domain-specific capability, with scaled execution and global reach.

No single asset reaches Era Three alone. Data at depth does not reason on its own. Semantic infrastructure has nothing to stand on without data beneath it. Together they become the Platform Value Office spanning the full arc: mobilizing data from operating partners, refining it into resolved, governed, semantic infrastructure, then activating agents on top.

Products stop at the silo. The services business does not: across asset classes, across geographies, into white space no incumbent has reached.

Three eras of real assets software

Real assets software has been built around one consumer at a time, in three eras. The first two are largely complete. The third is open. The combined entity is what consolidates Era One leadership with the foundation Era Three runs on.

Era One, 1984 to 2005

Built to capture

Consumer: Finance, accounting, audit.
Architecture: ERP and systems of record.
Exemplars: Yardi, MRI, RealPage (OneSite, RUM, Knock, G5), Entrata, AppFolio. Captured transactions, enforced workflows, stored operational data. The system of record served the human reading the ledger.

Era Two, 2005 to 2024

Built to inform

Consumer: Analysts via BI and dashboards.
Architecture: Data warehouse, cloud SaaS, analytics.
Exemplars: Snowflake, Databricks, Cherre Connect, sector analytics platforms. Aggregated and presented data for human interpretation. The dashboard served the human reading the report.

Era Three, 2024 and forward

Built to reason

Consumer: AI agents, autonomous workflows.
Architecture: Knowledge graph, context graph, decision graph.
Exemplars: Cherre's CONNECT, CORE (Luna · Meridian), and ALPHA (ATLAS · Agent STUDIO); RealPage's Lumina; agentic platforms still forming. The foundation serves the agent: reason against, decide from, act through. RealPage operates at Era One and reaches into Era Two. The combined entity consolidates Era One leadership with Era Three architecture.

The AI agent is not the next user. The agent is a different kind of consumer. It does not read reports; it reads data directly. It does not infer context from the conversation in the room; the context has to live in the architecture. It does not remember what was decided last quarter; the memory has to be queryable. Era Three software is what real assets architecture looks like when the consumer is the agent.

Artifact 1, the structural advantage shift

Three eras of real assets software, illustrative

ERA ONE ERP Compressing ERA TWO Analytics Commoditizing ERA THREE Agents Open, no incumbent STRUCTURAL ADVANTAGE TIME, COMPLEXITY OF THE CONSUMER ERP-era advantage Analytics-era advantage Era Three advantage (emerging) TODAY THE VOID, WHITE SPACE No incumbent owns the foundation layer for real assets Analytics displaces ERPs Agents displace analytics
ERP-era advantage
Analytics-era advantage
Era Three advantage (emerging)

Each era's advantage compresses as the next one opens. The combined entity sits at the Era Three inflection: an established Era One operator footprint (RealPage), the Era Two analytics layer that has not solved real assets at foundation depth, and the open Era Three foundation the combined entity is built to own.

The defensive case: reproducibility is the next audit standard

Audit committees, regulators, and IC processes are starting to ask a question they did not ask a few years ago: how was this decision made, by which model, against which version of which definition, with what supporting evidence. Today the question is answered with shrugs and post-hoc reconstructions. Within a short horizon the answer becomes an audit finding. The shift from "the report was correct" to "the reasoning was reproducible" is structural; it will not reverse with any administration's regulatory posture, because the fiduciary obligation driving it is older than the regulators. Institutions that cannot reproduce their AI-assisted decisions absorb the reproducibility gap as a fiduciary risk before it becomes a compliance one. The infrastructure that produces reproducibility is not the infrastructure that produces good dashboards. The two have been confused. They are not the same thing.

The structural problem

The data mobilization gap

Operating partners generate the data. GPs and LPs need to consume it: quickly, accurately, and at scale. The gap between generation and consumption is filled today by manual processes, brittle spreadsheet pipelines, and point solutions that don't talk to each other. Data Platform and Data Services are built to close this gap at the infrastructure layer.

The AI opportunity

Real assets is ripe for agentification

Industry leaders like Brookfield and Blackstone have already started the agentic journey. The rest of the market is catching up at varying levels of data and AI maturity. Firms that get their data infrastructure right now will be the ones who deploy AI meaningfully. Those that don't will be locked out when the window closes.

The answer

FDEs meet clients where they are

Forward Deployed Engineers (FDEs) offer readiness assessments, data engineering, agentic design and build, and optimization services. This is not a one-size-fits-all product play. It is a high-touch, expert-led capability that unlocks value at every point on the client maturity curve, from data clean-up to full agentic deployment.

The market maturity curve

Maturity level Where most firms are today What they need Entry product
Early, Data chaotic Manual collection, fragmented systems, no single source of truth for fund data Data mobilization, standardization, reliable pipeline from operating partners Data Platform
Developing, Data structured Data lives in systems but is not consistently validated, enriched, or governed Ongoing management, QA, and enrichment at scale without internal headcount burden Data Services
Advanced, Data ready Clean, structured data in place; executive pressure to get value from AI FDE use case design, deployment, and ongoing optimization Agentic Activation
Leaders, AI native Brookfield, Blackstone, actively building AI capabilities internally Co-development, frontier capability deployment, optimization of existing agents FDE Specialist Track

Why this converges

02
What We're Building

This work has a name and a home: the Platform Value Office. Three directions of value: back to the software business, across asset classes, across geographies. Two capability vectors operate beneath it, forming a closed loop.

→ Action

Both vectors stand up in Phase 01. Vector 1: the FDE cohort is hired in Q1. Vector 2: Data Platform onboarding recombines existing Cherre and RealPage teams; Data Services Support is built new under unified CX leadership. Phase 02, which opens the EMEA pod, does not begin until Phase 01 clears. See the integration sequence, Section 14.

Vector 1

Forward Deployed Engineering for AI

Build and scale an FDE capability to support AI readiness assessments (where is the client on the maturity curve?), AI agent deployment (design, build, and go-live of agentic workflows), and client optimization of AI tools including ATLAS. FDEs are technical, embedded, and outcome-accountable. They are not a support function.

Vector 2

Data Platform Deployment & Data Services

Scale a deployment and support capability for Data Platform that covers technical implementation (onboarding, integration, data pipeline setup) and ongoing support through Data Services. This is the operational backbone that sustains the data quality clients depend on before and during AI deployment. Data Services is the function where the combined entity inherits the largest asymmetry: Cherre delivers Data Services through partner relationships rather than natively; RealPage owns a native Data Services practice run from in-house delivery centers in India and the Philippines. Phase 01 work integrates the owned delivery centers and Cherre's partner network as a managed delivery portfolio: owned capacity for predictable, repeatable workloads at scale; partners for surge, regional reach, and specialty coverage. The mix is actively managed.

Vector 1, FDE for AI
Drives adoption and expansion. FDEs unlock Agentic Activation value and pull clients toward higher-value contracts and deeper integrations
+
Vector 2, Data Platform / Data Services
Sustains the data foundation. Without clean, managed data flowing through Data Platform, Agentic Activation cannot perform and FDE work stalls before it starts

Building sequence, what gets built first

Phase 01, 0 to 6 months
Stand-up
  • Define FDE hiring profile and 90-day ramp model
  • Establish Data Platform onboarding playbook from existing deployments
  • Launch Data Services as a packaged, contractable service offering
  • Run first AI readiness assessments with anchor clients
  • Instrument the product ladder; track expansion signals from Data Platform clients
Gate to Phase 02
First FDE-led agent deployment live; Data Services Annual Recurring Revenue (ARR) at threshold; Data Platform onboarding at target velocity
Phase 02, 6 to 18 months
Scale
  • Scale FDE team regionally (NA → EMEA)
  • Productize the AI readiness assessment: fixed scope, repeatable output
  • Build FDE for Agents specialist track with its own hiring and comp model
  • Data Platform Onboarding Team at capacity; Data Services Support Team established
  • First EMEA fund admin channel relationships active
Gate to Phase 03
Regional FDE leads in place; Agentic Activation reference clients live in NA and EMEA
Phase 03, 18 months+
Global
  • APAC FDE presence (Singapore hub)
  • Agentic Activation at scale across all Ideal Customer Profile (ICP) segments
  • In-house Data Platform data engineering capability; partner model for niche sourcing only
  • FDE as a durable competitive moat in every active market
  • Cross-regional knowledge transfer cadence established
Steady-state measure
ARR per FDE, Net Revenue Retention (NRR) in FDE-covered accounts, agent deployments live, Data Platform go-live velocity
03
What the Stack Must Install

The vectors set the strategy. The stack is what the strategy has to install underneath. For thirty years the data layer of institutional capital was built around one consumer, the human reading a report. The reporting stack worked because the human did the integrating, interpreting, contextualizing, and remembering, invisibly, every time. The agentic stack does not have a human in that position. The cognitive work that always existed becomes a set of codified layers.

→ Action

Build the foundation codified layers (03 to 05) once, as shared infrastructure reused across clients; build only the reasoning layers (06 to 08) per engagement. CORA participation, the industry coordination of that shared layer, is a Phase 01 gate. See Section 14.

From the reporting stack to the agentic stack

Systems of record do not change. Storage does not change. The consumption layer does not change in role. Its consumer does. What gets added are the layers that make explicit the cognitive work the human used to perform invisibly. Three classes carry through the stack: Persistent, what survives unchanged from the reporting era; Transformed, mechanical transport shed of the semantic arbitration that used to hide inside ETL; and Codified, what the human used to do invisibly, now made explicit because the agent has no head.

Read top-down: the consumer is layer 09; systems of record are layer 01. Click any row for the gap today and what is still missing at scale. The badge on each row carries two lenses: the architectural class and the product line that owns it. Data Platform owns layer 02 across three flows in: Submissions, Pipelines, Market Data. Layers 01 and 09 are client-owned.

Persistent 09 Action surface +
From the reporting stack

Was "Consumption." Same layer; the consumer expands from human to human + agent.

Gap today

Consumers are display-shaped. Architecture assumes a human reading a chart; agents need graph traversal, semantic queries, lineage retrieval.

Still missing at scale

·

Codified · Agentic Activation 08 Reasoning surface +
From the reporting stack

Was implicit. The human composed across entity, relationship, definition, context, and memory to produce reasoning.

Gap today

AI is bolted on. Models wrap around the dashboard. They see the outputs the human reads, not the foundation that produced them.

Still missing at scale

A reasoning surface agents can rely on. A stable interface that survives schema, vendor, and platform churn downstream.

Four reasoning models FDEs deploy against the foundation
Causal
"Why did renewal rates drop?"
Traverses the graph from the metric through operational events, PM transitions, and market conditions to surface the root factor.
Comparative
"How does Asset A compare to Asset B?"
Normalizes metrics via the semantic layer and aligns entity dimensions before comparing.
Predictive
"Which assets are likely at risk?"
Detects context-graph patterns that preceded prior defaults or underperformance and scores current exposure.
Diagnostic
"Why is this metric inconsistent?"
Traces lineage through semantic definitions and isolates where meaning diverges across sources.
Codified · Agentic Activation 07 Decision graph +
From the reporting stack

Was implicit. The human remembered what was decided last quarter and why it made sense.

Gap today

No decision lineage. Approvals, overrides, sign-offs live in emails, chat, IC minutes. None of it structured or queryable.

Still missing at scale

A decision layer that is queryable. Capturing decisions as events with standard schemas is tractable. The reason it isn't solved is not technical. Operational and political.

Codified · Agentic Activation 06 Context graph +
From the reporting stack

Was implicit. The human held situational context: what was active when this happened, what definition was in force.

Gap today

Context is reconstructed at query time. No native operation joins knowledge, events, and semantics; the work falls to whichever consumer asks the question.

Still missing at scale

A context graph that operates at population scale. Each firm's context graph stops at the firm's walls. The institutional memory worth having extends beyond any single firm.

Codified · Data Services 05 Semantic layer +
From the reporting stack

Was scattered. Definitions lived in modeling code, BI tools, analyst heads, IC memos.

Gap today

Semantic logic is scattered and unversioned. Same metric, multiple implementations. A definition active a year ago is not retrievable today.

Still missing at scale

A semantic registry that survives versioning. Today's semantic layers do not survive across tools, let alone across firms.

Codified · Data Services 04 Knowledge graph +
From the reporting stack

Was the human's mental picture of how entities related.

Gap today

The data is not graph-shaped. Real estate is a graph (entities and relationships) stored as tables. Multi-hop reasoning becomes expensive joins.

Still missing at scale

A canonical graph the industry agrees on. Each firm models its own; the graph of real estate itself does not exist at population scale.

Codified · Data Services 03 Entity resolution +
From the reporting stack

Was implicit. The human reconciled "this building" across sources at query time.

Gap today

No persistent canonical identity. Entity resolution happens at the BI or modeling layer, ad hoc, query by query. The same property has multiple IDs.

Still missing at scale

A canonical identity layer every consumer trusts. Each firm builds its own, often poorly. No shared canonical layer exists at population scale.

Transformed · Data Platform 02c Market Data +
From the reporting stack

Curated public and licensed feeds: comps, demographics, property records, market signals. Pre-resolved against client portfolio entities before they hit the foundation.

Gap today

Industry buys market data and runs entity matching at query time, every time. Pre-resolution at the ingest layer collapses query-time work and produces consistent answers across consumers.

Still missing at scale

Public and licensed feeds joined to portfolio entities as a continuous service, not a project. The combined entity inherits this from Cherre.

Transformed · Data Platform 02b Pipelines +
From the reporting stack

Direct connectors to ERP, deal management, leasing platforms, and warehouses. Pre-built integrations across the leading systems of record in real assets.

Gap today

Every client builds and maintains its own connectors. The connector library is the cost the industry pays repeatedly because no one consolidates it.

Still missing at scale

Connector library as a managed asset, not a project deliverable. The integration cost compounds in favor of the firm that runs it once for the industry.

Transformed · Data Platform 02a Submissions +
From the reporting stack

Operating-partner and third-party data collection. Map sources, define rules, automate ingestion, standardization, and validation.

Gap today

Submission flows run through email, FTP, and spreadsheet templates. State changes are overwrites. The audit trail is "what was the value at month-end snapshot," not "what happened, when, why, in what order."

Still missing at scale

Schema-validated, append-only ingestion with row-level provenance. The layer as a contract surface, not a connector library.

Persistent 01 Systems of record +
From the reporting stack

Unchanged. Property management, fund accounting, valuation systems. The authoritative ledger for transactions in both eras.

Gap today

·

Still missing at scale

·

The deployment question (what gets built inside the firm, what gets consumed from the shared canonical layer, and the one layer that sits locally but depends on the canonical layer beneath) is answered in Section 10 Delivery, where the nine layers collapse into three deployment buckets.

04
What We're Selling

Three products across three commercial models. Each serves a different point on the client maturity curve and generates a different type of revenue: recurring SaaS, data-as-a-service, and the emerging service-as-software model for agentic AI.

→ Action

Service-as-software is the unproven model of the three. The Platform Value Office owns Agentic Activation pricing and packaging, defined in Phase 01 before the first FDE engagement is scoped. SaaS and data-as-a-service keep their existing motions.

"Agentic Activation is not Software as a Service. It is Service as Software. Forward Deployed Engineers delivering outcomes that were previously produced by analyst teams and professional services organisations."

Product 01 · Software as a Service

Data Platform

The data-in surface for the combined entity: three flows that land in a single governed foundation. Submissions, Pipelines, and Market Data are the inputs Data Services resolves and Agentic Activation reasons against. Priced as recurring SaaS; the foundation on which Data Services and Agentic Activation sit.

Submissions
Operating-partner and third-party data collection. Map sources, define rules, automate ingestion, standardization, and validation.
Pipelines
Direct connectors to ERP, deal management, leasing platforms, and warehouses. Pre-built integrations across the leading systems of record in real assets.
Market Data
Curated public and licensed feeds, comps, demographics, property records, market signals. Pre-resolved against client portfolio entities and ready to query alongside operational data.
Product 02 · Data as a Service

Data Services

The foundation beneath the platform. Data Services resolves entities (Luna), governs the knowledge graph (Meridian), and maintains the semantic layer. It sustains data quality over time across operating partner changes, new sources, and evolving schema. Priced as a managed service contract; high retention, high-margin recurring revenue.

Cherre-inherited entity resolution at scale
2B+
Addresses
130M+
Buildings
100M+
Parcels
160M+
Tax records
12B+
Graph relationships
Product 03 · Service as Software

Agentic Activation

The combined entity's reasoning surface, where agents reason, decide, and act on operational truth, delivered as outcomes rather than software seats. Three components inside one product line: Agent STUDIO and ATLAS as the agent platform; the MCP server as governed context delivery to any model; FDE-led activations as the engagements that put agents in production. Forward Deployed Engineers are the delivery team. Value is measured in outcomes delivered, not seats sold.

Agent STUDIO + ATLAS
The platform. Agent STUDIO to build, deploy, and operate AI agents on the governed graph; ATLAS as the chat-based reasoning agent for governed answers with full lineage. Reasoning agents on production infrastructure, not prompts wrapped around spreadsheets. The model layer commoditizes; the trust layer underneath does not.
MCP server
Any AI, any model. Cherre's MCP server delivers governed context to Claude for Enterprise, ChatGPT Enterprise, and any MCP-aware agent. The trust layer the model layer queries.
Agentic Activations
FDE-led engagements that take a client from readiness assessment to first agent in production and onward through optimization. The work that turns the platform into outcomes.

Four reasoning models

The agents Forward Deployed Engineers deploy operate against four reasoning models, each tied to a class of operational question. Each connects the trust layer to operational decisions, applied across every deployment.

Causal
"Why did renewal rates drop?"
Comparative
"How does Asset A compare to Asset B?"
Predictive
"Which assets are likely at risk?"
Diagnostic
"Why is this metric inconsistent?"

Product ladder, the land-and-expand model

Stage Product Value delivered Expansion trigger
Land Data Platform Reliable data pipeline from operating partners; single source of truth for fund data Data volume grows or operating partner base expands. Data Services conversation opens
Expand 1 Data Services Data quality sustained at scale without internal headcount burden on the client Client asks "what can we do with this data?" AI readiness signal is live
Expand 2 Agentic Activation AI agents producing portfolio analytics, investor reporting automation, anomaly detection Agent outputs drive new use cases; cross-sell into adjacent data domains or asset classes
Optimize FDE Ongoing Continuous optimization of agents, new use case delivery, co-development on emerging capabilities Sustained partnership. FDE embedded in client data and AI roadmap; highest retention
05
Where We're Selling

Three regions, sequenced by market maturity, existing relationships, and go-live complexity. North America is the anchor. EMEA enters via the UK, with fund admin relationships in Luxembourg and the Netherlands as the continental wedge. APAC follows with Singapore and Australia as the entry points.

? Open question

Which EMEA fund admin relationships are in-flight today that should inform the sequencing of UK GM hiring?

Region 01, Anchor
North America
  • United States, primary market; deepest Cherre and RealPage relationships; existing AIM customer base anchors warm pipeline
  • Canada, strong LP and pension fund concentration; CPPIB, OTPP, OMERS in ICP
Investment priority
Maximum, anchor market for all three products and full FDE deployment
Region 02, Strategic
Europe
  • UK, London hub; GP and fund admin concentration; Phase 01 EMEA entry
  • Germany, industrial real assets; open-end fund structures; Data Platform fit
  • France, large institutional LP base; Axa IM, Amundi RE in ICP
  • Luxembourg, fund domicile; fund admin is the entry channel
  • Netherlands, pension LP concentration; APG, PGGM as anchor targets
Investment priority
Strategic, UK direct; Lux/NL via fund admin channel; DACH partner-assisted
Region 03, Emerging
APAC
  • Singapore, regional hub; GIC, Temasek, sovereign wealth in ICP
  • Australia, large superannuation funds; strong real assets allocation; AustralianSuper, QSuper in ICP
Investment priority
Phase 03, Singapore hub model; partner-assisted initial coverage in Australia; less mature competitive landscape, first-mover advantage is real; data localization requirements require early legal and infrastructure planning

Entry model by market

Market Entry model Regulatory / complexity note Priority product
United States Direct, existing team Existing legal entity; no additional compliance burden for SaaS delivery Full suite (Data Platform + Data Services + Agentic Activation)
UK Direct, Phase 01 EMEA English-first; UK GDPR compliance required; FCA considerations for financial data Data Platform + Data Services; Agentic Activation in Phase 02
Luxembourg / Netherlands Fund admin channel GDPR; fund domicile complexity; sell through existing fund admin relationships Data Services (fund admin-led)
Germany / France Partner-assisted Language complexity; local data sovereignty; GDPR Article 46 transfer rules Data Platform; Agentic Activation in Phase 03
Singapore Hub model, Phase 03 MAS regulations; PDPA; data localization requirements growing Data Platform + Agentic Activation (sovereign wealth ICP)
Australia Direct, Phase 03 English-first; Privacy Act 1988; superannuation fund regulatory environment Data Services + Agentic Activation (superannuation ICP)
06
Who We're Selling To

Four buyer archetypes, organized by position in the capital stack. Each archetype has a distinct door, buying motion, and conversion cadence. The Phase 01 anchor account (see Section 06) is a sovereign-scale LP engagement that brings 900 global operating partners into contact with a governed data foundation. Those 900 partners are the Phase 02 pipeline; each one is a GP, Asset Manager, or Operator in its own right, addressable through the foundation they already touch. The Institutional Asset Manager archetype is the primary direct-sales pipeline running concurrent with the anchor hub; the GP and Operator archetypes convert fast-cycle in parallel; mid-tier LPs follow as the LP archetype's broader pipeline from Phase 02. Warm pipeline from AIM, IMS, Cherre, and RealPage relationships is the channel into each archetype, not the archetype itself.

? Open question

Which accounts in the existing book carry mandates that already depend, in practice, on the foundation the combined entity provides, even where the requirement is unstated?

Buyer archetype Door (primary product) Buying signal Buying motion Conversion heat
Limited Partner
Sovereign wealth, large public pensions, endowments, insurance allocations; GIC, Temasek, CPPIB, ADIA, NBIM as the leading edge.
Cherre clients: OTPP, New York Life. In discussions: HOOPP.
Data Platform and Data Services; Agentic Activation in long arc CIO data commitment; manager rationalization; in-house data team build; cross-asset mandate CIO-led, often by committee with Head of Investment Operations. Long cycle, large contract values, heavy governance. 12 to 18 month cycle for sovereign-scale; 18 to 24 months for mid-tier. Highest for sovereign-scale
Medium for mid-tier
General Partner
Large real estate, infrastructure, and specialty sponsors; single-strategy or focused-platform firms below the multi-strategy mega-AM tier.
Cherre clients: Starwood Capital, Savills IM. In discussions: Almanac (Neuberger Berman), QuadReal, PEC.
Data Platform, ladder to Data Services Recent fund close above target; LP advisory pressure on reporting; growing operating partner base; fund admin RFP in flight CFO-led. Faster cycle than Asset Manager. Less procurement weight. 3 to 6 month cycle. High priority
Institutional Asset Manager
Multi-strategy alternatives mega-firms; Brookfield, Blackstone, GIP, Stonepeak, Carlyle, Apollo, KKR, Ares, Macquarie, Nuveen as the archetype profile.
Cherre clients: Brookfield, Nuveen, DWS, Sculptor (Rithm). In discussions: ICG.
Data Platform and Data Services, ladder to Agentic Activation Multiple operating partners; cross-asset portfolio; recent CDO or Head of AI hire; LP base demanding richer reporting Multi-stakeholder. COO or CDO sponsored, CEO signature, CFO blessed. 6 to 12 month cycle. Highest priority
Operator with institutional capital
Greystar, AvalonBay, Equity Residential, Prologis, institutional-JV operators; RealPage's existing top-tier book
Agentic Activation via Lumina; Data Platform if institutional sponsor agrees Existing RealPage relationship; institutional ownership; visible AI initiative; operational data depth COO-led. Fast cycle at the operating leadership level. 2 to 4 month cycle. High priority for Agentic Activation
Medium for Data Platform/Data Services

The largest sovereigns and reserve funds are the leading edge of the LP archetype. They have internal data sophistication, cross-asset mandates by construction, and procurement capacity to commit to foundation-grade infrastructure. Mid-tier LPs follow the leading edge by 18 to 24 months. The Phase 01 anchor account (see Section 06) sits in this segment; the 900 operating partners that report into its governed data foundation become a directly addressable Phase 02 pipeline at the GP, Asset Manager, and Operator levels. The dual cadence inside the LP archetype is structural, not a sequencing choice.

Phase 01 anchor
The single Phase 01 anchor is a sovereign-scale leading-edge LP. 900 global operating partners reporting into a governed data foundation. Late-stage qualification, committee-led with COO and Head of Investment Operations as named champions (see Section 06). The foundation landed at the anchor becomes the platform of record for every operating partner that reports into it.
+
Phase 02 motion
GIC's 900 operating partners are the Phase 02 pipeline, each addressable as a GP, Asset Manager, or Operator in its own right through the foundation they already touch. Concurrent: the Institutional Asset Manager archetype runs as direct-sales primary pipeline supported by AIM, PRODA, and Cherre warm relationships; the GP archetype converts fast-cycle through CFO buyers; the Operator archetype runs through the RealPage book on Agentic Activation. Mid-tier LPs follow as the sovereign-scale work seeds the segment.
07
How We're Selling

A direct enterprise sales motion anchored in existing relationships, co-sold by FDEs who demonstrate technical value early in the cycle. The product ladder drives expansion: land with Data Platform, grow through Data Services, graduate to Agentic Activation. Fund administrators and real assets consultancies are the high-leverage channel partners.

→ Action

Three entry paths into Phase 01 sales cycles, sequenced by client maturity: FDE-led AI readiness assessment for AI-ready clients; Data Platform onboarding for clients with data fragmentation; Data Services managed contract for clients with capacity gaps. The FDE-led assessment remains the default opening move but is not the only one.

Primary motion

Direct enterprise + FDE co-sell

AE owns the commercial relationship and pipeline. FDE engages early, often before a formal deal, to run a readiness assessment and demonstrate technical depth with the client's data and technology teams. The AI readiness assessment is the opening move: low-commitment for the client, high-signal for the combined entity, immediately differentiating.

Expansion motion

Product ladder-driven land-and-expand

Data Platform is the land. Data Services is the first expansion. Agentic Activation is the graduation. Each step is triggered by a client signal: growing data volume, internal headcount pressure, or an executive asking "what can AI do for us?" AEs and FDEs must be trained to recognize and act on these signals before a competitor does.

Channel motion

Fund admins and consultants as multipliers

Fund administrators (who touch every one of their clients' data flows) and real assets consultancies are high-leverage channel partners. A single fund admin relationship can introduce Data Platform and Data Services to their entire client base. Partner enablement requires patience and structure; the economics are compelling at scale.

Buyer map, who is in the room by product

Buyer persona Primary product Core concern Winning message
CTO / Head of Data Data Platform, Data Services Pipeline reliability; reducing manual data work; integration complexity with operating partners Medallion architecture; proven ingestion at scale; technical depth of FDE team demonstrated in assessment
CFO / Head of Finance Data Platform, Agentic Activation Reporting accuracy; LP data requests at scale; audit trail and data provenance Time saved on LP reporting; automation of existing manual reconciliation processes
Head of Portfolio Management Agentic Activation Faster portfolio insights; anomaly detection; decision support without adding analyst headcount Agent demos with their own data; reference clients in the same vertical and asset class
Fund Administrator (ops) Data Platform, Data Services Processing volume; client Service Level Agreement (SLA) fulfilment; reducing manual reconciliation and data cleaning Data Services eliminates their reconciliation burden; channel economics for referrals
Managing Director / CIO All products Competitive positioning against peers; AI strategy credibility; board-level narrative on data and technology Brookfield / Blackstone comparison; agentification as a competitive moat; FDE as the deployment proof point

GTM sequencing, what opens the door

08
Competitive Landscape

Competition splits into five structural cohorts that travel across regions. None occupies the foundation position the combined entity is built around; the integrated Data Platform + Data Services + Agentic Activation stack is the differentiator at every layer.

Motivated competitors have the capital. They do not have the foundation or the window. The acquisition closes both.

The competitive landscape splits into structural cohorts that travel across regions. Operator ERPs (Cohort 1) and operator AI surfaces (Cohort 2) compete at the application layer. Institutional reporting platforms (Cohort 3) and analytics platforms (Cohort 4) compete at the data layer. Horizontal AI infrastructure (Cohort 5) competes through both vertical acquisition (data platforms) and hyperscaler-LLM partnerships arming new vertical AI plays. Cherre's foundation position is at the layer that does not yet have an incumbent. Institutional in-house builds (sovereign wealth funds, large LPs) are addressed in Section 06 as buyer archetypes, not as competitors.

Cohort Representative competitors Position today Why the combined entity is ahead, and for how long
Cohort 1: Operator ERP Yardi, MRI Software, Entrata, AppFolio, Inhabit IQ (incl. ResMan) Deep property management workflows across multi-asset platforms (Yardi, MRI) and multifamily-focused operators (Entrata, AppFolio, Inhabit IQ); large installed base; growing AI investments across the cohort (Yardi Virtuoso, MRI Agora). Operational AI is bounded by the data silo it sits on. The combined entity governs the foundation they would need to acquire.
Cohort 2: Operator AI surface Yardi Virtuoso, MRI Agora, AppFolio Realm-X, Entrata ELI, Inhabit IQ (incl. ResMan), Elise AI, Funnel Leasing AI products embedded in or alongside Cohort 1 systems. Resident communications, leasing, marketing automation. Lumina is the only AI in this cohort that runs on a governed foundation. Surface plus foundation is structurally different.
Cohort 3: Institutional reporting eFront (BlackRock), Juniper Square, Allvue / iLevel, Mercatus, Chronograph LP reporting, fund admin portals, portfolio monitoring. eFront dominant in EMEA; Juniper Square strong in NA; Mercatus and Chronograph cross-region. Era Two adjacencies; each governs one slice. Reporting sits above the foundation. Multi-source operating partner aggregation is a different architectural problem; the foundation is multi-slice and decade-compounding, not a single roadmap extension.
Cohort 4: Analytics and valuation CoStar Group, MSCI / Real Capital Analytics, Altus Group, 73 Strings Commercial real estate analytics, market data, and valuation. CoStar dominates CRE analytics and market data at scale (also owns LoopNet, Apartments.com, Homes.com, STR). MSCI / RCA owns institutional transaction data and indices. Altus owns valuation; 73 Strings is the AI-native challenger. All consume data; none govern it. Analytics consumes the foundation; does not replace it. The AI-native version of analytics does not change the layer.
Cohort 5: Horizontal AI infrastructure Snowflake, Databricks, Palantir, plus hyperscaler-LLM partnerships (AWS + Anthropic, Microsoft + OpenAI, Google Cloud + Gemini) Two flavors. Data platforms (Snowflake, Databricks, Palantir) compete via vertical acquisition; Palantir closest in posture (FDE practice, vertical depth). Hyperscaler-LLM partnerships are the infrastructure behind recent vertical AI plays: the Anthropic + Blackstone + Goldman $1.5B real estate deal runs on Bedrock; the OpenAI + Brookfield + TPG $10B deal runs on Azure. Acquisition path closes for data platforms; the obvious target leaves the market. Partnership path: model access is commoditizing while the operational data those partnerships need to reason against is not. The trust layer is the defense in both cases.
09
Industry Coordination & the Meta-Layer

Multiple credible standards exist for the real assets data landscape: RETTC/MITS, REDI, NCREIF, OSCRE, INREV, and others. None connects to the others without manual reconciliation, and manual reconciliation does not survive contact with AI agents. CORA, the Common Ontology for Real Assets, does not compete with the standards bodies; it operates at the layer above them. It performs two functions no individual body is structured to perform for itself or its peers: n-way correspondence across the full landscape and longitudinal drift tracking. CORA is the standard of standards. Cherre's canonical layer is its operational implementation. See coradata.org for the public initiative.

What the meta-layer codifies: the drift taxonomy

Standards do not sit still. They evolve under their own governance, in their own cadence, against their own audiences. The drift across them is not noise; it is the structural problem the meta-layer exists to solve. CORA names three drift types.

Drift type 01

Structural drift

The schema reorganizes. Same concept, new shape. Mappings that relied on the prior structure break silently. Without a meta-layer that pins the prior structure and annotates the change, downstream agents inherit a broken dependency they cannot detect.

Drift type 02

Semantic drift

A concept's definition narrows, broadens, or is redefined. The label stays the same; the meaning does not. Comparability across time degrades silently. The audit trail says "OperatingExpense" was reported in both periods; only the meta-layer records that the definition moved between them.

Drift type 03

Coverage drift

The standard expands or contracts. The scope of what is covered changes. Whole-domain gaps appear in cross-standard reconciliation when one body extends into territory another has retreated from. The meta-layer is what makes the gap visible before downstream systems silently consume it.

CORA operationalizes the drift taxonomy as a drift register: a neutral, public, version-pinned, machine-readable record of where and when drift occurs across the full standards landscape. Each entry captures what changed, why it matters downstream, and traces back to source documents. The register is the citable artifact. Cherre's canonical layer is the operational implementation. The platform applies the n-way correspondence and the drift annotations in production so that an agent reasoning on top of it does not silently break when an underlying standard moves.

The bodies the meta-layer spans

Below the meta-layer sit the standards bodies the drift register tracks. Cherre is seated in the four most relevant to the combined entity's product portfolio.

Standard, seat secured

RETTC

Cherre seated. The Real Estate Technology & Transformation Center owns the MITS (Multifamily Information Transaction Standards) data model and APIs, covering property marketing, leasing, screening, and transactions across multifamily rental housing. Operator-side, multifamily-only. Aligns to RP Core.

Standard, seat secured

REDI

Cherre seated. The Real Estate Data Initiative is investor-led and produces a single data model that operationalizes existing standards (INREV, NCREIF) rather than replacing them. Covers closed-end, open-end (APAC, EU, NA), and SMA institutional vehicles. Aligns to Data Platform and AIM Core.

Standard, in discussion

NCREIF

In active discussions. National Council of Real Estate Investment Fiduciaries; the Chart of Accounts initiative standardizes operating revenue and expense coding for institutional performance benchmarking and the NCREIF Property Index. LP-side benchmark. Aligns to Data Platform and AIM Core.

Standard, in discussion

OSCRE

In active discussions. Open Standards Consortium for Real Estate; the Industry Data Model spans 130+ use cases and 1,000+ entities across operator and institutional real estate, with downloadable JSON and XML schemas. The broadest cross-segment model of the four.

The standards landscape maps cleanly onto Cherre's product cores. RP Core (multifamily, operator-side) sits where RETTC/MITS owns the rental-housing data exchange. Data Platform and AIM Core (LP, institutional) sit where REDI and NCREIF anchor the institutional reporting and benchmarking work. OSCRE spans both as the broadest cross-segment model. Cherre's seats are secured on the segment-specific bodies that mirror its product cores (RETTC on the operator side, REDI on the institutional side) and in discussion on the institutional benchmark (NCREIF) and the cross-segment model (OSCRE). Both halves of the product portfolio are already represented in the rooms where standards are being formed.

REDI's stated approach is to operationalize INREV and NCREIF rather than replace them. This is the same logic CORA codifies one layer up. The standards bodies that matter are already converging on the "connect existing standards, do not invent a new one" model. The structural risk a sophisticated diligence team will look for is whether a vendor-neutral standard could emerge that commoditizes the semantic layer. The answer reverses the question. Cherre is not a participant in the standards being commoditized. Cherre operates the layer above them: the n-way correspondence, the drift register, the operational implementation of the meta-layer. The firm that implements the meta-layer is not displaced by the standards converging beneath it. The firm that implements the meta-layer is the firm the convergence runs through.

Cherre is not a participant in the standards. Cherre operates the trust layer above them.

10
How We Deliver

Three delivery functions, each mapped to a product. The Onboarding team deploys Data Platform. The Support team sustains Data Services. FDEs deliver Agentic Activation. The Medallion data architecture, Bronze, Silver, Gold, is the shared development methodology that underpins all three functions.

→ Action

The Onboarding Team is built from Cherre's existing Data Platform deployment muscle plus RealPage's implementation team; the Data Services Support Team is stood up new under unified CX leadership; FDE is hired against a defined pod profile.

Delivery function 01 · Data Platform

Onboarding Team

Responsible for technical implementation of Data Platform: data source discovery, connector setup, schema mapping, validation rule configuration, and go-live. Onboarding velocity, time from contract signature to first live data, is the primary metric. Playbooks are standardized globally; client-specific configuration is documented and handed to Data Services on go-live.

Delivery function 02 · Data Services

Support Team

Manages ongoing data quality, pipeline reliability, and issue resolution for Data Services clients. This is not a helpdesk. It is a managed service organization accountable to data quality SLAs. It handles operating partner data changes, schema drift, new source onboarding within existing contracts, and platform-level escalations to Data Platform engineering. The post-close Data Services practice runs as a managed delivery portfolio. RealPage's owned delivery centers in India and the Philippines carry predictable, repeatable workloads at managed cost. Cherre's existing Data Services partner network covers surge capacity, regional reach, and specialty workloads where owned playbooks have not yet been built or do not economically scale. Both modalities are strategic assets; the mix is actively managed.

Delivery function 03 · Agentic Activation

Forward Deployed Engineers

FDEs design, build, and optimize AI agents for clients. They begin with a readiness assessment (is the client's data ready?), move to use case design (what should the agent do and why?), then build and deployment. Post-deployment, FDEs optimize agent performance and identify the next use case. The product and the practice are the same thing: outcomes delivered by Forward Deployed Engineers running causal, comparative, predictive, and diagnostic reasoning against the Data Services foundation.

Why FDE-shaped delivery, not Center of Excellence

A Center of Excellence holds capability centrally and routes domain requests up for review before outputs return to the field. The model scales for reporting-era outputs, where review cadence is weekly and approval latency is measured in business days. Agentic outputs are continuous. Approval chains become latency. The reviewing center becomes the bottleneck the agent was supposed to eliminate. FDE-shaped delivery keeps domain expertise and AI capability in the same accountable unit. Pods carry the governing principles into the engagement. Authority sits at the unit doing the work. The decision-maker receives governed outputs without the relay.

Approval chains were a feature in the reporting era. They are a bottleneck in the reasoning era.

The model has institutional precedent. Palantir's Forward Deployed Engineer practice scales by pod, not by hierarchy. McKinsey's QuantumBlack operates as embedded teams with platform-side principals, not as a center of excellence routing through a partnership. The argument is not that FDE is new; it is that the existing CoE-shaped CX functions inside both Cherre and RealPage need restructuring before they meet agentic delivery at scale.

Capability model, who owns what

Capability Onboarding Team Data Services Support Team FDE
Data source discovery & schema mapping Owns New sources in-contract Complex / novel schemas
Connector setup & pipeline configuration Owns Maintains post-go-live Custom / bespoke integrations
Ongoing data quality & SLA management Hands off at go-live Owns
AI readiness assessment Owns
Agent design, build & deployment Owns
Agent optimization & new use case expansion Owns
Client technical training & enablement Data Platform-level Data Services-level FDE-level
Platform bug escalation to product engineering Escalates Escalates Escalates + provides field context

Build, buy, or consume: the architectural layers

The in-house vs. partner question is sharper at the layer level than at the function level. Some layers each enterprise has to run for itself. Some are candidates for shared industry-level infrastructure, because the value of each scales with the number of participants. One layer sits locally but depends on the layers below being stable. This is the framework the combined entity uses to answer the audience's question about which services stay in-house and what in-house actually looks like.

Enterprise, build or buy

Operated inside the firm

Firm-specific operating models, security perimeters, and consumer surfaces. Each client owns them. RealPage owns at the operator side; Cherre owns at the institutional side. The post-close question is consolidation and standards, not provenance.

Industry-level, consume from canonical layer

Shared infrastructure

Value scales with the number of participants. A canonical identity layer with one institution's data is worth less than one with many. This is the layer the combined entity operates for the industry. CORA codifies the coordination; the platform implements it.

Local but dependent

The reasoning surface

Each client deploys agents against their own use cases, but those agents query a stable model-context surface that abstracts them from underlying schema and vendor changes. FDE-shaped delivery is what makes this layer reliable across clients without bespoke point-to-point integration.

The nine layers, mapped to the three buckets

The architecture defined in Section 03 collapses here into a deployment decision. Each row is one layer of the stack; the bucket column names where the combined entity operates it: built or bought inside the firm, consumed from the shared canonical layer, or sitting locally but depending on the canonical layer beneath.

# Layer Class Bucket
09 Action surface (consumption) Persistent · consumer expands to human + agent Enterprise · build or buy
08 Reasoning surface Codified · Agentic Activation Local but dependent
07 Decision graph Codified · Agentic Activation Industry · consume
06 Context graph Codified · Agentic Activation Industry · consume
05 Semantic layer Codified · Data Services Industry · consume
04 Knowledge graph Codified · Data Services Industry · consume
03 Entity resolution Codified · Data Services Industry · consume
02 Ingestion + submission Transformed · Data Platform Enterprise · build or buy
01 Systems of record Persistent Enterprise · build or buy

The audience flagged that existing deployment and CX functions are weak in both legacy organizations. The layer-level frame is the answer: the canonical layer (industry-level) is where the combined entity invests deepest, because that is where scale advantage compounds. Enterprise-side delivery (in-house implementation, Data Services support, FDE pods) is rebuilt as the Platform Value Office, led by the Platform Value Office GM. RealPage's owned Data Services delivery centers in India and the Philippines and Cherre's Data Services partner network are integrated as a managed delivery portfolio for the support function: owned capacity handles predictable, repeatable workloads; partners handle surge, regional reach, and specialty workloads. The portfolio is the strategic asset, actively managed. Partner participation is appropriate at the enterprise edges (regional implementation surge, jurisdiction-specific compliance, specialty Data Services coverage) and explicitly inappropriate at the canonical layer.

Medallion architecture, the shared development methodology

The Medallion architecture defines how data is ingested, validated, and made ready for AI, in three enforced layers. Every client deployment follows this pattern, which ensures data quality gates are met before AI agents operate on the data. It is the shared methodology across Onboarding, Data Services, and FDE.

Artifact, Medallion data flow with function ownership

From operating partner data to agent consumption, three layers with gates between

DATA FLOWS LEFT TO RIGHT · QUALITY ENFORCED AT EACH GATE SOURCE Operating partner data BRONZE Ingest Raw, lineage tracked Schema discovered Data Platform ONBOARDING SILVER Validate Normalized, resolved Cross-source reconciled Data Services SUPPORT GOLD Activate Agent-ready, API-consumable Agentic Activation FDE CONSUMER AI agents, workflows GATE completeness GATE accuracy GATE readiness PHASE 01 STAND-UP · PHASE 02 SCALE · PHASE 03 GLOBAL Phase 01: pipeline live Phase 01: rules Phase 02: SLA Phase 02: agents Phase 03: ICP

No layer is skipped, no gate is bypassed. Bronze is what Data Platform delivers at go-live. Silver is what Data Services sustains under SLA. Gold is what FDEs build agents against. Each function's quality gate is the next function's entry condition.

Bronze Layer, Raw Ingestion
Ingest
  • Operating partner data ingested as-is; no transformation at point of entry
  • All source formats supported: structured, semi-structured, document
  • Full audit trail maintained from source to destination from day one
  • Schema discovery and initial data profiling run automatically
  • Data lineage tracked continuously; source attribution preserved
Quality gate
Completeness, is expected data arriving at expected frequency and volume?
Silver Layer, Validated & Standardised
Validate
  • Schema normalization and field mapping to client data model
  • Deduplication and entity resolution across operating partners
  • Validation rules applied (defined per client by Data Services team)
  • Anomaly flagging and exception reporting generated automatically
  • Cross-source reconciliation and conflict resolution
Quality gate
Accuracy and consistency, does data meet the client's agreed validation ruleset?
Gold Layer, Business-Ready & Agent-Ready
Activate
  • Enriched, analytics-ready data sets available for downstream consumption
  • API-consumable for client systems, dashboards, and reporting tools
  • Structured for agent consumption: embeddings, retrieval, context packaging
  • Portfolio analytics and LP reporting outputs generated at this layer
  • FDE-built agent workflows operate exclusively at the Gold layer
Quality gate
Business completeness, does Gold layer data support the client's agreed use cases and agent workflows?

Beyond Gold: the decision graph

The Medallion layers serve the data agents reason against. The decision graph is what makes their reasoning auditable. Every meaningful action an agent takes, every approval, every override, every interpretation, every model output, is captured as a queryable event indexed by entity, by actor, by time, and by definition version. This is what reproducibility looks like in implementation. Without it, an agent that produces the right answer cannot prove how it produced the right answer, and the answer fails the audit standard the next era will impose.

The Gold layer makes the agent capable. The decision graph makes the agent accountable.

Delivery model decisions to resolve

11
Three Moves at Close

Three organizational moves at close: the Platform Value Office, a unified Client Experience function spanning Onboarding, Data Services Operations, and FDE, and an FDE Practice. The Office is led by the Platform Value Office GM with two engine leads beneath: Head of AI + FDE owns the FDE Practice; Head of CX + MDS owns Onboarding, Data Services Operations, and the unified CX horizontal. The Platform Value Office is the operating engine of value creation back to the software business. Its mandate runs three directions: back to software, across asset classes, across geographies. The incumbent BUs (Cherre product, RealPage product) are US-centric by construction; the Platform Value Office is built global from day one, the only way to ship Data Services in EMEA, FDE-led activations in Singapore, and operator implementations across regions on a single operating cadence. This is the structural reason services cannot be absorbed into an existing product BU.

→ Action

Stand up the Platform Value Office first; do not absorb existing teams unchanged. The Platform Value Office GM and the two engine leads (Head of AI + FDE; Head of CX + MDS) are the Phase 01 gate that opens the first regional pod in Phase 02. See Section 14.

Move 01

Platform Value Office

Stand up the Platform Value Office as a global function, led by the Platform Value Office GM with two engine leads beneath. Head of AI + FDE owns two engines: the AI engine builds the Agentic Activation platform (product development); the Domain engine is the FDE Practice deploying it with clients (field engineering). Head of CX + MDS owns Onboarding, Data Services Operations, and the unified CX horizontal. Delivery functions report through the Office rather than into the product organization. Data Services and Agentic Activation revenue reports distinctly from software subscription revenue: software and managed services land in opex on the buy side; FDE engagements sit with consulting in capex. Buyers can budget across both lines without ambiguity. Sales, Marketing, and Client Success interlock as cross-functional peers; the GM holds GTM authority over platform revenue lines.

Move 02

Unified Client Experience

The CX function is built fresh, not inherited. Onboarding, Data Services Support, and FDE share a single intake, a single client success owner per account, and a single escalation path back to product engineering. This is the function the audience flagged as weakest in the existing organizations; the brief recommends rebuilding it rather than merging the existing teams.

Move 03

FDE Practice

The FDE Practice sits inside the Platform Value Office and is structured as a federation of small embedded pods, each carrying domain expertise and AI capability in the same unit. Authority sits at the pod. Coordination across pods is lateral. The Practice ramps regionally on the same phase-gated sequence as the rest of the Office, with the NA cohort hired first and EMEA following at the Phase 02 gate.

Artifact, Platform Value Office structure

Platform Value Office (proposed)

GM Global Platform GTM HEAD OF AI + FDE HEAD OF CX + MDS AI ENGINE Product Development DOMAIN ENGINE Field Engineering CX ENGINE Client Experience MDS ENGINE Managed Data Services NORTH AMERICA · EMEA · APAC

The GM at top with two engine leads beneath. Each engine lead owns two engines. The Office is built global from day one; regional pods sequence through phases without changing the structure above.

12
Inside the Office

How the Platform Value Office actually runs: the three delivery functions inside, the capability map across them, the phased hiring sequence, and the organizational decisions to resolve before close.

Three delivery functions inside the Office

Three functions sit inside the Platform Value Office, each contributing services revenue. A fourth function, SaaS Support, sits adjacent: operationally co-located so signals flow into Onboarding, Data Services, and FDE, but funded as software COGS to keep software and services revenue distinct on the buyer's side.

Function 01 NRR
Onboarding exists
Project-based. Get client data to Gold-layer state. Scoped engagement with defined deliverable. Ends at go-live.
Metric: time-to-Gold
Function 02 ARR
Data Services Operations exists
Ongoing. Maintain Gold-layer state against SLAs. Data quality, pipeline reliability, operating partner change management.
Metric: SLA attainment, anomaly resolution
Function 03 NRR
FDE Practice build new
Project-based. Scoped agent design, build, and deployment. Defined deliverable, clear exit. Post-deployment monitoring hands to Data Services.
Metric: agents deployed, expansion per account

The client lifecycle flow

Onboarding delivers Support keeps it running Data Services defends the data FDE builds on top

How signals travel

Inside the same operating team, even when Support's costs sit in software COGS, signals travel by standup, by shared ticket queue, by the same manager asking "what are we seeing this week?" The cost-center boundary doesn't break the discipline; the BU is built so it doesn't.

Support sees a pattern
→ signals Onboarding
Three clients hit the same schema mapping confusion in week one. Onboarding updates the playbook before the next go-live.
Support sees a pattern
→ signals Data Services Operations
Tickets spike around month-end from the same operating partners. Data Services proactively monitors those partners before the next cycle.
Support sees a pattern
→ signals FDE Practice
Multiple clients ask "can I automate this report?" Support flags it as an agent use case. FDE scopes a repeatable engagement.

Client experience as a horizontal

CX is not a fifth org. It is a horizontal function inside the Platform Value Office that ensures the client sees one team, one intake, and one escalation path, regardless of which function is doing the work.

CX owns
Single client owner
One named person per account across all functions.
CX owns
Unified intake
Routes to the right function. The client says what's wrong; CX decides where it goes.
CX owns
Escalation path
One channel to product engineering from all functions.
CX owns
Account health
Full view: tickets + projects + SLAs + agent performance. One dashboard.

Capability map, what gets merged, what gets built, what gets sunset

Capability RealPage today Cherre today Post-close action
Implementation / Onboarding (operator data) Owns at scale Data Platform-specific muscle Merge under the Platform Value Office; RealPage muscle leads, Cherre Data Platform expertise embeds
Implementation / Onboarding (institutional data) Owns Cherre team carries forward; reports into the Platform Value Office
Data Services Established native practice via owned delivery centers (India and the Philippines) Partner-delivered, not native Owned and partner delivery integrated as a managed portfolio: owned capacity for predictable workloads, partners for surge and specialty; new hires staffed against EMEA / APAC ramp
FDE for Agents Early-stage practice Build new under FDE Practice; structured hiring against named ramp profile
Client Success (operator accounts) Owns at scale Restructure under unified CX function; not absorbed unchanged
Client Success (institutional accounts) Owns Cherre team carries forward; institutional-only CS specialization preserved
Sales engineering / Solutions consulting Owns Solutions function Open question, merge under sales, or fold into FDE pre-sales motion
Partner / channel enablement Partial coverage Fund admin channel only Open question, channel strategy precedes org design here

How revenue is reported

Software stream
Data Platform subscriptions ARR
Software subscription revenue with SaaS support funded as COGS. Lands in buyer opex with other SaaS spend.
Services stream
Data Services + Agentic Activation ARR + NRR
Data Services contract revenue (ARR) plus FDE engagement revenue (NRR). Managed services land in buyer opex; FDE engagements sit with consulting in capex.
Buyer-side budgeting
Two budget lines, one vendor
Software and managed services land in opex; FDE engagements with consulting in capex. The combined entity reports each stream distinctly so the buyer can budget cleanly across both.

The test: can the buyer's finance team allocate Data Services and Agentic Activation revenue cleanly across their opex and capex budgets? If yes, the revenue distinction is working.

Hiring sequence, what gets staffed first

Phase 01, 0 to 6 months
Stand-up
  • Platform Value Office GM confirmed (Cherre's existing senior leadership)
  • Head of CX + MDS confirmed (Cherre's existing CX + MDS leadership; owns Onboarding, Data Services Operations, and the unified CX horizontal)
  • Head of AI + FDE confirmed (Cherre's existing AI + FDE leadership; owns FDE Practice and agentic professional services)
  • First two regional pod members hired into NA; remaining ramp follows phase gate
  • India and Philippines Data Services delivery centers (RealPage-owned) and Cherre's Data Services partner relationships integrated as a managed delivery portfolio; capacity plan finalized against Phase 01 deal pipeline
Gate to Phase 02
Platform Value Office operational with revenue reported distinctly from software; first FDE-led agent engagement live; unified CX intake stood up
Phase 02, 6 to 18 months
Scale
  • EMEA Platform Value Office regional lead in market (UK based); first EMEA FDE pod ramping
  • FDE Practice scales to defined NA cohort target; productized AI readiness assessment in market
  • Data Services Support regional team established in EMEA
  • Unified CX function fully merged across legacy RealPage and Cherre teams
  • FDE comp model and attribution rules finalized; first cross-region FDE deployment executed
Gate to Phase 03
Platform Value Office EMEA operations running at target capacity; first APAC institutional account in contract; partner-channel strategy resolved
Phase 03, 18 months+
Global
  • APAC presence stood up (Singapore-led, Australia secondary)
  • FDE Practice scales to durable global headcount with cross-region rotation
  • Partner ecosystem operational at target ratio of in-house to channel delivery
  • Unified CX function operating at steady state across three regions
  • Platform Value Office revenue (Data Services + Agentic Activation) at defined contribution to combined entity, reported distinctly from software
Steady-state measure
Platform Value Office revenue per FDE, NRR contribution from FDE-covered accounts, Data Services gross margin at target, CX NPS

Organizational decisions to resolve before close

13
Capability Model

Three functions. Distinct scope. No overlap on accountability. The capability model defines what Onboarding, Data Services Support, and FDE each own, where they support each other, and where their mandate ends. It is the operational answer to what the combined entity actually delivers. It is the frame against which every hire, every ramp, and every client engagement is evaluated.

Function 01 · Data Platform

Onboarding

Purpose: Get the client's data to a certified Gold-layer state, on schedule, with a documented handoff. Onboarding owns everything from first kickoff to go-live sign-off. Its success metric is time-to-Gold.

Scope limit: Onboarding ends at go-live. Post-go-live data quality and pipeline health transfer fully to Data Services Support. Onboarding does not maintain; it delivers and moves.

Function 02 · Data Services

Data Services Support

Purpose: Maintain the Gold-layer state Onboarding delivered, continuously, against agreed SLAs. Data Services Support owns data quality post-go-live, pipeline reliability, and operating partner change management. Its success metric is SLA attainment and anomaly resolution time.

Scope limit: Data Services Support does not onboard new clients; it manages within active contracts. Net-new client onboarding returns to the Onboarding function.

Function 03 · Agentic Activation

Forward Deployed Engineering

Purpose: Design, build, and optimize AI agents that operate on Gold-layer data. FDE owns the full agent lifecycle, from readiness assessment through deployment and ongoing optimization. Its success metric is deployed agent count, agent NRR, and new use case expansion per account.

Scope limit: FDE does not do data engineering or pipeline management. It operates on a stable data layer. If the layer is not stable, FDE surfaces that to Data Services and Onboarding before proceeding.

Onboarding delivers a state. Data Services defends that state. FDE builds on top of it.

Capability model

Functional areas define the phases of delivery. Capabilities (pills) will be mapped to each area and color-coded by which function owns them: Onboarding, Data Services Support, or FDE. Project Management spans all four delivery phases as a cross-cutting discipline.

Maturity levels: baseline to excellent

Baseline is the floor: what each function must be able to do before touching a client account. Excellent is the Phase 02 target state. The gap between them is each engine lead's development agenda.

Function Baseline (floor at hire) Excellent (Phase 02 target)
Onboarding Executes documented onboarding playbook; achieves Bronze and Silver layer certification within committed timeline; completes clean Data Services handoff with documented validation ruleset Runs parallel onboarding tracks without quality degradation; proactively surfaces schema complexity before kickoff; contributes new patterns to playbook; zero regression on Data Services handoffs; sub-30-day time-to-Silver for standard ICP clients
Data Services Support Monitors pipeline health against SLA thresholds; resolves data quality exceptions within agreed window; manages within-contract source additions without re-engaging Onboarding for playbook guidance Predictive anomaly detection that surfaces issues before clients flag them; owns schema drift taxonomy and responds to regulatory changes without escalation; SLA attainment above 99%; maintains Gold-layer readiness for FDE without manual coordination; monthly data quality report published to client stakeholders
FDE Delivers productized AI readiness assessment; scopes and deploys first agent against a defined use case within agreed timeline; achieves client acceptance on first submission Runs multi-use-case agent programs across concurrent accounts; proactively identifies expansion use cases before renewal; contributes reusable patterns to FDE practice library; generates measurable NRR lift on every covered account; deployable cross-region without a re-ramp period

Ramp model: 30 / 60 / 90 day milestones

Every new hire enters a structured ramp regardless of seniority. Milestones are binary: either the deliverable exists or it does not. They are the basis for ramp-period performance review and the gate before a hire receives primary account ownership.

Onboarding

Day 30
Tools
  • Data Platform platform certified: can configure connectors, run schema mapping, execute Bronze ingestion independently
  • Onboarding playbook read; can reproduce the go-live checklist without reference
  • Has shadowed at least one full engagement from kickoff to Data Services handoff
  • Can explain the Silver validation ruleset framework to a client without support
Day 60
First Lead
  • Co-lead on at least one live onboarding engagement
  • Has identified and documented at least one schema edge case not covered by existing playbook
  • Completed first Data Services handoff with no playbook gap flagged by the receiving team
  • Handling day-to-day client communication on onboarding status independently
Day 90
Full Ownership
  • Primary lead on one or more engagements end-to-end
  • Time-to-Silver tracking within 10% of team average
  • At least one new pattern or edge case contributed to shared playbook
  • Data Services handoff accepted on first submission, no rework required
Ramp complete
Primary account ownership granted; time-to-Silver within benchmark; zero playbook regressions on first three solo handoffs

Data Services Support

Day 30
Tools
  • Monitoring tooling certified: can independently read pipeline health dashboards, exception queues, and SLA tracking
  • Has shadowed at least three exception resolution cycles with a senior team member
  • Understands SLA definitions for all active accounts in assigned portfolio
  • Knows the escalation path to Onboarding (new sources) and Product Engineering (platform bugs)
Day 60
First Portfolio
  • Primary owner of a defined portfolio of Data Services accounts under supervision
  • Resolving anomalies and exceptions independently within agreed SLA windows
  • Has handled at least one within-contract new source addition end-to-end
  • Weekly SLA report produced accurately and on schedule for assigned accounts
Day 90
Full Ownership
  • Unsupervised portfolio ownership; SLA attainment at or above team average
  • Has triaged at least one schema drift event without escalation to Onboarding
  • Proactively flagging data quality risks before clients report them
  • FDE team receiving Gold-layer data without manual coordination
Ramp complete
Unsupervised portfolio ownership; SLA attainment at team average; at least one proactive client-facing data quality action documented

Forward Deployed Engineering

Day 30
Tools
  • ATLAS platform certified: can configure, deploy, and test agent workflows in sandbox
  • AI readiness assessment framework internalised; can deliver it independently on a practice account
  • Has reviewed at least two completed FDE engagement case studies end-to-end
  • Domain context established: understands the real assets operating model for their assigned ICP segment (GP, LP, Operator, or Institutional Asset Manager)
Day 60
First Deployment
  • Co-lead on at least one live FDE engagement; has run the readiness assessment with a real client
  • First agent use case scoped, designed, and approved by client stakeholder
  • Agent deployed to Gold-layer data in client environment; acceptance criteria met
  • Has coordinated at least once with Data Services Support on a Gold-layer data question and resolved without escalation
Day 90
Full Ownership
  • Primary FDE lead on at least one account; managing the readiness → build → deploy → optimize cycle independently
  • First expansion use case identified and in scoping with client; NRR motion active
  • At least one reusable agent component or use case pattern contributed to the FDE practice library
  • Can represent the FDE function credibly in a sales cycle, with readiness assessment as the opening play
Ramp complete
Primary lead on one live engagement; first agent in production; expansion use case in pipeline; practice library contribution made

A ramp milestone is not a check-in. It is a binary gate. Either the deliverable exists or it does not.

Capability model decisions to resolve

14
Operational Execution

A compact view for the integration committee. The body of the brief argues the strategic commitments and operating model. The artifacts below name the integration sequence, the decisions being asked to ratify, and the integration risks under active management. The sequence, Decision Ledger, and Risk Register are populated; each fills further as additional technical content lands.

Integration sequence

Phase model from Section 02 crossed against the delivery, capability, and standards tracks. Each cell names the milestones and gates the track has to clear before the next phase opens.

Artifact, the integration sequence

From Org stand-up to global scale, the critical path across tracks

PHASE 01 · STAND-UP 0 to 6 months PHASE 02 · SCALE 6 to 18 months PHASE 03 · GLOBAL 18 months+ Data Platform Playbook standardized Anchor account in flight EMEA cadence Product ladder live Data eng at scale Multi-region velocity Data Services Portfolio integrated Support live, ARR target met SLA discipline mature Partner roles managed Multi-region coverage SLA above target FDE Forward Deployed Engineers Q1 cohort + first agent AI readiness productized EMEA pod in market specialist track defined APAC presence durable competitive moat STANDARDS CORA canonical layer CORA framework Drift register operational Register populated n-way correspondence live Standard of standards Industry coordination layer ORG Platform Value Office PVO GM + engine leads Cherre leadership confirmed Pods staffed NA + EMEA Escalation paths live PVO at scale Cross-regional cadence Critical-path node Critical-path edge Cross-track dependency Within-track

The critical path runs Org Phase 01 (Platform Value Office GM and engine leads named) into FDE Phase 01 (Q1 cohort and first agent), then through FDE Phase 02 (EMEA pod) and FDE Phase 03 (APAC presence). CORA participation in Phase 01 is the gate the institutional sales motion in Phase 02 references; misses here compound across phases. Data Platform and Data Services run in parallel; the foundation they deliver underpins the FDE work but does not gate it.

Decision Ledger

Decisions for the integration committee to ratify. Priority indicates urgency, not sequence; each is argued in the body of the brief.

# Decision Recommendation Owner Priority
01 Platform office scope GM holds GTM authority over platform revenue lines (Data Services + Agentic Activation); product BUs keep SaaS GTM TBD High
02 PVO GM leadership Confirm Cherre's existing senior leadership in the Platform Value Office GM role TBD High
03 CX leadership Confirm Cherre's existing leadership in the Head of CX + MDS role TBD High
04 FDE leadership Confirm Cherre's existing leadership in the Head of AI + FDE role TBD High
05 Non-software revenue model Design the combined entity's accounting and compensation framework to handle software (SaaS), managed services (Data Services), and engagement-based revenue (Agentic Activation) distinctly. Sales attribution and delivery team compensation both flow from this framework. Existing Cherre and RealPage structures are SaaS-oriented; non-software revenue needs explicit rules. TBD Medium
06 Partner network Owned capacity for predictable workloads; partners for surge, regional reach, specialty coverage; criteria locked before EMEA pod ramp TBD Medium

Risk Register

Integration risks for the integration committee to track. Impact indicates severity, not sequence; each carries a mitigation argued in the body of the brief.

# Risk Mitigation Owner Impact
01 Talent retention Retention packages in place; named champions identified at both legacy orgs TBD High
02 Product integration latency Integration scope is the semantic infrastructure (Sections 03, 09) and Cherre's Luna and Meridian products; Phase 01 gate requires first FDE-led agent engagement live before Phase 02 opens; layered architecture allows independent build streams TBD High
03 Client churn Account-by-account communications plan; named single owner per client through transition; SLA continuity guarantees TBD High
04 Resource constraint Phased hiring against named profiles; partner network as overflow capacity for surge work; existing offshore delivery centers (India, Philippines) extend in-house capacity; pod composition optimized to share scarce expertise TBD Medium
05 Adoption / cultural friction Bridge domain gaps (Cherre institutional depth, RealPage operator depth) through joint pod composition; manage operating model friction between legacy software company process and startup velocity through shared cadence and named field-level signal routing TBD Medium
06 Incumbent capabilities / gaps Capability map identifies gaps by function (Section 13); Phase 01 hiring sequence (Section 12) addresses gaps; FDE Practice fills frontier capability needs TBD Medium
Delivery Design
Capability Model
Filter by team
Discovery
Facilitate Discovery Sessions Develop Use Cases Gather Requirements Document Business Requirements
Design
Create Technical Requirements Design Workflows Create Orchestration Framework Determine Test Harness Design Business Rules Create Canonical Definitions and Decision Trees Document Design Frameworks and Decisions
Development
Build Agentic Workflows Create MCPs Codify Business Rules Formalize Judgement Build Canonical Definitions and Decision Trees Build Agentic Scaffolding Data Engineering Pipeline Monitoring Data Modeling
Deployment
Configure Systems Train Users Manage Testing Manage Rollout
Operate
Monitor Data Quality Troubleshoot Issues Maintain Systems Escalate Issues
Project Management
Manage Project Plans Manage Budgets Create Forecasts Manage Resources Track Project Hours Develop Status Reporting Conduct Status and Steerco Meetings Manage Stakeholders Track Submission Status