How {yourcompany}os Brings the Palantir Model to SMEs

Palantir built one of the most successful enterprise software businesses of the last decade on a simple structure: model the customer's world in an ontology, embed engineers who encode the customer's knowledge into the platform, and position the result as the operating system the whole organization runs through. The model works. It also excludes small and mid-sized companies by design. This guide explains how the Palantir playbook actually functions, why its structure prices out SMEs, and how {yourcompany}os applies the same model with each pillar inverted.

The core inversion in one sentence: Palantir keeps the ontology closed because the ontology is Palantir's moat. For an SME, the moat is its own processes, so the schema should be open and shared, and the processes should be what each company keeps and refines.

What makes Palantir work

1. The ontology

At the center of Palantir Foundry sits the ontology: a semantic layer that maps a company's real-world objects (customers, orders, assets, incidents) and the actions on them into one shared model that software and AI can operate on. It is what turns scattered data into something decisions can run through. It is also proprietary and closed. Once your operations are encoded in Palantir's ontology, leaving means re-modeling your entire business somewhere else. The lock-in is not a side effect; it is the business model.

2. Forward deployed engineers

Palantir does not sell shrink-wrapped software. It embeds engineers with the customer, on-site, who learn the domain, encode it into the platform, and feed what they learn back into the product. This is why implementations succeed where generic software fails, and it is why the price starts in seven figures: you are paying for elite engineering time, flown in and taught your domain from zero.

3. Operating-system positioning and land-and-expand

Palantir lands with one painful workflow, proves value fast, and expands until Foundry is the layer everything runs through. "Operating system for the modern enterprise" is not just marketing; it describes the end state the delivery model is built to reach.

Why SMEs cannot buy this

Nothing about the model requires a Fortune 500 customer, but everything about Palantir's implementation of it does:

• Price. Contracts routinely start in the millions. A 30-person planning firm or a regional manufacturer is not in the addressable market and never will be.
• Lock-in risk. A large enterprise can negotiate exit clauses and run parallel systems. An SME that encodes its operations into a closed ontology has bet the company on one vendor's pricing and roadmap.
• No local presence. Forward deployed engineers are generalists who must be taught the domain, the language, and the regulatory context on the customer's budget. That overhead is absorbable at enterprise scale and prohibitive below it.

Palantir proved the model works. Its structure excludes SMEs by design, not by accident.

The same model, inverted

Open ontology instead of closed

The schema behind {yourcompany}os is published openly on GitHub as the Pragmatic Data Contract. Anyone can inspect it, build against it, or export their data out of it. For an SME this flips the risk calculation: the semantic layer is a commons, not a cage. And because multiple companies share one open schema, patterns built on it compound across the ecosystem instead of being rebuilt inside every closed deployment.

Processes as the secret sauce, patterns as the commons

Here is the subtle point. Every firm approves invoices, onboards projects, and tracks contacts. Those are shared patterns, and on {yourcompany}os they exist as ready building blocks: CRM light, task management, document handling, project creation. But who approves, at what threshold, with which exceptions, in what order - that is the operating knowledge that makes one firm better than its competitor. In Palantir's model, that knowledge gets encoded into Palantir's platform. In {yourcompany}os, it lives in your BPMN process definitions: readable, auditable, exportable, and yours. The commodity layer is shared; the differentiating layer stays with the company it belongs to.

Forward deployed partners instead of forward deployed engineers

Palantir flies in brilliant generalists and teaches them your domain. {yourcompany}os works through local partners who already have the domain: they speak the industry's language and, in the Swiss market, literally the language, the KMU culture, and the regulatory context (Swiss hosting, GDPR, audit requirements). The mechanism is the same, embedded expertise that encodes how the company actually works. The knowledge flow is inverted: instead of importing engineers who learn the domain, the domain expert gets a platform.

Explainable processes instead of black-box AI

One more inversion matters. Palantir's pitch increasingly centers on AI operating over the ontology. {yourcompany}os puts BPMN in control of agents, not the other way around: AI steps do bounded work like research, classification, and drafting, while gateways and people decide what happens next, and every step is logged. For an SME without a data-science team to supervise a black box, explainability is not a preference; it is the only governable option.

Palantir vs {yourcompany}os at a glance

	Palantir	{yourcompany}os
Ontology / schema	Proprietary and closed; the vendor's moat	Open, published on GitHub; a shared commons
Where the secret sauce lives	Customer knowledge encoded into the vendor's platform	In the customer's own process definitions, on shared patterns
Delivery model	Forward deployed engineers, flown in and taught the domain	Local partners who already know the industry and regulation
Entry price	Seven-figure contracts	Discovery Sprint at CHF 6'000, first process live in 1-2 days
AI approach	AI operating over the ontology	BPMN in control of agents; explainable, human approves
Exit	Re-model the business elsewhere	Export data and processes on the open schema

What this looks like in practice

The land-and-expand motion translates directly to SME scale. A Discovery Sprint (CHF 6'000) takes one core process from a structured workshop through mapping to a live implementation in 1-2 days: the same "prove value with one workflow first" logic Palantir uses, at a price a 20-person firm can decide on in one meeting. Ongoing operations start from CHF 3'000 per year for SLA, Swiss hosting, and support, and scaling is priced by workload, not by seat count. Your CRM, email, and specialist tools stay where they are; {yourcompany}os orchestrates between them.

The end state is the same one Palantir sells: an operating system the company runs through. The difference is who owns what when you get there. With {yourcompany}os, the schema is open, the patterns are shared, and the processes, the part that actually makes your firm competitive, remain yours.