Concept card

Escrow & Source Code Protection

As of 2026-07-24T14:00:00Z

Edited by Cédric-Olivier Mâry, sole editor and publication director — 27 years in the SAP ecosystem. How these pages are produced, and where automation is used.

What is Escrow & Source Code Protection?

Escrow protects clients against bus-factor risk on custom IP — but it only earns its €1.5-5k/year cost above roughly €200k engagement value or genuine long-term dependency.

An escrow agreement solves a specific and uncomfortable problem for a client buying custom SAP analytics work: what happens to their Datasphere views, Business Data Cloud pipelines or Joule agent configurations if the consultant who built them becomes unavailable — through bankruptcy, incapacity, or simply walking away from the market? Under a tripartite escrow arrangement, the consultant deposits the source code, configuration scripts and documentation with a neutral third party, the escrow agent, who releases them to the client only if a defined trigger event occurs. It is, in effect, insurance against bus-factor risk on intellectual property the client cannot easily reproduce themselves.

Five trigger events cover the overwhelming majority of escrow agreements in this market: consultant bankruptcy or insolvency; death or permanent incapacity; material and uncured breach of the support obligations in the underlying services agreement; cessation of business; or a mutually agreed termination that leaves no qualified replacement in place. The drafting discipline that separates a workable escrow clause from a disputed one is whether each trigger is objectively verifiable — a court ruling, a death certificate, a documented and time-stamped breach — rather than requiring subjective judgment or arbitration to establish. Ambiguous triggers are the single biggest source of escrow disputes, because they surface exactly when tensions are already highest.

When escrow earns its cost versus when it is unnecessary overhead

Why it matters

Five release triggers (bankruptcy, death, material breach, cessation, unreplaced termination) must be objectively verifiable — vague triggers create disputes exactly when escrow matters most.
Below the value/dependency threshold, escrow overhead exceeds its protection value — it's not a default clause to include everywhere.
Solo consultants typically negotiate down to annual-refresh-plus-final-deposit cadence rather than full per-release deposits, which inflate cost without proportional benefit.

Key points

Tripartite contract (depositor · agent · beneficiary) for source/docs/config.
Five verifiable triggers: bankruptcy · death · material breach · cessation · joint instruction.
Threshold: engagement > €200k + custom IP.
EMEA agents: NCC (default) · Iron Mountain · EscrowTech · notarial (FR/DE).
Setup €500-2k + annual €1.5-5k.
Cadence: initial + annual refresh + final-state deposit.
Code without docs is unusable on release; deposit must include README + deployment guide.
Align scope with C062 liability cap.
Escrow & Source Code Protection is mastered only when it changes a named buyer decision.
Start with the semantic contract and control model before demonstrating the tool.

Terms used on this page

Escrow agreement: Tripartite contract depositing critical assets with a neutral agent for release on specified triggers.
Depositor: Party depositing the assets (consultant).
Beneficiary: Party receiving the assets on release (client).
Escrow agent: Neutral third party (NCC, Iron Mountain, etc.) holding deposit + administering release.
Verifiable trigger: Release condition with objective evidence (court ruling, death certificate, cessation filing). Preferred over vague triggers.
Annual refresh: Yearly deposit update verifying current state. Default cadence for solo consultants.
Final-state deposit: Last deposit at engagement end with permanent retention. Captures complete operational state.
BAIT: BaFin Anforderungen an die IT — DE banking-IT regulation. Influences escrow-agent selection for DE financial-services clients.

Sources

Full card available to members. What the full card adds: the full decision framework · the SAP vs Snowflake / Databricks / Fabric comparison · the common pitfalls and their fix · the cheat sheet · the architecture schemas · the code blocks · the facts worth quoting.

Open in the app →