What to preserve before shutting down software

The decision to shut down software starts a time‑sensitive process

When a product line ends, an acquisition closes, or a legacy platform is retired, the data it holds does not disappear automatically. The data remains an asset that can be licensed, sold, or archived safely. The window for action is narrow. Systems may be turned off. Credentials can be lost. Staff with contextual knowledge may leave. Acting early preserves value and reduces risk.

Identify the data landscape before any system changes

A clear view of what exists is the foundation of every later step. Begin with a high‑level map that lists:

• Production databases and their backup schedules.
• File storage locations, including network shares and cloud buckets.
• Legacy applications that still hold transaction logs or configuration files.
• Source‑code repositories, schema definitions, and data dictionaries.
• Third‑party services that retain copies of data (analytics, monitoring, SaaS).

Use the Data Inventory service page (/collections/data-inventory/) as a reference for the level of detail we recommend. Record the owner of each system, the date of the last backup, and the expected decommission date.

Capture rights and provenance early

Ownership and licensing rights are often scattered across contracts, internal policies, and vendor agreements. Without a consolidated view, you cannot determine what may be transferred. Create a Rights and Provenance Register (/collections/rights-and-provenance/) that captures:

• The legal entity that owns the data.
• Any third‑party licenses that restrict redistribution.
• Dates when data was collected and any consent terms attached.
• The chain of custody for each dataset.

A concise register helps counsel and potential buyers understand what is permissible.

Map personal and sensitive information

Personal data triggers privacy obligations. Conduct a PII Mapping exercise to locate:

• Names, email addresses, phone numbers, and any government‑issued identifiers.
• Health, financial, or location data that may be subject to sector‑specific rules.
• Data that is already anonymized or aggregated.

Document the legal basis for retaining each record. If a record cannot be retained without breach, plan for secure deletion or de‑identification. This step protects you from inadvertent violations during later transfers.

Deduplicate and clean the data

Redundant copies inflate storage costs and create confusion during licensing. Run deduplication checks on large file sets and database tables. Remove obvious duplicates, but keep a log of what was eliminated and why. The log becomes part of the metadata recovery documentation that buyers will review.

Recover missing metadata

Older systems often lose field descriptions, data lineage, or usage notes. Where possible, reconstruct this information by:

• Reviewing system documentation, change logs, and release notes.
• Interviewing engineers or analysts who built or maintained the system.
• Extracting schema information directly from databases.

Even brief annotations improve the credibility of the data package.

Build a sample manifest

Buyers rarely request full datasets before signing a term sheet. Provide a small, representative sample that demonstrates structure, quality, and provenance. The Dataset Preparation service (/preparation/dataset-preparation/) outlines the format we recommend for these manifests. Include:

• A data dictionary for the sample fields.
• Provenance notes for each table or file.
• Any redaction or masking applied.

A well‑crafted sample reduces the time spent on due diligence.

Checklist for a shutdown‑ready archive

Use this list to verify that each critical element is in place before you move to a licensing discussion.

1. System inventory – All production, backup, and legacy systems listed.
2. Ownership register – Rights and provenance recorded for each dataset.
3. PII map – Personal data identified, classified, and treatment decided.
4. Deduplication log – Redundant copies removed and documented.
5. Metadata recovery – Field definitions and lineage captured.
6. Sample manifest – Small, clean subset prepared with a data dictionary.
7. Access controls – Credentials secured, audit logs enabled for any export.
8. Retention schedule – Legal and regulatory hold periods noted.
9. Stakeholder sign‑off – Internal owner, legal counsel, and IT approved the package.
10. Readiness review request – Completed via the brief form (/accession/).

Cross each item before you approach a potential licensee.

Conduct a self‑assessment with the readiness check

Our online tool (/preparation/recovery-planner/) walks you through the same items listed above. It produces a first-pass inventory checklist and identifies questions to resolve. The tool is free and does not require any commitment. Completing it gives you a concrete baseline and a clear path forward.

Align internal teams around the process

Technical staff, legal counsel, and business owners must work together. Assign a single internal owner for the data project. This person coordinates:

• Extraction schedules with IT.
• Rights verification with legal.
• Sample preparation with data analysts.

A single point of contact reduces duplicated effort and ensures accountability.

Protect the data during transition

While you are preparing the archive, keep the data secure:

• Limit access to a defined list of users.
• Use encrypted storage for any exported files.
• Enable audit logging on all export actions.
• Retain original timestamps and file hashes to prove integrity.

If a breach occurs, you may face liability that outweighs any potential licensing revenue.

When to involve external expertise

If the data set is large, highly regulated, or spans multiple jurisdictions, consider engaging a specialist. Dormant Data Partners provides end‑to‑end support, from inventory to licensing documentation. Our How It Works page (/preparation/process/) outlines the stages we follow. You can start with a confidential brief (/accession/) to get an initial feasibility assessment.

Timing considerations

The overall timeline depends on the shutdown schedule, and the following milestones are an Illustrative example, not a promised schedule:

• Week 1–2: System inventory and rights register initiation.
• Week 3–4: PII mapping and deduplication.
• Week 5: Metadata recovery and sample manifest creation.
• Week 6: Internal sign‑off and readiness check completion.
• Week 7: Submit brief for a readiness review.

Adjust the schedule if your shutdown date moves. The earlier you start, the less likely you will lose access to critical systems.

Common pitfalls and how to avoid them

Pitfall	Impact	Mitigation
Turning off servers before inventory	Data loss, incomplete catalog	Freeze decommission dates until inventory is complete
Relying on memory for data context	Inaccurate provenance, buyer distrust	Conduct short interviews and capture notes in writing
Overlooking third‑party license restrictions	Inability to transfer, legal exposure	Review all SaaS contracts and embed findings in the rights register
Exporting full datasets without a sample	Excessive risk, higher cost	Start with a curated sample manifest and expand only after buyer approval

Being proactive about these issues saves time and protects value.

How Dormant Data Partners can help

We do not act as a marketplace, nor do we represent buyers. Our role is to help the data owner create a documented, reviewable asset that can be licensed or archived safely. Our services include:

• Detailed data inventory and rights mapping.
• PII identification and compliance guidance.
• Deduplication, metadata recovery, and sample preparation.
• A clear, written register that satisfies due diligence.

The process is described in full on our How It Works page. For deeper industry perspectives, see the broader collection of articles in our Insights section.

If your organization is preparing to shut down software and wants to know whether the remaining data is a viable asset, start with a confidential seller brief (/accession/). Provide the systems involved, an estimate of data volume, and the anticipated shutdown date. We will return a concise assessment of feasibility and next steps.

---