Build the map before the move

Abstract

A system migration moves more than files; it moves ownership, retention, and access decisions into a new environment. An information asset inventory gives the migration a map of what exists, what is authoritative, and what should not travel. This review draws on public records and security guidance to set a bounded inventory method for an executive office. It also explains how to test the inventory against real requests before cutover [1][2][3].

Methodology

The review uses public records management and cybersecurity framework guidance as a structure for inventory design, not as a source of local data. It defines an information asset as a body of information with value that needs lifecycle management. The method separates systems, official records, and working files, and assigns an owner and a retention class to each asset. Testing uses a small sample of real requests before and after migration [1][2][4].

Why an inventory comes before migration

A system migration is usually planned around applications, but the risk sits in the information those applications hold. Files with no owner, duplicate folders, and records past their retention date move into the new system unless someone identifies them first. An information asset inventory gives the migration a map of what exists and what should not travel. Without it, the new system inherits the old disorder [1][2]. The inventory also supports access design. A migration is the moment to decide who should see which records, rather than copying every permission forward. That decision requires a description of the asset, its owner, and its sensitivity. The inventory is the evidence for those choices [3][4].

Start the inventory with the questions the migration must answer: which records are authoritative, who owns them, and what must be retained. Those questions shape the fields and prevent the inventory from becoming a file count. A count is easy to produce and hard to use. A decision map is harder to build and far more valuable.

Define what counts as an information asset

An information asset is a body of information that has value and needs management through its lifecycle. It can be a contract repository, a board pack archive, a vendor list, or a set of working files that supports a recurring process. The definition should be broad enough to include informal stores, such as a shared inbox or a personal drive. Those are often the hardest to migrate and the easiest to overlook [1][2]. Group the assets by purpose rather than by application. A contract may exist in a legal system, an email thread, and a shared folder, and all three copies need a decision. The inventory should name the authoritative copy and the duplicates. That distinction prevents the migration from preserving three versions of the same record [2][4].

Include the places where information lives between systems, such as shared inboxes, chat channels, and local drives. These are often the source of the documents people actually rely on. If they are excluded, the inventory will describe the official systems but miss the working reality. Add them with a clear owner and a plan.

Separate systems, records, and working files

A system is the tool that holds information, a record is the information that must be retained, and a working file is the intermediate material used to produce it. The three need different treatment. Systems are migrated, records are retained under a schedule, and working files are usually disposable after the work is complete. Mixing them creates unnecessary cost [1][3]. Ask the owner to identify the record that would be needed to reconstruct a decision or prove compliance. Everything else is likely a working file. This test is faster than classifying every document. It also gives the migration team a defensible reason to leave low-value material behind [2][4].

A record can exist in several formats and locations, so the inventory should identify the authoritative version rather than every copy. The other copies can be treated as convenience or working files. This reduces the volume that must be migrated and the risk of conflicting versions. State the rule for choosing the authoritative copy.

Assign an owner and a retention class

Every asset needs a business owner who can explain its purpose and confirm its retention class. The owner is not necessarily the person who created it. For a board archive, the board secretary may be the owner, while the support desk maintains the files. The inventory should name the role rather than only the individual [1][2]. Use a small set of retention classes and define the trigger for each. A contract may be retained for the term plus a set period, while a routine schedule may be kept for a shorter operational window. The class should be visible in the inventory so the migration team knows what to carry forward [3][4].

Retention classes should be few enough to remember. A long taxonomy invites inconsistent choices and slows the migration. Use a short list, define each class with an example, and let the owner confirm the class. The inventory records the decision; it does not make it.

Test the inventory against real requests

A good inventory should help someone find a real document. Pick five recent requests and trace how each would be answered using the inventory. If the map does not lead to the record, the inventory is incomplete or the ownership is unclear. This test is more useful than a theoretical completeness review [2][4]. Record the time and the path for each test request. If the answer depends on one person's memory, that dependency should be documented as a risk. The migration plan can then include a fix, such as consolidating duplicates or assigning an owner [1][3].

Include requests that failed as well as requests that succeeded. A failed search shows where the inventory or the ownership breaks down. Ask what the requester did next and where they eventually found the document. That path is often the most useful evidence for the migration plan.

Plan migration by dependency, not by folder

Some assets depend on others. A board pack archive may need the meeting calendar and the decision log to be useful, while a vendor repository may need the contract index. Migrating in dependency order reduces broken links and missing context. A folder-by-folder plan often separates records that belong together [2][4]. Define a minimum usable set for each asset and test it after migration. The set should include the records, the index, and the access rules. If the set is incomplete, the asset is not ready to go live. This definition prevents a migration from being declared complete when only the files have moved [1][3].

Document the dependencies in the inventory so the migration team can see why an order was chosen. A short note is enough, such as the calendar being needed to interpret board pack dates. Without that note, a later team may reorder the work and break the context. The dependency is part of the asset description.

Keep the inventory alive after cutover

An inventory that is built once and never updated becomes a historical artifact. Assign a review cadence and an owner for the inventory itself. Update it when a new system is introduced, a record type changes, or a business owner leaves. A quarterly review is enough for most offices [1][4]. Link the inventory to the retention schedule and the access review process. That way a change in one place prompts a check in the others. The inventory should be a working control, not a project deliverable. Its value is in the decisions it supports after the migration [2][3].

Assign the inventory review to the same role that owns records management, not to a temporary project team. Give that role the authority to request updates from business owners. A quarterly reminder is usually enough, provided the owners know what changed. The inventory should outlive the migration project.

Limits and responsible use

The inventory cannot make a legal retention decision for the office, and it cannot guarantee that every informal store has been found. It also depends on owners providing accurate descriptions. A review should state the coverage and the known gaps rather than claim completeness [1][3]. Protect sensitive information by restricting the inventory itself and by describing assets at the right level. Do not list personal or privileged details that are not needed for migration planning. Review the inventory with legal and security where the scope requires it. The method supports good judgement rather than replacing it [2][4].

State clearly what the inventory does not cover, such as personal devices or vendor-held data that the office cannot inspect. Those gaps should be listed as follow-up items rather than hidden. A reviewer who knows the boundary can judge the risk. An inventory that claims completeness invites false confidence.

Key findings

  • A migration should start with an information asset inventory, not an application list [1][2].
  • Separate systems, official records, and working files because each has a different lifecycle [1][3].
  • Name an authoritative copy for every asset and assign a business owner and retention class [2][4].
  • Test the inventory against real requests and migrate in dependency order [2][4].
  • Keep the inventory updated after cutover and restrict access to its sensitive detail [1][3].

Sources

  1. Guidelines for the Selection, Configuration, and Use of Transport Layer Security Implementations, National Institute of Standards and Technology
  2. Records Management, U.S. National Archives and Records Administration
  3. The NIST Cybersecurity Framework, National Institute of Standards and Technology
  4. Standards for Internal Control in the Federal Government, U.S. Government Accountability Office

Return to the Research Content library