Document lifecycle management is the control of a document through every stage from creation or capture to final disposition, governing what may happen to it at each stage, who may act on it, and what evidence is kept when it moves between stages.
The defining feature is the transitions rather than the stages. Storage systems model states; lifecycle management models the moves between them, because that is where obligations attach.
The stages
The naming varies by framework, but the substance is consistent.
Creation or capture. The document enters the estate, either authored internally or captured from outside, including from paper. What is recorded here bounds everything later: identity, source, and for captured content the parameters of the capture itself.
Classification. What the document is, which determines its handling. Usually the first point at which retention becomes determinable, and often performed by document classification rather than by hand.
Active use. Editing, review, approval, distribution. Versioning and access control apply here, and most systems do this stage well.
Declaration. The point at which a document becomes a record and its handling changes: typically immutable, retention clock started, deletion requiring authority. Covered under records management.
Retention. The document is held for a defined period, usually anchored to an event rather than to a creation date.
Migration. Not a stage most models include, and one every long-retained document passes through as systems and formats are replaced. Its evidential consequences are the subject of digital preservation.
Disposition. Destruction, transfer to archive, or review for extension.
Lifecycle management is not workflow
The two are adjacent and frequently confused.
Document workflow automation routes a document through a business process: an invoice to approval, a claim to assessment, a contract to signature. Its scope is a process with a defined start and end, and it is finished when the process completes.
Lifecycle management governs the document for as long as it exists, which is usually far longer than any process that touched it. A workflow may run for four days; the lifecycle may run for twenty-five years, across several systems, with no owner named for the intervening period.
The practical consequence is that workflow completion is routinely mistaken for lifecycle completion. The invoice is paid, the case is closed, the project delivered its report, and the document is treated as done while most of its lifecycle has not yet happened.
The transitions that carry obligations
Four transitions do most of the work, and are the ones most often unrecorded.
Into the estate. For captured content, the capture parameters and what was deliberately excluded. A future reader cannot otherwise distinguish an intentional exclusion from an oversight, and the two mean entirely different things under audit.
Into record status. When declaration happened, on what basis, and what the retention trigger is.
Between systems. What was migrated, what changed in the process, and what was verified afterwards. Migrations that record only success or failure lose the detail that later matters.
Out of the estate. What was destroyed, on whose authority, against which retention rule, and what was held back. The evidence has to outlive the content, because the content is gone.
Manifestations complicate the model
A single intellectual record can exist as more than one manifestation: a paper original and a scanned surrogate, or a native file and a rendered archival copy. Lifecycle models that assume one document equals one object handle this badly.
The two manifestations frequently carry different retention rules, and their stages are not synchronized. An original may reach disposition while the surrogate is still in active use. Treating them as one object means one of the rules goes unapplied; treating them as unrelated objects means the link between them is lost, and with it the ability to prove the surrogate represents the original faithfully.
What commonly goes wrong
Retention anchored to creation date. Most real schedules anchor to an event: contract termination, employee departure, case closure. A system that only understands creation-date retention cannot implement them, and the workaround is usually a manual process that stops being followed.
No stage after active use. The estate has creation, use and storage, and nothing else. Documents accumulate indefinitely because no transition out of the estate was ever built.
Disposition configured but never executed. Retention expires and nothing happens, because nobody will authorize deletion. This is the most common state and it is worse than having no schedule, since the policy documents a standard the organization is visibly not meeting.
Lifecycle owned by the project that created the content. Ownership ends when the project does, and the remaining stages have nobody accountable for them.
Contellect One governs classification, retention and audit across these transitions rather than at a single stage, in document and content management.
