Reviewed and updated October 8, 2026.
A workflow step should earn its place through the problem it solves today. Before removing an old approval or automating a manual check, find out why it exists, who relies on it and what could fail without it. Age alone is not evidence that a control is wasteful.
Workflow archaeology is a practical review heuristic: trace the origins of a process, then test its current purpose. It is not a validated research method or a promise of a particular saving. Its value is the evidence it helps a team gather before changing work.
Choose a completed transaction with a clear start and finish, such as a supplier bank-detail change. Walk through the records with the people who performed it. Record the handoffs, systems, decisions, waiting time and rework. Compare the documented procedure with what people actually did, including exceptions and workarounds.
Ask when each disputed step was added. Look for a policy revision, audit finding, incident record or system limitation. Interviews can suggest an explanation, but record an uncertain recollection as uncertain. If no one remembers the rationale, the next task is to identify the present risk, not to assume there is none.
The following is an illustrative example, not a client case or measured outcome. An accounts-payable team wants to remove a duplicate approval from supplier-detail changes.
That record exposes a useful distinction. Two people may appear to duplicate work while checking different risks. A redesign can reduce repeated data handling while preserving independent verification.
Write the hypothesis in operational terms: “Replacing the duplicate full review will reduce waiting without increasing missing supplier confirmations.” Specify which cases qualify, which remain on the old path, who can stop the trial and how quickly the team can restore the procedure. Do not test a control change on transactions whose consequences the team cannot contain.
ASQ describes Plan-Do-Check-Act as an iterative change model that includes a small-scale test and review. Here, the review should consider the quality of the work as well as speed. A quicker process that shifts repair work to another team has not demonstrated a net improvement.
After the trial, record the decision, supporting evidence and unresolved questions. Update the procedure with the control's purpose and owner, not just its sequence of clicks. Revisit it when the system, risk or dependency changes. This gives the next reviewer a usable history instead of another unexplained rule.
Bring one disputed approval, its original rationale and the systems that depend on it. Those details let the review focus on a change that can be tested safely. Explore ALLTIPLY AI roadmap and enablement, or request a project review with the workflow, systems and constraints you need to assess.
Optional Google Analytics stays off until you choose Accept analytics.