Every architecture firm has a rework problem. It shows up in different places — in cost overruns against project budgets, in schedule compression at the end of phases, in exhausted teams working through weekends to recover from something that went wrong earlier.
Most firms track it by project, in aggregate: "this one went over on hours." Fewer have done the analysis to understand the specific, recurring causes in their practice. That analysis is worth doing, because the causes are addressable.
The Most Common Root Causes
Undocumented phase decisions that get relitigated. When SD decisions aren't formally documented and approved, they get revisited in DD. When DD decisions aren't formally approved, they get revisited in CDs. Every time a phase decision is reopened and changed, the work that implemented the original decision is rework. This isn't the client's fault: it's a failure to formalize decisions at phase boundaries.
The fix is phase-close discipline: a written record of what was decided at the end of each phase, a formal client approval of that record, and a clear protocol that changes to approved decisions are scope additions subject to fee impact.
Version conflicts generating duplicated work. A designer produces detailed work based on a drawing that's already been revised. When they discover the conflict, the work has to be redone. This is a document control failure with a direct rework cost, visible and measurable if anyone is tracking it.
Late consultant input on already-detailed work. The structural engineer reviews the mechanical room layout after the architect has already detailed the ceiling, ductwork routing, and finish schedule at that location. Their feedback requires changes to work that's already complete. This is a coordination sequencing failure: the consultant wasn't engaged early enough to provide input before work was detailed.
Scope additions absorbed as normal design changes. A client asks for "just one more option" or "a small change" that requires reworking drawings already at CD level. When this is absorbed into the fee as a design iteration rather than priced as an additional service, the rework cost comes directly out of project margin. It happens because scope boundaries aren't defined clearly enough to make the addition recognizable when it occurs.
Specification-drawing conflicts caught late. Specs that reference materials or products that don't match the drawings require revision to one or both, typically under deadline pressure during CD production. The rework is concentrated, expensive, and almost entirely preventable with better coordination discipline during spec development.
The Leverage Points
Pick one.
If your rework most often comes from phase decisions being relitigated: invest in phase-close formalization. Written approval, signed or confirmed in email, treated as binding. This one change, consistently applied, dramatically reduces the "but I thought we were still exploring that" conversations in DD.
If it's version conflicts: invest in document control. A clear current/archive structure, automated notifications when files update, a team culture where nobody works from a drawing they pulled from an email.
If it's late consultant input: invest in coordination sequencing. Consultants engaged earlier, with clear input deadlines by design phase, and a coordination review cadence that catches conflicts before the detailed work is done.
Each of these is addressable. None of them requires a complete practice overhaul. They require identifying which cause is driving the most rework at your firm and building the process specifically targeting it.