Decision Logs with Rejected Alternatives for the CCAR-P Exam

Recording the why, not just the what

The handoff recipient needs the reasoning behind a design, and the decision log is where that reasoning lives. The CCAR-P exam treats it as an understand-level skill with a specific requirement: a decision log records not just the decision that was made, but the alternatives that were rejected and the trade-off each rejection resolved. The rejected alternatives are the part most logs omit, and they are the part that makes the log usable to someone who was not there.

The reason is straightforward. A decision recorded on its own tells a successor what shape the design took. The rejected alternatives tell them why it took that shape, which is the information they actually need to change it safely. A log that lists only chosen options is a record of outcomes with the reasoning stripped out.

Why a decision without its alternatives is unreadable

Consider a decision to use a particular context strategy. Logged as "chose strategy X," it looks complete, but a successor reading it later cannot tell whether strategy X was chosen for performance, for cost, for a compliance constraint, or simply out of habit. Logged with its rejected alternatives, "chose strategy X over strategy Y because Y would have moved regulated data out of region, and over strategy Z because Z added a retrieval layer we did not need", the successor immediately sees which considerations are load-bearing. The rejected alternatives are the difference between a decision a reader can reason about and one they can only accept or blindly overturn.

The failure this prevents

The concrete danger of omitting rejected alternatives is that a successor reverses the right decision for the wrong reason. Acting reasonably on an incomplete log, they change a choice that was actually load-bearing for a constraint they could not see, and reintroduce a problem the original design had solved. This is exactly the failure traced in diagnosing rationale loss in a handoff failure: the decision looked like a preference, so the successor changed it, and it turned out to be compliance-critical. The rejected alternatives are what would have flagged the decision as non-negotiable.

"Obvious" is where rationale dies

The reason decision logs get written without rejected alternatives is that the reasoning feels obvious to the person who lived the design. When you made the choice, the alternatives you rejected and why are vivid, so writing them down feels redundant. But that vividness is exactly what a successor lacks, and it walks out the door with you. Assuming the reasoning is obvious and therefore does not need recording is the specific habit that loses load-bearing rationale at handoff. The discipline is to write down the alternatives precisely when they feel too obvious to bother with.

What the exam trips candidates on

The first trap is logging only the chosen option and omitting the alternatives that were seriously considered and rejected. The exam presents a tidy log of decisions and asks whether it is adequate for handoff; the credited answer notes that without the rejected alternatives, the log cannot be understood by someone who was not in the room.

The second trap is assuming the reasoning behind a decision is "obvious" and therefore does not need to be written down. The exam rewards recognising that obviousness is a property of the person who made the decision, not of the document, and that the rationale most worth recording is often the one that feels self-evident.

Common misreadings to avoid

How this shows up on the exam

Questions show a decision log and ask whether it will survive a handoff, or which element it is missing. The reliable reading is that a usable log records the decision, the rejected alternatives, and the trade-off each resolved, that omitting the alternatives makes the log unreadable to an outsider, and that "the reasoning was obvious" is not a reason to skip it.

This serves the handoff recipient identified in the three readers of architecture documentation, is one of the two artifacts checked by the documentation completeness test, and its absence is the root cause in diagnosing rationale loss in a handoff failure.