Where to write down why, so it survives

Deciding to write down the reasoning behind a decision is only half the problem. The other half, less discussed and just as important, is where it gets written. A reason recorded in a place nobody will ever think to look is functionally the same as a reason never recorded, except it wastes the effort of writing it in the first place.

The wrong places are everywhere, and they all feel reasonable at the time

A note in a personal notebook. A comment in a Slack or WhatsApp thread that scrolls out of view within days. A separate document titled "decisions log" that lives in a folder nobody but its creator ever opens. An email sent to a colleague that gets buried under the next few hundred emails. Each of these feels like a legitimate act of documentation in the moment. None of them will be found by the person who needs the answer two years from now, because none of them are attached to the thing they explain.

The principle that actually matters: proximity

A reason survives if it's attached to the record it explains, at the place someone would naturally look when trying to understand that record. This is a stronger requirement than "written down somewhere," and it's the entire difference between documentation that gets found and documentation that technically exists but might as well not.

If someone is looking at a customer's account and wondering why their terms are non-standard, the reason needs to be visible from that account — not in a separate spreadsheet, not in an email from three years ago, not in a colleague's memory. The test is simple: when someone encounters the thing that needs explaining, is the explanation right there, or does it require already knowing where to look?

Why separate systems for "the reasons" tend to fail

A dedicated decision log, kept apart from the records it explains, requires someone to remember it exists, remember to check it, and cross-reference it back to the specific record in question. Every one of those steps is a place the habit can break down, and in practice, most of them do, within months of the system being set up with good intentions.

This is the same reason general-purpose SOP documents scattered across shared drives tend to go stale — see the SOP that nobody has opened in a year. Distance from the point of use is the actual failure mode, not lack of effort in creating the documentation in the first place.

Where reasoning actually belongs

Directly on the record, wherever your systems allow a note or comment field against a specific transaction, customer, supplier or configuration setting. This is almost always better than a separate document, because it's found by anyone looking at the record, without having to know a separate system exists.

In the approval step itself, if your process has one. A required reason field at the point of approving an exception means the explanation is captured exactly where and when it's needed, without requiring anyone to remember a separate habit.

As close to the moment of decision as your tools allow, even if that means a slightly awkward workaround — a note field that wasn't designed for this purpose, used consistently, beats a well-designed separate system used inconsistently.

What to check in your own business

Pick three exceptions or non-standard arrangements currently active, and try to find the reasoning behind each one using only the record itself — no asking around, no checking old emails. If you can find it directly, your documentation is in the right place. If you can't, the reasoning either doesn't exist, or it exists somewhere nobody would think to look, which for practical purposes is the same problem.

Common questions

Why isn't a separate "decisions log" document a reliable place to keep reasoning?

Because it requires anyone who needs the answer to know the log exists, remember to check it, and cross-reference it back to the specific record in question. Each of those steps is a place the habit can fail, and in practice a document disconnected from the records it explains tends to go unused within months.

What makes a location for documented reasoning actually work?

Proximity to the record it explains. If someone looking at a customer account, a supplier term, or a configuration setting can see the reason directly, without already knowing where to look, the documentation will actually get found. Distance from the point of use is the main reason documentation efforts fail, not lack of initial effort.

Is a Slack message or email an acceptable place to record a decision's reasoning?

Not reliably. These channels scroll and get buried quickly, and finding a specific message from years ago requires remembering it exists and searching for it — which almost nobody does when they actually need the answer. They're fine for the conversation itself, but the conclusion needs to end up somewhere attached to the record it explains.

How can a business check whether its documentation is actually in the right place?

Pick a few live exceptions or non-standard arrangements and try to find the reasoning using only the record itself, without asking anyone or searching old messages. If it can't be found that way, the documentation either doesn't exist or is stored somewhere nobody would naturally look, which has the same practical effect.

Related: the sop that nobody has opened in a year · a decision log is cheaper than the mistake it prevents · configuration without a reason attached