Project Decision Audit Record Checklist

Write down the kinds of decision you will record, such as anything changing what is built, what is bought or who does something, and hold to that definition.

A stated boundary is what stops a decision record turning into a diary, which is the usual way these records are abandoned.

Record decisions that close off an option even where nothing visible changes, since those are the hardest to explain when somebody asks about them later.

The decision not to do something leaves no trace on site at all, which is exactly why it needs a written one.

Include the decisions you made by yourself without discussing them, because a choice made quietly is still a choice somebody may ask you about.

Owners record the decisions taken in meetings and forget the ones they made at a kitchen table, which are often the more consequential.

Note the decisions that were made for you and reported afterwards, and record them plainly as such rather than as decisions you took.

Where a decision was taken on your behalf, ask the appointed professionals what authority it was taken under and record their answer.

Avoid recording every preference and passing thought, since a record crowded with opinions stops being read and therefore stops being useful.

The value of the record is that it is short enough to read at the moment somebody is standing in front of you asking a question.

Give each entry a short plain title of its own, so a decision can be referred to without anybody having to read the whole entry aloud.

Untitled entries get referred to by description, which is how two people end up discussing two different decisions with confidence.

Record what information was in front of people when the decision was made, including which drawing or document was being looked at.

A decision is only fairly judged against what was known at the time, and that context evaporates faster than anything else.

Write down the reason given at the time rather than the reason that seems obvious afterwards, even where the two appear to be the same.

Reasons are reconstructed unconsciously, and the reconstructed version is usually tidier and less accurate than the original.

Note where the decision rested on advice and who gave it, without repeating that advice as though it had become your own conclusion.

Advice belongs to the professional who gave it; record who said what rather than converting it into a statement of fact in your own record.

Record which options were considered and set aside, and why, since the discarded option is what somebody will propose again later.

Without this, a rejected idea returns as a fresh suggestion and the whole discussion has to be held a second time.

Note anything that was unknown when the decision was made and was expected to be resolved later, and keep that as an open point.

Where a decision depends on something still to be established, ask the appointed professionals to confirm the position once it is known.

Record where a decision depended on something being true about the property that nobody had yet established.

Facts about existing construction must be established by qualified professionals; a decision resting on an unchecked assumption should be revisited.

Write down decisions reached during a site walk before you leave the property, since these are precisely the ones nobody follows up with an email.

Site decisions are made confidently and remembered vaguely, and the gap between the two opens almost immediately.

Note where a decision was reached by a nod rather than by words, and confirm it in writing before anybody acts on it.

Ask the appointed professionals which route counts as an instruction on your project; agreement in conversation may not be one.

Record who else was present when a verbal decision was reached, since that is what makes it recoverable if the two of you remember it differently.

A third person who heard the exchange is often the only way a disagreement about a site decision gets resolved without ill feeling.

Note decisions taken by telephone, including who called whom, and write down what you each said you would do at the end of the call.

Telephone decisions leave no artefact whatsoever, which makes them the least recoverable of all and the easiest to record while fresh.

Write down decisions made while looking at something physical, and photograph whatever was being looked at as part of the entry.

A decision about a junction, a level or a position makes very little sense in words alone, even to the person who wrote them.

Copy decisions made in a messaging thread out of that thread and into your record, rather than trusting that you will find them again.

Messaging threads scroll, get muted and search badly, and they are where the most consequential quick decisions tend to live.

Record the decisions that took effect because nobody objected, and note who was in a position to object at the time.

Absence of objection is treated as agreement by everybody afterwards, which makes it worth capturing as though it were an active choice.

Note where a choice was closed because something was ordered, and record when you learned that it had been.

Ordering is a decision with a physical consequence, and it frequently precedes the conversation an owner thought was still open.

Write down where a question of yours was never answered, and treat the position that resulted as a decision that was made anyway.

An unanswered question does not leave a project in suspense; it leaves it proceeding on whatever assumption was most convenient.

Record where you were asked to decide without being given what you needed to decide, and note what you did in that situation.

Where you are pressed for a decision without adequate information, say so in writing and ask the appointed professionals what is missing.

Note where a decision was taken because it was the last moment to take one, since those deserve revisiting more often than they get it.

A decision made under time pressure is not necessarily wrong, but it was made without the consideration the others received.

Record where a standard or default option was used because nothing had been specified, and note who chose that default.

Defaults are invisible in the finished work and completely invisible in the paperwork, until somebody asks why it is like that.

Record the name and role of whoever made each decision rather than the practice or company alone.

People move on and a company name answers nothing later, whereas a name and role can usually still be followed up.

Note who was present when a decision was made and, separately, who was told afterwards, because the second list is where things go wrong.

Decisions fail far more often in distribution than in substance, and only a written list makes the omission visible.

Record where one household member decided on behalf of others, so the position is clear if somebody later says they were not consulted.

Household disagreements about decisions surface late and are much easier to settle where the original entry says who decided.

Note where somebody made a decision you had understood was yours, and raise it at the time rather than letting the position stand by default.

Where decisions are being taken outside the route you agreed, raise it with the appointed professionals and take unclear wording to your own adviser.

Record where a decision needed somebody who was not there, and what was done in their absence.

Where a decision needs a professional who was not present, ask them to confirm the position rather than treating the site view as settled.

Write down who is responsible for passing each decision to the people who have to act on it, and how you would know that had happened.

Nobody is responsible for distribution unless somebody is named, and everybody assumes the person who made the decision has done it.

Keep the original entry and add a new one when a decision changes, rather than editing the record so it shows only the current position.

A record showing only the current position cannot answer any question about how the project arrived there, which is what people actually ask.

Record what prompted the reversal, since a change caused by new information is an entirely different matter from a change of mind.

The distinction matters to everybody involved and becomes impossible to establish once the project has moved on.

Note what had already been done on the basis of the original decision at the point it was reversed.

Where work has already been carried out on a reversed decision, ask the appointed professionals what has to happen to it before anything continues.

Record who was told of the reversal and, more importantly, who was not, because a party still acting on the earlier decision is the immediate problem.

Reversals travel more slowly than original decisions, because the people who need them most are usually already at work.

Note decisions that have been revisited more than once, since a decision that will not stay settled usually has an unresolved question behind it.

Repeated reversal is a symptom rather than a nuisance, and the underlying question is normally identifiable once the pattern is visible.

Ask whether a reversal changes any document, and record which documents were reissued as a result of it.

Where a reversal affects the drawings or specification, ask the appointed professionals to reissue them rather than relying on correspondence.

Read the record before each meeting and bring out the entries that relate to whatever is going to be discussed.

Arriving with the earlier decision in front of you changes the conversation from recollection to confirmation, which is much faster.

Read the relevant entries again when the work reaches an area you decided about early, since the reasoning fades much faster than the decision itself.

Owners remember what they chose and forget why, which leaves them unable to defend a decision that was perfectly sound.

Answer questions about what was agreed from the record rather than from memory, so you do not create a second version alongside the written one.

An owner who answers from memory and later finds their own note disagreeing has weakened both accounts at once.

Note where your record and somebody else's account of a decision differ, and raise the difference rather than quietly adjusting your own entry.

Where accounts of a decision differ, put both to the appointed professionals and take any question of obligation to your own legal adviser.

Keep the record after the work has finished, since questions about why something was done a particular way arrive long afterwards.

The most common reader of a decision record is somebody planning further work on the same property, often long afterwards.

Review the entries marked as depending on something still to be established, and check whether it ever was.

Open points recorded honestly at the time are the single most useful thing in the record, provided somebody goes back and reads them.