Most consultations don't fail during the call. They fail in the week after, when the advice meets a full inbox and quietly disappears. A decision log prevents that. It records each decision, the reason for it, the first concrete step and the number that will show whether it worked, then sets a date to check.
This guide shows you how to set one up in about fifteen minutes and keep it going without adding meetings.
Why good advice gets lost
Good advice gets lost for predictable reasons, and none of them are about the quality of the advice. Knowing them helps you design around them.
- No first step. "Improve follow-up" is a direction, not an action. Nothing happens on Monday.
- No owner. Everyone agreed it was a good idea; nobody was asked to do it.
- No measure. Without a number, the decision can't be judged, so it fades.
- No review date. Even a measured decision drifts if nobody looks at the number again.
- No memory. Six months later nobody recalls why the decision was made, so it gets reversed on a whim.
A decision log fixes all five with one page.
The four-column log
Set up a spreadsheet, or rule a notebook page, with these columns:
| Decision | Why | First step (who, by when) | Measure and review date |
|---|---|---|---|
| Reply to every web enquiry within two business hours | 40% of enquiries waited over a day; slow replies close less often | Office manager sets up a shared inbox and phone alert by [date] | Close rate on web enquiries; review in six weeks |
Add a fifth column later for outcome, filled in at the review.
Writing each column well
Decision. One sentence, specific enough that someone else could tell whether it's being done.
Why. The reasoning in a line or two, including the number that prompted it. This is the column that saves you months later.
First step. A single action with a named person and a date. Not the whole plan; just the first move.
Measure and review date. One number, ideally one you already track, and the date you'll look at it.
A worked example from start to finish
Suppose a consultation with a home renovation company produced three recommendations. Here's how they'd go into the log.
Decision 1: Raise the minimum job price from $800 to $1,000. Why: small jobs tie up crews for a full day at a loss once travel is counted. First step: owner updates the price sheet and tells the office by Friday. Measure: average job value and number of small jobs turned away; review in eight weeks.
Decision 2: Stop the $1,200-a-month print advertising. Why: nobody could trace a single customer to it in six months. First step: owner cancels before the next billing date. Measure: total monthly enquiries, to confirm they don't drop; review in two months.
Decision 3: Ask every satisfied customer for a Google review. Why: a competitor has three times as many recent reviews. First step: office manager drafts a short message template by next Tuesday. Measure: new reviews per month; review in one quarter.
Now the arithmetic. If the minimum-price change lifts average job value by $150 across 25 jobs a month, that's $3,750 more revenue a month. The print saving adds $1,200. Together that's nearly $5,000 a month, from two decisions that took ten minutes to log. Without the log, there's a good chance neither would have happened.
Setting review dates that fit the change
Different changes show up at different speeds, so review dates should vary too.
- Follow-up, pricing and scripts: four to eight weeks. They act on enquiries you already receive.
- Paid advertising changes: one or two billing cycles.
- Website changes: six to twelve weeks, depending on traffic volume.
- Organic search and content: quarterly. Growth here compounds slowly; Roma Heating & Cooling grew organic traffic 400% over 12 months, a result from our HVAC work that shows the realistic timescale.
Put the review dates in your own calendar the day you log the decision.
What to do at a review
At each review, answer three questions and write the answers in the outcome column:
- Did we do the first step? If not, why not, and who will do it now?
- Did the number move? Up, down or flat, and by how much?
- Do we keep, change or reverse the decision?
If the number moved the wrong way, don't panic and don't ignore it. Check whether something else changed at the same time, such as a slow season or a lost staff member. If the decision itself was wrong, reverse it and note what you learned.
Keeping a partner or team aligned
The log is also a communication tool. Share the relevant rows with anyone carrying out a decision. They see the reasoning, the deadline and the measure, which removes most of the "why are we doing this?" conversations.
For owner partnerships, a monthly fifteen-minute look at the log replaces a lot of debate. You're discussing evidence, not opinions.
Common mistakes with decision logs
- Logging too much. Ten decisions at once means none get done well. Three is plenty.
- Vague measures. "Better marketing" can't be reviewed. "Enquiries per month" can.
- Skipping the why. It feels obvious today; it won't in six months.
- Never closing entries. A decision that's been reviewed and settled should be marked done, so the log stays readable.
A fifteen-minute setup checklist
If you want to start today, this is the whole job:
- Open a new spreadsheet and name it "[Business name] decisions".
- Add five column headings: decision, why, first step, measure and review date, outcome.
- Enter the top three recommendations from your most recent conversation with any advisor.
- Give each one a named person and a date for the first step.
- Put every review date in your calendar before you close the file.
How this fits with working with Ali
In a paid strategy session, the written summary is designed to drop straight into a log like this. In ongoing advisory, Ali keeps a running decision log with you and reviews it each quarter.
If you haven't had your consultation yet, start with our guide on preparing for it, and when you're ready to talk, the contact page explains how to book by call, text or email.
Frequently asked questions
Do I need special software for a decision log?
No. A spreadsheet with four columns is enough, and a notebook works too. What matters is that every decision has a date, a reason, a first step and a measure.
How often should I review the log?
Set a review date for each decision when you record it. For follow-up or pricing changes, four to six weeks is usually enough to see a trend. For organic search or content, review quarterly.
What if a decision turns out to be wrong?
Record that too, with what you learned. A reversed decision isn't a failure if you caught it early because you were measuring. That's the whole point of the log.
Should my staff see the log?
Anyone responsible for carrying out a decision should see the relevant entries. It saves explaining the reasoning again and keeps everyone working from the same plan.