It is late November, and the partners are around a table. One says the newsletter really landed this year. Another says referrals have gone quiet. The third is sure the networking breakfasts were a waste of mornings. Nobody has opened the lead log, so the meeting ends the way these meetings usually do: with a feeling, a vague plan and the same activities running again in January.
Why opinions make a bad year-end review
Opinions follow memory, and memory favors whatever was loudest or most recent. The webinar that took three weeks to prepare feels important because it hurt. The quiet referral partner who sent two good clients in spring is forgotten because that work arrived without any effort from you. If you decide the next year from how the last one felt, you will keep the noisy activities and starve the ones that quietly worked.
The fix is not a bigger analytics setup. For a small service firm, the fix is a single document that records where every enquiry came from, plus one honest meeting where you read it. Bring the CRM or the lead log. Leave the opinions at the door, or at least make them earn their place by pointing to a row.
Before the meeting: list everything you actually ran
Start with an inventory, and be unkind about completeness. Write down every channel and activity the firm ran this year, including the ones nobody officially owns. Think of the website and its blog, email, social profiles, paid ads, directory listings, referral arrangements, events, speaking, partnerships, outbound emails and any agency or freelancer retainer. If someone spent working hours on it, it goes on the list.
This step alone is often uncomfortable. Firms find activities that have been running on autopilot for years, with no one able to say why. Put those on the list too. An activity with no owner and no record is exactly the kind of thing a review exists to catch.
Three questions for every line
For each item on the inventory, answer the same three questions, in writing, before anyone argues.
- What did it cost in time? Count the hours of the people who ran it, not only outside spend. Time is the scarce resource in a small firm, and it is the one most often left off the books.
- Did it produce conversations or leads you can trace? Check the lead log. A traceable conversation is a named person who got in touch and can be tied back to this activity. If the log cannot tell you, write that down as the answer.
- Would the team miss it? Ask honestly. Some activities produce no trackable leads but hold something up, such as a directory listing that confirms you exist or a client newsletter that keeps past clients warm. Others would not be missed by anyone.
Notice that the third question is the only subjective one, and it is deliberately last. It is a check on the data, not a replacement for it. When the log says an activity produced nothing and the team says they would miss it, that gap is worth a few minutes of discussion. Usually one side is missing information.
Sorting into keep, kill and test
Keep
Keep what produced traceable conversations at a time cost you can live with, and keep the quiet foundations the team would genuinely miss. Keeping is not the same as leaving alone. For each keeper, write one sentence on what would make it better in Q1, such as a faster reply to enquiries, a clearer call to action or a fixed schedule. A good channel run sloppily is still a waste.
Kill
Kill what consumed real time, produced nothing you can trace and would not be missed. This is the hardest column because every activity was started by someone for a reason. Be firm anyway. A kill decision needs an owner and a date. Without both, it becomes a polite intention that dies in a shared document while the activity carries on. The owner is the person who stops it, cancels it or hands it back. The date is when that happens.
Killing is also where you free the hours that make testing possible. A firm that adds new activities without removing old ones is not planning, it is accumulating.
Test
A test is something you have not proven either way, or a keeper you want to push harder. Limit Q1 tests to one or two. This will feel too conservative, and that is the point. A small team running five experiments at once cannot read any of them, because attention and follow-up are spread too thin to tell a bad idea from a badly executed one.
For each test, write down before you start what you will count and when you will look. If you cannot say what a useful result looks like, it is not yet a test. It is a hobby.
What if the lead log is thin or missing
Many small firms will open the log and find gaps: enquiries with no source, phone calls never recorded, a spreadsheet abandoned in spring. Do not paper over this. The most valuable finding of the review may be that you cannot tell what worked.
In that case, do two things. First, reconstruct what you can by looking at recent clients and asking the people who handled them how they first made contact. Treat the result as rough, and mark it as rough. Second, make fixing the log a keeper-level priority for Q1: one field for source, filled in at the first conversation, by whoever takes it. That is a small habit with a large payoff by next November.
If you want a lightweight way to keep that habit alive between reviews, a <a href="one-page-weekly-scorecard-for-solo-marketers-and-small-service-firms.html">one-page weekly scorecard</a> does the job without turning into a reporting project.
Running it as one meeting
This works as a single meeting because the preparation happens beforehand. Someone, ideally whoever owns marketing, builds the inventory and fills in the time and lead columns from the log before anyone sits down. In the room, the group does three things: challenges the draft sorting, assigns owners and dates to every kill, and picks the one or two Q1 tests.
Keep the debate focused on the exceptions. Where the log and the team's feeling agree, move on. Where they disagree, ask what evidence would settle it, and decide whether it is worth finding out or whether the cheaper move is to run it as a test.
Write the result on one page
The output is a single page with three lists: keep, kill, test. Next to each kill goes an owner and a date. Next to each test goes what you will measure. That page is the bridge into the new year, and it slots straight into a <a href="writing-a-one-page-marketing-plan-founders-will-actually-read.html">one-page marketing plan</a> that founders will actually read.
If you plan in quarters, the same page becomes the opening input for the next cycle. Our <a href="fractional-cmo-quarterly-planning-guide.html">quarterly planning guide</a> covers how to turn it into priorities and check-ins through the year, so the review is not a once-a-year event that everyone forgets by February.
The unpopular truth of a good year-end review is that it mostly produces subtraction. You will leave with fewer activities, clearer ownership and a short list of things to learn. That is a better position than a longer list of plans nobody has the hours to deliver.
See exactly what each Fracmo plan ships. Every deliverable and every price is public — $249, $999 and $2,490 a month, month-to-month, no discovery call to see a number.
See Fracmo pricing →Keep reading
- What is a fractional CMO? — the plain-English 2026 guide
- Fractional CMO cost in 2026 — real numbers, including ours
- AI CMO vs fractional CMO — how the models actually differ
- Compare Fracmo to agencies, in-house hires and DIY tools
- All Fracmo blog guides