Most marketing plans you will be handed are decks. They have goals, a market overview, a list of channels, and a slide that says “measure and optimize.” Nothing in them tells you who is doing what by when, and nothing in them has been decided — because listing eight channels is not a decision.
A plan you can actually run has eight parts. Here is what each one has to contain before it is worth anything, and then a short checklist for reviewing one.
1. The ICP, narrow enough to disqualify someone
“Mid-market B2B companies” is not an ICP. It is a category. A usable definition names the industry, the revenue or headcount band, the buying trigger, the person who signs, and the person who blocks the deal.
The test is disqualification. Read the definition, then name three real companies you have talked to in the last year that it rules out. If you cannot, the ICP is not doing any work — it is a description of everyone, and every later decision about channels and messaging will have nothing to push against.
2. Positioning and the messages that follow from it
One sentence on what you are, who it is for, and what the alternative is. Then the two or three claims you will repeat all quarter, and the proof you have for each.
The proof column is where most positioning falls apart. If a claim has no customer, no number, and no demonstration behind it, it is an adjective. Either find the proof or drop the claim — a quarter spent asserting something you cannot support is a quarter spent teaching the market to discount you.
3. A channel mix short enough to execute
Three channels, maybe four. Each one gets a named owner, a budget, and a hypothesis you can be wrong about: who it reaches, what it is supposed to produce, and roughly how much of it.
A long channel list is the most reliable tell that nothing was prioritized. Eight channels across one marketing manager and a freelancer means eight channels run badly, and at the end of the quarter you will not be able to say which of them worked, because none of them got enough budget or attention to produce a readable signal.
This is the stage we call Rank, and the job at Rank is not ordering the list. It is deciding what not to do.
4. Budget, allocated by line
Not a total. A split: how much goes to each channel, how much to content and creative, how much to tooling, and how much is held back for whatever the first six weeks turns up. Reserve something. A plan with every dollar committed on day one cannot respond to a pilot that works.
5. An owner and a date on every initiative
One name per initiative, not a team. Teams do not have calendars.
Dates should be delivery dates for something specific — the landing page is live, the sequence is sending, the first report exists — rather than “ongoing” or “Q3.” Anything marked ongoing is not an initiative; it is a job description, and it belongs somewhere else.
6. KPI baselines, written down today
For each number you will be judged on, record what it is right now, how it is calculated, and where it comes from. Qualified leads, pipeline created, conversion rate by stage, acquisition cost — whichever few actually map to how the business makes money.
Do this before anything launches. Baselines agreed after campaigns are running get negotiated, because by then everyone has something to defend. This is the one section that is nearly impossible to add retroactively.
7. What you are not doing this quarter
An explicit list, in the plan, of the things that were considered and declined, plus the things currently running that will be stopped.
This section is the single best signal of whether real work happened. Saying yes to five initiatives costs nothing. Saying out loud that the conference is being skipped, the podcast is being shelved, and the display spend is ending is where the capacity for the five actually comes from. A plan with no stop list has added work to a team that was already full.
8. The review cadence
When you look at the numbers, who is in the room, and what triggers a change. Monthly against the baselines, re-planned at the end of the quarter, with a stated rule for killing a pilot early. Without that rule, failing initiatives run the full 90 days because nobody wants to be the one to stop them.
How to review a plan in ten minutes
Four questions, in this order:
- Does every initiative name an owner, a date, and one number it moves? Count the rows that fail. If it is most of them, the document is a deck.
- Is anything being stopped? Look for the stop list. If it is absent, ask what gets dropped to make room, and listen for whether the answer exists already or is being invented on the spot.
- Can you tell what gets cut if the budget halves? A real plan is ordered, so the answer is visible without a meeting. If halving the budget would require starting the planning over, nothing was ranked.
- Could this plan have been written for your competitor? Swap your company name out. If it still reads fine, the ICP and positioning sections are generic and the rest is built on them.
A plan that passes those four will still be wrong about some things. That is fine and expected — being specific enough to be wrong on schedule is the whole point, and it is why the baselines and the review cadence matter more than the forecast.
Where this comes from
At Deep Quiver the plan is the output of the 30-Day Marketing Diagnostic — a fixed-scope audit of your market, funnel, content, data, and martech that ends with a ranked roadmap, owners, and agreed KPI baselines. The diagnostic exists first because every section above is easier to settle before anyone is attached to a plan. For how the ongoing engagement works after that, see the guide to fractional CMOs and our fractional CMO services.
Take your current plan and run the four review questions against it. Talk to a CMO if you want a second read on what comes back.
