How to Build a 30-Day Marketing Calendar That Actually Works

Most 30-day marketing plans die around day nine. Not because the ideas were bad, but because the plan asked for thirty separate decisions. By the second week the owner is out on a job, the only decision available is to skip today — and once you have skipped twice, the calendar stops being a plan and becomes a reminder that you are behind.

The fix is not more discipline. It is making one decision instead of thirty. Design a single week — a few recurring post types, each pinned to a fixed weekday — then run that same week four times. The material changes every week. The format never does.

Regular beats relentless

One piece of evidence is worth having before you decide how big to go. Buffer analysed more than 100,000 users across ten platforms over 26 weeks, published in January 2025. Accounts posting in 5 to 19 of those weeks got 340% more engagement per post than accounts posting in four weeks or fewer. Accounts posting in 20 or more got 450% more.

The gap between those two numbers matters more than either one. Going from barely posting to posting regularly is worth +340%; going from regular to near-constant adds only about 26% on top. Almost the entire benefit comes from being regular at all — which is an argument for the calendar you will still be running in November, not the ambitious one you abandon in March.

Buffer’s consistency study across 100,000+ users and 10 platforms

Design the week, not the month

Pick three slots. Three posts a week is about thirteen a month: enough to look alive, few enough to produce in one sitting. If you have abandoned a calendar before, start with two. You can add a slot later — cutting one back feels like failure and usually takes the whole thing with it.

Then give each slot a permanent format. A format is a repeatable question, not a subject. “What did we fix this week” is a format; it produces a post every week for the rest of your life. “Boilers” is a subject, and subjects run out by March.

Four formats that cover almost any SME

• Proof — one job from last week: what was wrong, what you did, how long it took. The format that makes a stranger believe you are good at this.
• Answer — a question a customer actually asked on the phone, in their words, answered properly — even when the honest answer is that they do not need you.
• Face — the people, the vans, the workshop, the apprentice. Cheap to make, and the reason someone is comfortable letting you into their house.
• Ask — what you are taking bookings for and how to reach you. Once a month, plainly, without apologising for it.

Proof and Answer carry the weight — they are what a prospect reads before ringing you. Face fills the gaps and costs almost nothing. Ask replaces a slot once a month rather than being added on top.

A worked 30-day marketing plan for one business

Take a six-person heating and plumbing firm outside Ghent: boiler servicing, repairs, bathrooms, a rising number of heat pump enquiries. The owner — call him Tom — is on the tools four days a week and does his quotes on Sunday evening. A composite of accounts we have run, not one client.

His week: Tuesday Proof, Thursday Answer, Saturday Face. One platform — Facebook, because that is where his customers are. Adding Instagram before Facebook works is how businesses end up doing two things badly. Thirty days from a Monday:

1. Day 2, Tue — Proof: a 22-year-old boiler pulled out on Friday, photographed beside its replacement. What failed, and what the last two repairs cost.
2. Day 4, Thu — Answer: “My radiators are warm at the bottom and cold at the top.” What it means, and the ten-minute fix a homeowner can do alone.
3. Day 6, Sat — Face: both vans loaded at seven, with a line naming who is out where today.
4. Day 9, Tue — Proof: a finished bathroom, before and after, plus the old lead pipework behind the wall that added two days.
5. Day 11, Thu — Answer: “What does a heat pump actually cost to run?” His honest range for a semi-detached house, and the two things that change it most.
6. Day 13, Sat — Face: the apprentice at six months, and the jobs he can now do unsupervised.
7. Day 16, Tue — Proof: a Sunday-night emergency callout, and the one thing the customer did right before ringing — finding the stopcock.
8. Day 18, Thu — Answer: “How often does a boiler really need servicing?” What the inspection rules require, versus what he actually sees fail.
9. Day 20, Sat — Face: the merchant’s counter at half seven, and why three pump models live on the van instead of being ordered in.
10. Day 23, Tue — Proof: a job he turned down — a quote that made no sense on a boiler with two good years left in it.
11. Day 25, Thu — Ask, replacing the Answer slot: four servicing slots left this month, what it costs, and the number to ring.
12. Day 27, Sat — Face: the Friday debrief in the yard, and the thing the team argued about this week.
13. Day 30, Tue — Proof: the month counted up. Eleven callouts for the same fault in four weeks, what causes it, how to spot it early.

Thirteen posts, four decisions. Tom never asks what to post — only which job goes in Tuesday’s slot. That is a two-minute question.

The batching workflow

Five jobs, kept separate, because mixing them is why writing one post somehow takes an hour.

1. Capture, continuously — ten seconds at a time. A shared phone album everyone in the business can add to: before, after, anything odd. No captions, no judgement about whether it is good enough. This is the step everyone skips, and it is where calendars actually die — on planning day you have thirteen slots and four photos.
2. Plan, once a month — 45 minutes. Sit with the album and the thirteen empty slots and assign material to days. Write nothing yet. A slot with no material gets marked and filled from the reserve.
3. Produce, once a month — two to three hours, or two blocks of ninety minutes. Write all thirteen by format: the four Proof posts in a row, then the four Answers. The repetition is what makes it fast.
4. Schedule — 30 minutes. Load the lot into your platform’s own scheduler in one pass. Then the month runs whether or not you feel like it, which is the point.
5. Reply, daily — 10 minutes. The only part that cannot be batched. A comment left sitting four days undoes the post that earned it.

Four hours a month once the habit is in, plus ten minutes a day on replies. The first month takes closer to six, because you are designing the formats as well as filling them.

What to do when a week collapses

It will be week three. A job overruns, someone is off sick, the album is empty, and Tuesday goes past without a post. This is the normal case, not the exception, and a plan with no answer for it ends here.

• Skip forward, never backfill. Thursday’s post goes out on Thursday. The Tuesday post is gone — not owed, not late.
• Keep a reserve of three. Produce sixteen in a good month and publish thirteen. The other three are evergreen and undated, sitting there for exactly this week.
• Have a minimum week: one post, in whichever format is cheapest for you — usually Face. A week with one post still counts as a week you posted, which is what Buffer was measuring.
• Resume, do not restart. There is no clean slate and no fresh month to begin on. You just do Thursday.

A calendar you keep at seventy percent beats a calendar you abandon at a hundred.

— Ben Van Aken, Co-Founder & CTO

The 30-day review

At day 30, give it twenty minutes. In this order, because the order is the point:

1. Did the posts go out? Count filled slots out of thirteen — in month one, the only number that means anything. Nine or more and the system works. Under six and the shape is too big: drop to two slots and run it again.
2. Which slot was hardest to fill? There is always one, and it is a signal about the format, not about your willpower. Swap it for one that runs off material you already generate.
3. Which posts started a conversation? Replies, DMs, saves, and the customer who mentioned it on the phone — not likes. At this size, a comment from someone with a broken boiler is the only real data.
4. What did you fail to capture? Almost always the finished job: the van has left before anyone takes the after photo. Fix the capture habit, not the writing.
5. Change exactly one thing. A calendar redesigned every month is thirty decisions again in a nicer wrapper.

What you should not do at day 30 is judge results. Thirteen posts over four weeks tells you nothing about whether social media brings you work — Buffer’s numbers ran across 26 weeks. Month one tests whether the system runs. Month six tests whether it earns.

The calendar is the easy part

You can build the shape above in an afternoon. Three slots, four formats, a shared photo album, one planning block a month. None of it needs a template you have to pay for.

The hard part is month eleven — producing Thursday’s post in the week your best fitter resigns, and again in the week the accountant needs everything by Friday. Nobody abandons a content calendar because they ran out of ideas. The ideas were never the constraint. The hour was.

That is what a dedicated team is for. Not ideas — you have those, they are in the van — but someone whose actual job is that Thursday’s post exists on Thursday, in your voice, and that Tuesday’s comment got a reply. Our packages start at €595 a month for a dedicated Filipino team doing the planning, writing, design and community management. To test whether a steady cadence changes anything first, €1 a day — €30 a month — buys one post a day on one platform. Three slots a week is the right number when you produce them yourself; when someone else produces, daily becomes sustainable, because the only job left on your side is filling the album while you work.

See what each package includes

Either way, build the week before you build the month. Then run it four times, miss one, and keep going.