Build a 4-Week Calendar Shell: Pillars, Then the 70/20/10 Split
Objective: Define 3-5 content pillars and apply the 70/20/10 split to build a 4-week calendar shell with slots labeled, not filled, before a single post is drafted.
You're the sole social marketer at a Klaviyo-style ecommerce-tools startup. Leadership wants '3 posts a week starting Monday' with no pillars and no plan, and you know an unstructured calendar collapses by week 3.
Define pillars, apply the 70/20/10 split, and produce a 4-week slot shell before any content gets written.
Before you start
What you'll need
Free path (everything below is enough to finish)
Free, shareable, sufficient for planning before any scheduling tool is needed
Paid upgrades (optional, faster/deeper)
Templates speed up batch production once the pillars and slots are locked
The process
2 steps
Step 01 of 02
The lesson says every post must map to exactly one of 3-5 pillars; if a post can't be assigned a pillar, it doesn't belong in the calendar yet.
For a B2B ecommerce marketing tool, what 4 pillars cover the range of things worth posting without overlapping each other?
Procedure
- List candidate pillars: customer wins, product education, founder POV, industry data
- Write one sentence per pillar describing what it accomplishes (awareness, trust, or conversion)
- Check for overlap, if two pillars could both claim the same post idea, merge them
PILLARS 1. Customer wins - trust, real logos and numbers 2. Product education - conversion, feature explainers 3. Founder POV - awareness, opinion and industry takes 4. Industry data - awareness, benchmark stats worth sharing
Healthy
Every pillar has a clear job (awareness, trust, or conversion) and no two pillars would claim the same post idea.
Unhealthy
Pillars are named 'general updates' and 'misc,' which means every post can be filed under either one and neither pillar means anything.
What this means
A pillar without a stated job is just a label; the job is what lets you check whether the calendar is balanced.
So what do I do about it?
| Symptom | Action | Effort |
|---|---|---|
| Two pillars keep claiming the same post ideas during planning | Merge the overlapping pillars into one and rewrite its job statement to be more specific | 5 min |
Step 02 of 02
The lesson's 70/20/10 framework reserves 70% of slots for planned evergreen content, 20% for reactive content left empty until the week arrives, and 10% for experimental formats.
At 3 posts/week for 4 weeks, that's 12 slots total. How many go to each bucket, and which slots stay empty on purpose?
Procedure
- 12 total slots: 70% = 8-9 evergreen, 20% = 2-3 reactive (left blank), 10% = 1 experimental
- Assign a pillar to each of the 8-9 evergreen slots now
- Leave the 2-3 reactive slots' Pillar column empty, they fill in the week they're used, not now
- Mark 1 experimental slot for a format never tried before, e.g. a poll or a behind-the-scenes video
WEEK 1 Slot 1 (Mon): Evergreen, Customer wins Slot 2 (Wed): Evergreen, Product education Slot 3 (Fri): Reactive, [empty, fill this week] WEEK 4 Slot 12 (Fri): Experimental, [format TBD - first poll]
Healthy
9 slots are pre-filled with pillar assignments, 2 slots are deliberately blank, and 1 is flagged experimental, all decided before week 1 starts.
Unhealthy
All 12 slots get evergreen content assigned in advance because 'reactive' felt too vague to plan for, leaving zero room for a real trend or customer story.
What this means
The reactive slots are supposed to stay empty, an empty cell in the shell is the plan working, not a gap to fill in.
So what do I do about it?
| Symptom | Action | Effort |
|---|---|---|
| Every slot in the 4-week shell already has content assigned before week 1 | Clear 2-3 slots per week back to empty and label them Reactive, resist filling them until the week arrives | 5 min |
Final deliverable
A 4-week, 12-slot calendar shell with pillars assigned to evergreen slots and reactive/experimental slots deliberately left open.
See a reference example
Robinhood 4-week shell (excerpt) PILLARS: Investing basics, market moments, product features, community stories WEEK 1 Mon: Evergreen, Investing basics Wed: Evergreen, Product features Fri: Reactive, [empty]
Success criteria
You're done when you can:
- 3-5 pillars are defined, each with a stated job (awareness, trust, or conversion)
- 70% of the 12 slots are pre-assigned to a pillar, 20% are deliberately left blank, 10% is flagged experimental
- No pillar overlaps with another pillar's stated job