The content calendar in an afternoon

Ask any burnt-out creator what killed consistency and the answer is rarely writing — it's deciding. Every day you don't know what to post, you pay the decision tax before you can even start, and most accounts die of that tax, not of laziness. The fix predates AI: plan a month at a time. What AI changes is the price — planning a month used to be a weekend; it is now an afternoon, and this lesson is the recipe.

Pillars first — the part you decide

A calendar generated from nothing produces random content; a calendar generated from pillars produces a strategy. Pillars are the three to five repeating themes your account rotates through — for Kesar & Co.: the making (workshop, artisans, process), the home (products styled in real rooms), the knowledge (block-printing history, care guides, how-to-spot-handmade), and the shop (launches, restocks, the only openly commercial pillar). Pillars are strategy, so they come from you — lesson 1's paragraph plus five minutes of honesty about what you can sustainably produce. If you have no product shots this month, a product-heavy calendar is fiction. Tell the model your real constraints and it plans around them; that's the difference between a plan and a wish.

The generation pass

With voice doc installed (lesson 2), the ask looks like this:

Plan my Instagram content for March. Pillars: making (35%), home (25%),
knowledge (25%), shop (15%). Cadence: 3 feed posts + 2 Reels + daily
stories per week. Constraints: new lamp collection launches March 18 —
build up to it; workshop is closed Holi week, so use archive making-
content there. For each item: date, pillar, format, the one-line idea,
and what I need to shoot for it.

Two minutes later there's a month on the table: thirty-odd items, each with a date, an angle, and — the quietly valuable column — a shot list, so one workshop morning with a phone covers three weeks of visual raw material. The first draft will be 70% usable, and the 30% you reject teaches the next prompt (add the rule, regenerate — same loop as the voice doc). Push back in conversation exactly as with any draft: "week two is all product, rebalance", "give me a Holi-adjacent knowledge post", "the March 18 build-up needs a teaser Reel a week out".

The calendar is a hypothesis, not a contract

Two habits keep the artifact honest. Batch against it: the calendar's real gift is that content creation becomes a scheduled work session instead of a daily scramble — Meera drafts next week's captions every Friday in one sitting (lesson 4 makes this fast), shoots against the shot list in one weekly workshop visit, and the daily decision tax drops to zero. Revise against reality: when a post outperforms, ask the model to lean the remaining month toward what worked ("the artisan-hands Reel did 4× our average — adjust the rest of March"). The plan serves the feed's feedback, never the other way round — a rhythm lesson 11 turns into a proper monthly review.

And keep the plan honestly yours: the model will fill any gap you leave with plausible-sounding generic ideas ("share a motivational quote!") — every one you accept without judgment is lesson 1's slop, scheduled in advance. The calendar multiplies your strategy; it cannot substitute for one.

Do this today: write your pillars with rough percentages, then run the generation pass for the next four weeks with your real constraints in the prompt. Reject freely, regenerate once, and put the shot list in your phone's notes. You've just deleted a month of daily what-do-I-post decisions.