An Instagram content plan comes down to four decisions, and none of them is the list of post ideas. First: a target action you measure the channel by, meaning profile visits, clicks on your link, newsletter sign-ups or enquiries. Second: three to five content pillars from your own business, meaning from the questions your customers actually ask you, and not from a template that reads "motivation, behind the scenes, offer". Third: a cadence you pin to your measured production time and not to a recommendation you read somewhere. Fourth: a fixed production day on which you batch several weeks in one pass. On top of that comes a monthly review that does not judge single posts but pillars and formats, and from which one change follows. Because the point where most plans fail is not the idea, it is the cadence: two posts a week that arrive every week are worth more than seven in week one and none in week three.
I no longer run my channel on instinct but on pillars and cadence, and the draft for every post is written by an AI employee. What I collect about AI at work sits on the AI page, who does what at my place is in My AI Workforce, and how I decide what to hand over is in Which Tasks You Can Hand Off to AI. How a single reel comes together is in Can AI Create Instagram Reels?, the opening inside the video in How to Write Strong Instagram Reel Hooks, the text underneath it in How to Write Instagram Captions. This post is the level above all of them: what should exist over the next eight weeks.
Deriving content pillars from your own business
A pillar is not a topic, it is a recurring kind of post from which you could write twenty pieces without research. That is my test, and most template pillars fail it: from "behind the scenes" you get three posts, from "the questions that always come up in a first call" you get twenty. Four sources work for this, and all of them sit in your business, not on the internet.
Three pillars are enough, five is my ceiling. That is not a specification from Instagram, it is the range that has proven itself in my own template. Pillars are also the reason not to post something: the holiday greeting, the repost and the quote graphic fit into none of them. An idea without a pillar is not a post, it is a notion.
Writing down your own pillars is the point where most people stall, because you stand too close to your own business to see it. If you would rather not guess, put your three candidates up for discussion in my community or bring them into a call. From the outside it takes ten minutes to see which pillar carries twenty posts and which one is empty after three.
Cadence: two that arrive beat seven that stop
This is where it is decided whether the plan survives week eight, and here almost everyone makes the same mistake: they set the cadence by a recommendation instead of by their own production time. So measure first how long a post really takes you, once, with the clock running, including the quarter hour in which you watch the video three times and change nothing.
The example values are neither mine nor yours, they are only there so the table is not empty. Enter your own, and then calculate in a direction nobody enjoys: how many minutes a week do you really have left for content, in a normal week, not a good one? Divide those minutes by your total. The result is your cadence, and now comes the part that makes the difference: round down, not up. Whoever lands on 2.4 posts plans two. The reserve is the buffer for the week with the emergency appointment, and that week decides whether the plan carries on.
Whoever shows something twice a week for half a year has forty posts to learn from and a basis for comparison over time. That a particular cadence brings more reach is nowhere stated by Instagram; the smaller claim is the safe one: a cadence you keep produces posts and numbers.
Batching four weeks in one pass
The second reason plans die is setup time. Whoever makes one post at a time switches mode five times: thinking, writing, speaking, editing, typing. Every switch costs a run-up, and whoever makes four posts together pays it once. So order the production day by steps, not by posts; the same cycle applies if you have the reels made for you and only supply the raw material.
Without the ongoing stock of raw notes the production day turns into a brainstorm, and brainstorms run longer than planned. Two house rules hold the day together. First: plan five when you need four. The fifth is the reserve and the slot for whatever is current. Second: the plan is not a contract. If something better happens, you push the planned post back. It is finished anyway.
What you plan with
The smallest decision of all, and the one people talk about most. The plan itself needs no more than a table with five columns: date, pillar, idea, format, status. For scheduling, and for the numbers in the next section, the same condition applies: a professional Instagram account, business or creator. Meta's developer documentation states explicitly about the insights that they exist only for professional accounts, not for private ones. Which tool you pick changes nothing about your plan; the pillars and the cadence change everything, and which tool is enough for which task has its own comparison.
What to measure in order to adjust the plan
Do not start with the platform numbers, start with the question of what the channel is supposed to do for your business. Set a target action you can count: qualified profile visits, clicks on your link, newsletter sign-ups or enquiries. That is the number the plan hangs on. Everything else Instagram shows you is a platform early indicator: it tells you how your posts are distributed and received, not whether somebody enquires or books.
Which target action fits you is nobody else's call to make. If you are torn between two candidates, bring the question into my community or into a call: the people there have the step behind them and will tell you quickly which number turned out to be useful for them.
For the early indicators the rule is: judge pillars and formats instead of single posts, once a month. Instagram recommends the same direction for its own insights, namely to watch for recurring patterns and to break down the accounts you reached into followers and non-followers (published 31 May 2023). Three metrics are enough.
Accounts reached that do not follow you yet, per pillar. That is reach outside your following, not growth in itself: whether enquiries come out of it is answered by your target action, not by this number. Still the most useful breakdown in the tool, because a pillar that has reached almost nothing but followers for two months binds, but it does not grow.
Reshares and full views. Instagram names as its most important predictions for reels how likely you are to reshare a reel, watch it all the way through, like it, and go to the audio page (published 31 May 2023). Resharing sits at the front, liking at the back. For Explore the same source names likes, saves and shares as the most important predicted actions, and in the feed a viewer's own activity includes what they have saved. A fixed weighting for every account does not follow from that. That I rate saves higher than likes is my house rule: twelve likes and seventeen saves are worth more to me than forty and two.
Watch time and completion rate. Rough signals for whether a format carries. An average watch time tells you how long people watched on average, not at which second somebody drops off; a specific drop-off point can only be read from a retention curve, and the insights do not supply one. What they do supply is the comparison: Instagram names views, watch time, followers and interactions as metrics for video performance and lets you place up to three reels side by side (published 19 August 2026). If a format regularly comes last in that comparison, I look at the opening first: How to Write Strong Instagram Reel Hooks.
Out of the review comes exactly one change per month, another house rule: whoever changes three things at once will not know next month which of them worked. And the setback belongs in the first sentence, not in a footnote.
Who fills the plan at my place
In my business that is an AI employee. Conny takes a content assignment, reel, hook, script, carousel or caption, and delivers a draft ready for approval in my own voice, checked against audience, content pillars and the house rules. I film, I speak and I approve. The decisive words in that sentence are content pillars: every draft runs against the same structure described above in this post. I set the pillars myself, because only I know which questions I get asked. Everything after that is hers, and her suggestions do not come from a hunch: behind her rulebooks sit 15,536 analysed Instagram hooks.
What an adjustment looks like is shown by a case from my own account that sits on her page: reels reached a median of 485 accounts, carousels 57, and carousels then repeatedly fell below 200 reach with zero saves. The response was not another round of tweaks to the old format but a clean cut: production shifted its focus to reels, and the carousel came back in a reworked form.
The limit is firm: Conny invents no personal facts about me. No anecdote, no number, no client detail. If something is missing from my note, the spot stays open and I fill it in thirty seconds. She drafts, I approve. How to set up such a role with a rulebook instead of a pile of prompts is in Hiring an AI Employee: The Process, the craft behind it in CLAUDE.md: Claude Code's Memory and the basics in What Is an AI Employee?.
Frequently asked questions
How do I create an Instagram content plan?
In four steps: write down your target person and one countable target action, derive three to five content pillars from the questions and cases in your own work, calculate your cadence from your measured production time and round down, then set a fixed production day. Ideas come last and only get sorted in.
How often should I post on Instagram?
As often as you manage in a normal week, not in a good one. Work out your production time per post, divide the minutes you genuinely have left by that figure and round down. A cadence you keep gives you posts and comparable numbers over time. A cadence that is too high gives you neither.
How many content pillars do I need?
Three to five is the range that has proven itself for me, not a specification from the platform. Fewer feels monotonous, more means no pillar gets enough posts for viewers to recognise a pattern. The test: could you write twenty posts from one pillar without doing research? If not, it is a topic.
Can I schedule Instagram posts in advance?
Yes, with a professional account. Meta offers a planner in Business Suite for this, and there are third party services alongside it. Meta's developer documentation states an API limit of 100 published posts per account within a 24-hour moving period, which is irrelevant for a one-person channel. The tool is the smallest decision in the whole plan.
Should I plan stories in advance too?
Not the content, but the formats. Stories live on what happens today, and four weeks of lead time takes exactly that away. Plan recurring occasions: the start of the week, the question box on Thursday, a look at the work in progress. Then you know when a story is due without fixing what goes in it.
How do I know my plan is not working?
By two signs. First: you postpone the production day for the second time in a row. Then the cadence is too high, not your discipline too low. Second: your target action does not move for months although reach is rising. Then you are not reaching the right people, and the pillars need reviewing.
What to do next
Start with the measurement, not with the planning. Take the next post you are making anyway and time every step. That number tells you which cadence is even possible for you. After that write down your pillars, using the twenty-post test, and put a fixed production day in the calendar. If you then want to stop deciding on every single post what fits you, you need a rulebook instead of an idea list. How I set that up, what Conny's template looks like and who else is working on the same pillars right now is in my community.
