Build a table of content ideas in 6 columns
A table of content ideas only works when each row carries a decision. The six columns, the order to fill them in, and the test that catches dead rows.
A table of content ideas works when every row already carries the decisions you would otherwise make on posting day. Six columns do that: the network, the asset you already have, the point, the form, the next step, and the date. A table that only holds topics stalls by week three. Here is how to build one, fill it in the right order, and test the rows before you trust them.
What goes in a table of content ideas
The unit of the table is one post on one network. Not one theme, not one week. That single decision changes everything downstream, because a row that says "customer stories" is a category, and a category cannot be published.
Most templates you can download have three columns: date, pillar, idea. Three columns describe intentions. Nothing in them tells you what to do on Thursday. The six below do, because each one takes a decision off your plate at the moment you are least able to make it.
| Column | What it holds | The decision it removes |
|---|---|---|
| Network | One network, one row | Copying the same words everywhere at the last minute |
| Have | The photo, note, screenshot or figure already on your phone | The shoot you keep postponing |
| Point | One sentence a reader could repeat to someone else | The rewrite on posting day |
| Form | Single image, carousel, or text post | The format hesitation |
| Next step | What the reader is meant to do after reading | The caption that ends on nothing |
| Date | Filled last, and only for rows that are ready | The dated row you were never going to publish |
One network per row is the column people fight hardest. It looks like duplicated work. It is the opposite: the same point does not land the same way in an eighty-word Instagram caption and a two-hundred-word LinkedIn post, and adapting one message to each network is a different job from having the idea. Keep them apart in the table and each row stays small enough to finish.
Fill the column nobody writes first
Start with Have, not with Point.
The usual order is idea first, everything else later. It reads well, and it is why saved ideas never ship. An idea written in the abstract silently assumes an asset you do not have yet. "Behind the scenes of a custom order" assumes photographs taken during a job that finished three weeks ago. The row is not blocked by motivation. It is blocked by a photo nobody took.
So open your camera roll, your notes app and your invoices for the past seven days, and write one line per usable thing. A half-finished piece on the bench. A question a client asked twice. A figure from a quote you sent. Then, and only then, write the point that asset can carry. Where the raw material comes from in the first place is a separate question from what the table does with it.
Can this step be skipped? Only if you already photograph your work every week without deciding to. Almost nobody does.
Run the same row twice. If you get two unrelated posts, the row is a topic, not a brief.
Write rows that survive being run twice
Here is the test that separates a row from a wish, and it takes about a minute.
Take one finished row. Hand it to any AI writing tool and ask for the post. Then open a fresh conversation and ask again from exactly the same row. Compare the two drafts.
If they are about different things, the row is a topic, not a brief. This is not a flaw in the tool. It is the row reporting how much it left undecided. A generator fills every blank with the most average plausible answer, and two runs land on two different averages. The same thing happens to you at seven in the evening, only slower: you reopen the row, the row says "behind the scenes", and you invent a new post on the spot.
The fix is never a better adjective. It is a concrete noun plus one sentence.
- "Behind the scenes" becomes "the forty minutes spent re-cutting the second sleeve on the navy jacket".
- "Client testimonial" becomes "a client asked whether the finish survives a dishwasher, and here is the answer, with the mug".
- "Share our values" becomes "why the workshop closes on Wednesdays".
Run the test again on the rewritten row. The two drafts converge, because there is now only one post that row can produce. A row that passes this test is a row you can hand to someone else, or to something else, and get back what you meant.
Sort by readiness, then attach the dates
Most tables are sorted by date from the first minute. That is why they break the first week something goes wrong: one missed Tuesday and the whole grid is out of step, and the table turns into a record of what you did not do.
Sort on the Have column instead. Rows with an asset go on top. That block is this week. Rows without one go to a second sheet, which is not a calendar but a shot list: things to photograph or write down when the occasion turns up.
Dates go on the top block only, and they go on last. A month prepared in one sitting beats a grid you improvise against every morning, but the order matters. Readiness first, dates second. A date on an unready row is a promise to yourself that you will do the missing work later, at a worse moment.
Mark what shipped, and which column was empty
At the end of the week, two marks per row. Shipped or not shipped. And for the rows that did not ship, which column was empty when you opened it.
Do that for three weeks and one column will be empty far more often than the others. Which one it is tells you where the system leaks. An empty Have means capture is the bottleneck, and the fix is a habit, not a better idea. An empty Next step means you have not decided what a post is for, and that decision belongs upstream of the table. This review is the only part of the method that improves it over time. The six columns themselves never change.
What you end up with is not a fuller idea list. It is a queue where a row and a post are the same object at two different stages, and where the work left is review rather than invention. That is also the point at which the writing becomes something you can delegate, because the decisions were already made in the table.
FAQ
How many rows should a table of content ideas have?
Enough for about two weeks per network you actually post on. Beyond that, rows go stale before their turn: the asset gets used elsewhere, the offer changes, the season moves.
Is a content ideas table the same as a content calendar?
No. A calendar answers when. The table answers what, and with which asset. A calendar built on rows that are still topics gives you dated blanks.
What should I build it in?
Anything that sorts columns. A spreadsheet is enough. The tool matters far less than the six columns and the order you fill them in.
Do I need one table per network?
One table, one row per network. Splitting into separate tables hides duplicated points, and you end up publishing the same sentence twice.
How often should I refill it?
Once a week, in the same sitting where you mark what shipped. Refilling monthly means writing rows from memory, and memory produces topics rather than assets.
