The fear runs in both directions. Publish your best thinking for free and you worry you have destroyed the reason to buy. Hold it all back and nobody has any evidence you know what you are talking about. Most people resolve this by hedging: they publish thin content that neither helps anyone nor sells anything.
The way out is to stop thinking about how much to give away and start thinking about what kind of thing free content is. Once the categories are clear, the quantity question mostly answers itself.
Free content earns trust; paid content saves time
Useful free content proves two things at once: that you understand the problem, and that you are competent enough to be worth paying. It cannot do that if it withholds the good part. Deliberately incomplete content, the kind that explains that a thing matters and then stops, does not create desire. It creates the impression that you have nothing to say.
What people pay for is almost never information. It is compression. The template instead of the explanation of what a good template contains. The tool that does the thing instead of the description of how to do the thing. The done version instead of the doable version. You can explain your entire method in public and still sell the implementation of it, because the buyer is short of time, not ideas.
A practical dividing line: give away the what and the why in full, and charge for the how, done for them. Explain the strategy, the reasoning, the mistakes, the mental model. Charge for the artefact that removes the work.
Where the line actually falls, by product type
The line moves depending on what you sell, and it is worth being explicit about it rather than deciding post by post.
- Selling software: give away the entire methodology. The product is the labour saved, and people who implement it manually usually come back once they feel the effort.
- Selling templates or assets: give away the principles and a genuinely usable sample. The value is in breadth and polish, not secrecy.
- Selling a course: give away the curriculum and the core insights. What people pay for is sequence, accountability and feedback.
- Selling services: give away as much thinking as you like. Nobody who can afford you was going to do it themselves.
- Selling a community or membership: the free content is the audition. The paid part is access and other members, which cannot be published anyway.
Notice that in none of these is the answer "hold back the good stuff." The paid product is a different category of thing, not a longer version of the free one.
The cannibalisation worry is usually backwards
The people who consume all your free content and never buy were rarely going to buy. They are doing it themselves, and they are also the people most likely to recommend you to someone who will not. Treating them as lost revenue leads to defensive, worse content, which costs you the buyers as well.
The more common failure is the opposite one. Content so promotional that it never demonstrates competence produces an audience that does not believe you can solve anything. You cannot ask for a sale you have not earned the right to ask for, and the earning happens in public, for free, over months.
There is a real version of the worry, though: giving away a substitute for the product rather than a proof of it. A free spreadsheet that does everything your paid tool does is not generous, it is a competing product you built yourself. The test is whether the free thing removes the need or demonstrates the need.
Ask for something in exchange for the best free work
Not everything free has to be public. The middle layer, work that costs something to produce and delivers real value, works well in exchange for an email address. This is not a trick. It is a fair trade, and it converts a passive reader into someone you can reach without an algorithm’s permission.
Keep the gate proportionate. A one-page checklist behind a form annoys people. A substantial teardown, a working template or a small tool is worth an email, and the people who trade for it are the same people who buy later.
Let revenue tune the ratio, not a rule
Every guide on this topic offers a ratio. Four fifths free, one fifth paid. Nine to one. The ratios are made up, and they cannot know anything about your audience. The honest answer is that the right mix is a measurement, and you can take it.
Tag your posts by type before you publish, so that later you can group them: teaching posts, story posts, product posts, direct offers. Then look at revenue attributed to each group over a full conversion window. What usually shows up is that teaching posts produce a large share of the eventual revenue while producing very few immediate clicks, and product posts convert the traffic that teaching posts warmed up. Judging either category alone makes both look worse than they are.
seenpaid exists for this measurement: it schedules across the networks you publish on and connects to Stripe read-only, so each payment can be traced back to the post that started the chain rather than only the last one before checkout. Once you can see which free posts precede paid conversions, the balance question stops being philosophical.
A test for any individual piece
When you are unsure whether something belongs in the free pile, ask what a reader can do after consuming it. If they can now understand the problem better, decide whether it applies to them, or try a first version themselves, it is free content and it is doing its job. If they can only benefit by spending several hours executing something you could hand them finished, that finished thing is the product and the explanation is still free.
The second question is whether publishing it makes the paid product less necessary or more obviously necessary. Explaining, in detail, everything that has to go right for a process to work usually makes the shortcut more attractive, not less. People underestimate how often a thorough free explanation is the strongest sales argument you have, precisely because it shows the size of the job.
A third, more practical check: would you be annoyed to see a competitor publish this? If the answer is no, it is probably too thin to be worth publishing at all. Free content that costs you nothing to give away usually returns nothing either.
A default worth starting from
If you have no data yet, start here and correct later. Publish generously and consistently about the problem, in full, with nothing held back. Once a week, show the product doing something specific for someone specific. Once or twice a month, ask plainly for the sale. Keep one substantial free asset behind an email form.
Then run that for a quarter and look at what the revenue says. The most common correction is not that you gave away too much. It is that you asked too rarely.
And keep the measurement running while you correct. seenpaid schedules across the networks you publish on and attributes Stripe revenue back to individual posts, which is what turns the free-versus-paid argument from a matter of taste into a question with an answer.