Most founders we onboard have the same hidden problem. It's not that they're out of ideas. It's that the idea shows up at the wrong time — in the shower, mid-call, scrolling at 11pm — and by the time they sit down to write, it's gone. The blank page wins again.
A content swipe file fixes the timing problem. It's the single most underrated system in a founder's content stack, and almost nobody runs one on purpose.
Let's be precise about what it is, because it gets confused with two other systems we've written about.
What a swipe file is — and what it isn't
A swipe file is a library of external raw material you've reacted to: posts, hooks, structures, screenshots, stats, contrarian takes, and turns of phrase that made you stop scrolling. It's not your work. It's the stuff that provoked your work.
That distinction matters, because we run three different capture systems with clients and they are not the same:
- A hook bank is your own first lines — the openers you've written or want to write.
- A template library is your own reusable post skeletons — structures you've reverse-engineered from your winners.
- A swipe file is other people's raw material — the external spark that gives you something to react to in the first place.
The hook bank and template library solve "how do I write this." The swipe file solves the harder, earlier problem: "what do I even have to say today." It's the input side. Without it, the other two systems are well-organized empty rooms.
The reason this works is the same reason ghostwriting works at all. The bottleneck in founder content is never the writing — it's the raw material. A swipe file is a standing inventory of raw material you can pull from on a dead week, instead of trying to generate insight from a cold start.
What actually goes in it
A swipe file fails when it becomes a graveyard of "good posts" you'll never look at again. The fix is to capture with intent to reuse, not intent to admire. We tell clients to only save something if they can answer: what would I do with this?
Five things earn a slot:
1. A reaction you had. You saw a take and thought "that's wrong" or "that's only half true." That disagreement is a contrarian post waiting to happen. Save the post AND your one-line reaction — the reaction is the actual asset.
2. A structure that worked on you. Not the topic — the shape. A post that opened with a number, or used a two-column comparison, or buried the lesson in a story. You're not stealing the content; you're noting the container.
3. A stat or benchmark. Industry numbers are gold for operators because they let you say "here's the benchmark, here's where most brands actually land." Screenshot it with the source.
4. A phrase or framing. "Merchandising, not design." "You're renting the placement." Language that compresses a whole idea into four words. You won't reuse the exact words — you'll reuse the move.
5. A question someone asked. Comments, DMs, Reddit threads, sales calls. Every real question is a post that's already validated by demand. Most of our highest-engagement client posts started as a question someone else asked out loud.
Notice what's not on the list: polished posts you admire with no angle attached. Admiration isn't a slot. Reaction is.
The 30-second capture rule
The whole system lives or dies on friction. If capturing takes more than 30 seconds, you won't do it, and the idea evaporates — same as the shower thought.
So the rule we give every client: one inbox, one tap. Pick a single destination — a note pinned to your home screen, a Slack channel you DM yourself, a specific Apple/Google note. Not five tools. One. When something sparks a reaction, you do exactly two things:
- Capture the thing (link, screenshot, or one line of text).
- Add your one-line reaction — why you saved it.
That's it. No tagging in the moment, no folders, no "I'll organize this later." The reaction line is non-negotiable because future-you won't remember why present-you cared. A naked screenshot six weeks later is useless. "Saved because this is exactly backwards for $1M+ brands" is a post.
The forwarded DM counts too. When a client sends us a screenshot with "this is dumb, here's why" — that's a swipe-file entry and a finished angle in one move.
Turning the file into posts: the weekly pull
Capture is half the system. The other half is the pull — and this is where most founders' swipe files die as digital hoarding.
Once a week, during your content planning block, open the file and do a 10-minute pass. You're looking for three entries that pair with something you actually know. The swipe-file entry is the spark; your operator experience is the fuel. The post is the combination.
The pattern looks like this:
- External spark: a stat or take you saved.
- Your angle: "here's what that misses / confirms / costs in practice."
- Your proof: a number, a client situation, a decision you made.
A swipe-file entry on its own is commentary anyone could write. Paired with your specific experience, it's authority only you can publish. That pairing is the entire game. You're not reposting the spark — you're using it to unlock something you already knew but hadn't found the doorway into.
A healthy file produces 2–4 usable angles per weekly pull. That's most of a posting week, sourced from reactions you already had, captured in 30-second increments, with zero blank-page time.
How to keep it from rotting
Three maintenance rules keep a swipe file alive instead of becoming a junk drawer:
Prune on pull. When you do your weekly pass, delete anything you scroll past twice without a reaction. If it didn't spark a post in two passes, it never will. A swipe file should feel like a sharp deck of cards, not an archive.
Tag lightly, and only on pull. Don't tag at capture — that's friction. When you pull, loosely sort entries by your content pillars so you can find "the contrarian one" or "the data one" fast. Heavy taxonomy is procrastination disguised as organization.
Keep it small on purpose. A swipe file with 400 entries is a place ideas go to die. We tell clients to aim for a rolling 30–50 live entries. Past that, you're not curating, you're collecting — and collecting is a hobby, not a system.
FAQ
How is this different from just saving posts on LinkedIn? Saved posts are a black hole — no reaction attached, no pull ritual, buried in an app you don't open with intent. A swipe file forces the reaction line and lives in a destination you actually check weekly. The mechanism is the reaction, not the bookmark.
Won't using other people's posts make my content derivative? Only if you reuse the content. You're reusing the spark — a structure, a stat, a question — and pairing it with your own operator proof. The output is more original, not less, because it's grounded in your specific experience instead of a cold-start guess at what to write.
How big should it get before I start pulling from it? Start pulling in week one. The system isn't "build a giant library, then mine it." It's "capture three things this week, pull one into a post next week." A swipe file is a flywheel, not a warehouse.
Should my ghostwriter run this, or should I? Capture is yours — the reaction has to be authentically yours, in the moment. The pull and the writing can be shared. The best client setups have the founder dumping reactions into one inbox all week, and the writer running the weekly pull to turn three of them into drafts.
The founders who never run dry aren't more creative. They've just stopped trying to manufacture insight on demand. They capture reactions as they happen and pull from inventory when it's time to write.
If you're tired of staring at a blank post twice a week while good ideas die in the shower, this is the system that fixes it. We build the capture-and-pull workflow into every client engagement — because a founder with a full swipe file is a founder who never misses a posting week. Want us to set yours up and run the pulls?