Free content idea generator
Say what your account is about and get six specific post ideas back. Each one names its format, writes its literal opening line, and says why it would work for you, which is the part worth reading first. About five seconds, no account and no credit card.
Free forever No login No credit card
This is as far as we can get without your material.
Paste any one of these and regenerate:
Want these made, not just listed?
Take all six, they are yours. The harder half is doing them: Selzee turns ideas like these into the actual posts, from your own brand and your own customers' words, and keeps going after the first six.
Have them made, freeThat is the free batch for this page. Reload for another set, or Start free and get ideas from your own brand.
A content idea generator answers the question that stops most accounts: what do I post this week. This one is free, runs without a login, and asks about your account rather than your industry, because a generic input can only produce the idea list everybody has already read. Six ideas come back, each with the format, the opening line and the reason it would work here.
It is built for the people who post their own content and keep running dry: small brands, creators, founders, and social managers with a calendar to fill. The median post earns very little engagement on any platform, as Socialinsider's benchmarks across millions of posts show, so a bigger pile of ideas is not the goal. Enough good ideas to keep posting while you find out what your audience responds to is.
Not the industry, the account. "Small-batch coffee roastery that ships subscriptions" is a brief. "Coffee" is a category, and a category gets you the listicle everybody has already read. This is the only field the tool cannot work without, because it is the only one that cannot be inferred from the others.
A good idea usually survives the move between surfaces, so the platform is optional here and the format line tells you which one each idea suits. Pick a platform when the constraint is real, for instance when you know the next three posts have to be Reels, and every format that comes back will be one that platform supports.
These two change the ideas more than anything else, because an idea worth posting usually sits where what you sell meets what your audience is already worried about. Paste anything you have in the notes as well: what worked last month, the question that keeps arriving, the objection you answer on every call.
Each idea comes back as its format, what the post is, the literal first line, and why it would work for this account. That last part is the one worth reading first, because it is the part you can disagree with. Copy any idea in a click. No account, no email, no credit card.
Same idea, different artefact. Every one runs without a login and nothing you make in them costs anything.
The hub for the seven. One idea into posts, captions, carousels, scripts or a week of ideas.
Try it free 📣One idea into three posts, each written for the platform you pick, with its angle named.
Try it free 📸Describe the photo and get five captions at the vibe, length and emoji level you choose.
Try it free ðŸŽA topic into 5, 7 or 10 slides written in reading order, plus the caption to go under them.
Try it free 🎬A Reels, TikTok or Shorts script beat by beat: the line to say and the shot to film beside it.
Try it free 💼One thing you know into three LinkedIn posts that read like a person, not a press release.
Try it freeThe case for generating ideas is not that a model understands your audience better than you do. It is that the week you have nothing planned is the week you post nothing, and six specific ideas with their openings written is the cheapest way to make sure that week does not happen.
Almost nobody stops posting because a post failed. They stop because they sat down on a Monday with nothing to say and did something else instead. Six ideas with their opening lines already written removes that particular Monday.
The unit most people are short of is not one caption, it is a plan for the next few posts. That is why this returns six rather than one, and why each one is specific enough to act on without a follow-up question.
The same point lands very differently as a talking head, a carousel or a single photo with a long caption. Getting the format attached to each idea makes the difference between a list of topics and a list of posts.
Twenty generic prompts are less useful than six ideas that each say why they would work here. The reason is what lets you reject four of them quickly, which is most of what choosing well actually is.
Read the reason before the idea.
Every idea comes back with why it would work for your account. When that reason is thin, the idea is thin, whatever the title promises.
Search for content ideas and you will find lists of a hundred prompts: share a customer story, post a behind the scenes, do a myth versus fact. They are not wrong, they are just not ideas yet. Every one of them still requires you to decide what the story is, which myth, and why anybody would care this week. The work that was supposed to be removed is the work that is left. Drafting with a model is ordinary now, with SurveyMonkey finding 88% of marketers using AI day to day, and the same trap applies: a generic prompt produces the generic list back.
What makes an idea usable is that it is specific to one account and carries a reason attached. Sarah Levinger, Founder of Tether Insights, points at where that specificity comes from: “I wish more brands would use the data they have and start analyzing for emotional insights instead of just transactions.” Once you have an idea, the AI social media post generator writes it up, the carousel generator turns it into slides, and the video script generator turns it into something to film.
Who it is for, what it sells, what makes it different, in one or two sentences. Every vague word you leave in comes back as a vague idea, and every specific one comes back as something you could film this week.
If two posts did far better than the rest, say what they were. The ideas that come back will lean towards that shape instead of spreading evenly across formats you already know do nothing for you.
The third line on every idea is why it would work for this account. When that reason is thin, the idea is thin no matter how good the title sounds. This is the fastest way to cut six ideas down to the two worth making.
The single most effective second run is the same account with a much narrower audience: not "people who like coffee" but "people who bought a grinder and still make it badly". Narrow audiences produce specific ideas, and specific ideas are the only kind anybody engages with.
The first month is easy and the second is where it stops. A list of six that fits your niche is the thing that gets you past it.
You know the material. What you do not have is fifteen minutes on a Monday to decide what this week is about.
Run it per pillar, keep what fits, and the calendar stops being the blank part of the job.
Six ideas with reasons attached is a conversation with a client. A list of topics is a conversation about a list of topics.
Every idea returns in the same four parts, so six ideas are six decisions rather than six topics. This is what the tool above produces, not a wish list.
Reel, carousel, single image or text, named up front, because the format changes what the idea even is.
One sentence describing what is actually in it, not a topic you would still have to invent a post from.
The literal first line or first shot, as it would appear. The hardest part of any post, already drafted.
Why this would work for this account, stated so you can disagree with it. A reason nobody could argue with is not a reason.
Yes. Six ideas per run, no account, no email, no credit card. There is a small cap per page load so the shared model key stays available for everyone, and reloading gives you another batch.
A published list is written for everybody, so it stops at the level of "share a customer story". These are written for the account you described, they name the format, they include the actual opening line, and they say why each one suits you. The specificity is the whole difference.
Deliberately not. Anything built on a trend or a sound dates within weeks and no generator can know what is current today, so an idea that depends on one would be wrong more often than right. Every idea here is meant to work whenever you get round to making it.
No. The prompt assumes one person with a phone, so nothing that comes back needs a studio, a paid tool or anyone else on set.
No. It only uses a number, a name or a quote if it appears in what you gave it. An idea built on an invented statistic is worse than no idea, because you would find out it was wrong after publishing it.
It gives you six ideas at a time, which is a week or two for most accounts. For a month, run it once per content pillar and keep what fits. What it does not do is track what you have already posted, which is exactly the part the product handles.
This is one run from one description you typed. The product works from your own brand: it researches, comes up with concepts, makes the content, then learns from how it performed, so the ideas keep coming and they get better as it sees what worked.
A batch of ideas fixes one week. What it cannot do is remember what you posted last month, notice which of those ideas actually landed, or stop suggesting the shape that has never worked for you, because it only ever sees the description you typed. Selzee works from your own brand instead: research, concepts, the content, then what it learns from performance, running continuously rather than one batch at a time.
The generator is free right now, with no signup and no credit card. When you want ideas that come from your own brand and get better as it learns, start free.
Start free and get ideas from your own brandPrefer to explore on your own? Get started or browse free ecommerce tools.
Keep exploring: the AI social media post generator, the Instagram carousel generator, the video script generator, Instagram post ideas.
