The advice to “be everywhere” is how solo founders and small teams end up mediocre on nine platforms. Real accounts that work are usually deep on two or three and mechanically present on the rest. The hard part is not accepting that. It is working out which two or three, without spending a year finding out.
This is a decision you can make with evidence in about six weeks. The method below is deliberately unglamorous: cast a wide net cheaply, measure the thing that pays, then cut.
Stop asking which platform is best
There is no answer to that question, and every article claiming one is describing its author’s business. The right platform is a function of three things about you: where your buyers already are, what format you can sustain, and how quickly you need the return.
If you sell developer tooling, LinkedIn will feel like shouting into a boardroom and Mastodon or a niche subreddit will feel like a conversation. If you sell a B2B service to operations managers, LinkedIn is where the buyer is and TikTok is a hobby. If you sell a physical product with visual appeal, Pinterest and Instagram do work that no text platform can replicate. None of this is a ranking. It is a match.
The second filter, format sustainability, is the one people skip and then quit over. If you hate being on camera, a TikTok-first strategy will die in five weeks regardless of its theoretical upside. Choose a platform whose native format is something you would still do on a bad week.
Content half-life changes the maths completely
Platforms differ enormously in how long a post keeps working, and that difference matters more than reach.
- X and Threads: effectively finished within hours. High volume, fast feedback, no back catalogue value.
- LinkedIn and Facebook: a day or two of real distribution, occasionally longer if comments keep it alive.
- Instagram and TikTok: days, with an unpredictable chance of a much longer resurfacing.
- Pinterest: months. Pins are saved and re-found through search.
- YouTube and blog posts: years, because both are indexed and searched.
A platform with a long half-life is an investment; a platform with a short one is a job. Neither is wrong, but you should know which you are signing up for. Someone with limited hours per week often gets more compounding return from one indexed YouTube Short or one Pinterest pin than from twenty posts that expire the same evening. Someone who needs conversations and fast feedback should be on the short half-life platforms and should not feel bad about it.
The six-week test
Here is the actual procedure. It is cheap because the first phase deliberately refuses to optimise.
Weeks one to four: cross-post everywhere at low effort. Publish the same core ideas across every platform you could plausibly be on, with light per-platform adaptation and no agonising. The point is not to do great work on each. The point is to generate a real sample of your own data instead of reading someone else’s. This is exactly the phase a cross-posting tool exists for, because doing it manually across ten networks is what makes people give up in week two. seenpaid publishes one post to 21 networks and tracks each one separately.
Weeks five and six: stop adding and start reading. For every platform, gather three numbers. How much time did it cost. How many tracked clicks did it send. How much revenue can be attributed to it. Time is the number people never record and it is usually the decisive one.
What being everywhere badly actually costs
People underestimate the price of spreading thin because the visible cost is only the posting. The real costs are the ones that do not appear on a calendar. Every extra platform is a set of format rules to remember, an inbox of replies you are ignoring, a set of norms you are half-following, and one more place where a stale profile is visible to someone evaluating you.
There is also a quality tax. Attention divided nine ways produces posts that are technically present and clearly phoned in on all of them. Audiences read that instantly. A dormant account with four posts from eight months ago says more about your business than no account would.
The distinction that resolves this is between platforms you work and platforms you cover. Worked platforms get native formats, replies, and real thought. Covered platforms get an automated copy of what you already made, a filled-in profile, and nothing else. Covering is cheap and honest. Half-working nine platforms is expensive and reads as neglect.
Judge on revenue per hour, not followers
Rank the platforms by revenue divided by hours spent. This single ratio resolves most arguments. A platform that produced modest revenue for two hours of work beats one that produced slightly more for fifteen hours, and follower counts will tell you the opposite.
Be careful with early zeros. Some channels genuinely pay late, especially the indexed ones, so a Pinterest or YouTube zero at six weeks is not proof of failure the way an X zero is. Give long half-life platforms a longer window before judging, and be explicit with yourself about which category each one is in.
Also watch conversion rate separately from volume. A channel sending two hundred visitors that convert well is more valuable than one sending ten thousand that do not, and it will look worse in every dashboard that reports reach. This is the single most common reason people abandon a channel that was working.
Then cut, and automate what is left
Keep the top two or three and go deep: native formats, real replies, platform-specific work, proper effort. This is where quality actually pays, because concentrated effort on a matched audience compounds.
Do not delete the rest. Keep them on automated cross-posting at near-zero marginal cost. The upside is real: search visibility for your brand name, a placeholder claim on your handle, and the occasional surprise where a channel you had written off starts producing a year later because the platform changed or your niche moved there. The cost is a few seconds per post if it is automated, and hours per week if it is not, which is the whole argument for the tooling.
Re-run the measurement every quarter. Platforms change, your product changes, and the answer from March is not automatically the answer in September. The point of doing this with data is that re-deciding costs an afternoon instead of an existential debate. seenpaid produces the per-platform revenue number that makes the sort possible, by connecting read-only to your Stripe account and matching payments back to the posts that preceded them.