The thing that does not compound
A post that does well brings a burst of followers. Another good post brings another burst. Neither makes the next post easier, and this is the trap: an account can publish well for a year and stay flat, because every post starts from zero.
What separates a growing account from a busy one is whether anything carries over between posts. Three things do.
A reason to follow rather than to like. Someone enjoying a post is not the same as someone expecting the next one. The follow happens when a reader can predict what they will get — which requires that your posts be about something, consistently, rather than being individually good about different things.
A surface that keeps working after the post stops. Search, recommendations and other people's links keep sending people to a post long after the feed has moved on. A post written to be findable later is worth several written only for the day it ships.
Being remembered. People follow on the second or third encounter far more often than the first. That is a function of frequency and recognisability, which is where consistency stops being a virtue and starts being a growth mechanism.
Everything below is about making those three carry over.
Be specific enough to be followable
The most common growth problem is not quality. It is that the account has no answer to "what is this about?"
An account that posts about a specific thing, from a specific angle, is followable because the reader can predict the next post. An account that posts good content about whatever came up that week is enjoyable and not followable — there is nothing to anticipate.
Specificity feels like it costs reach, and in the short term it does. The trade is real and it is worth making: a narrow account converts a far higher share of the people who see it, and it accumulates a reputation for one thing rather than a vague impression of competence.
Two practical tests:
- Can you finish the sentence "I post about ___, for ___" without using the word "and" more than once?
- If someone read three of your posts at random, would they be able to guess the fourth?
If either answer is no, the growth problem is positioning, and no amount of extra posting will fix it.
Write for the second reader, not the first
Most of a post's lifetime audience arrives after the day it was published — through search, through a recommendation surface, or through someone linking to it months later. Very little of what gets written is built for those people.
The difference is mostly structural. A post that opens with "following on from yesterday" or "as promised" is unreadable to anyone who arrives cold. A post that states what it is about in the first line works for both the feed and everyone who finds it later.
This is also why the "evergreen versus timely" framing is misleading. The question is not whether the topic ages; it is whether the post is comprehensible to someone with no context. A post about a current event can be written so that it still makes sense in a year. Most are not, for no reason other than habit.
Frequency is a growth input, not just a discipline
Being remembered requires repetition, and repetition requires showing up. This is the least interesting reason consistency matters and probably the most important one.
The mechanism is unglamorous: someone sees a post and does not follow. They see another two weeks later and half-recognise the name. The third time they follow. An account posting twice a week reaches that third encounter roughly three times faster than one posting twice a month — and the account posting twice a month is not three times worse, it is invisible.
This is the argument for a cadence you can actually hold. An account that posts daily for six weeks and then stops has generated a burst and no memory. One that posts twice a week for a year has been in front of the same people about a hundred times.
What to do about engagement
Replying to comments and taking part in other people's threads is the highest-leverage growth activity that most accounts skip, and it is worth being precise about why it works.
It is not that the algorithm rewards it, though it sometimes does. It is that a reply puts you in front of a reader who has already demonstrated interest in the exact topic you post about, in a context where you are a person rather than a broadcast. The conversion from that encounter is far higher than from a post impression.
The version that does not work is volume commenting — leaving short agreeable remarks on large accounts. It is visible, it reads as marketing, and it converts close to nothing. The version that works is answering a specific question in a way that could only have come from someone who knows the subject.
Budget it like any other task. Twenty minutes, two or three times a week, spent on genuine replies, will usually do more than a fourth weekly post.
Stop measuring followers
Follower count is a poor steering metric. It is slow, it lags the work by weeks, and it is inflated by people who followed once and never looked again.
Better things to watch, roughly in order of usefulness:
- Repeat engagement — the same names appearing across different posts. This is the clearest early sign an audience is forming rather than passing through.
- Saves and shares over likes. A like is a reaction; a save is a judgement that the post is worth returning to, and a share is a reader spending their own reputation on it.
- Where new followers arrive from. If growth comes entirely from one viral post, it will stop when that post does.
- Reach among non-followers, where a network reports it — this is the findability signal.
A useful monthly question: did the number of people who engage with more than one post go up? If yes, the account is compounding even when the follower graph looks flat.
What does not work, and why it keeps being recommended
Two tactics come up constantly and are worth dismissing explicitly.
Follow-for-follow and engagement pods. These inflate the number and degrade everything the number is a proxy for. An audience that does not care about the subject reduces the rate at which posts are engaged with, which is what recommendation surfaces actually measure. The count goes up and the reach goes down.
Posting the same text everywhere. It is the single most recognisable signal of an account that is broadcasting rather than participating, and each network's conventions differ enough that identical text reads as wrong on at least two of them. Adapting a post for each network takes a couple of minutes and is the difference between looking native and looking syndicated.
A realistic timeline
Growth is slow in a specific pattern that is worth expecting, because the shape is what makes people quit.
The first stretch produces almost nothing measurable. This is not failure; it is the period in which there is no back catalogue for anyone to find and nobody has seen you often enough to remember. Then growth arrives unevenly — a post does unusually well, the account gains a step, and then it flattens again at a higher level. Compounding shows up as a staircase, not a curve.
The practical implication is that the decision to continue has to be made on inputs rather than outputs for longer than feels comfortable. Did you post at your cadence? Did you reply to people? Is the positioning still specific? Those are answerable every week. The follower graph is not.
Sources consulted: LinkedIn Help (linkedin.com/help) · Meta Business Help Centre (facebook.com/business/help) · TikTok Creator Portal (tiktok.com/creators) · YouTube Help (support.google.com/youtube)
Common questions
How long does it take to grow a social media following?
Expect the first stretch to produce very little measurable growth, because there is no back catalogue to find and nobody has encountered you often enough to remember. Growth then tends to arrive as a staircase — a post does unusually well, the account steps up, then flattens at the higher level. Judge progress on inputs like cadence and repeat engagement rather than on the follower graph.
Do I need to post every day to grow?
No, and daily posting without enough material to support it usually hurts. What matters for being remembered is the number of encounters with the same people over time, which a sustainable twice-weekly cadence delivers better than a daily burst that stops after six weeks. Choose the cadence your supply of genuine material supports.
Should I post the same content on every network?
Publish the same idea, not the same text. Each network has different length conventions, formatting and audience expectations, and identical text is the clearest signal that an account is broadcasting rather than participating. Adapting a post takes a couple of minutes and materially changes how it reads.
Are hashtags still worth using?
Their value varies by network and has generally declined where recommendation systems infer topic from the content itself. Use a small number of genuinely descriptive tags where a network still uses them for discovery, and do not pad. Large blocks of loosely related tags mostly signal low-quality content.
Why did my follower count stop growing?
A plateau after early growth usually means nothing carries over between posts — each one starts from zero. The three things that carry over are specific positioning that makes you followable, posts written so they still make sense to someone arriving cold, and enough frequency to be remembered. A plateau is normally a positioning problem rather than a quality problem.
PostWharf is priced per workspace rather than per connected channel, and every plan carries unlimited channels. See per-brand pricing.