Getting started
Which social platforms should you actually be on?
8 min readPublished
Which social media platforms should my company be on?
Two or three, chosen because your buyers are already there and you can sustain them. Platform count is the most over-decided variable in marketing and the least correlated with results: an account posted to badly is worse than no account, because it is public evidence that nobody is home. Pick where your buyers already spend attention, then earn the right to add a third.
Key takeaways
- Fewer platforms, done natively, beats more platforms done mechanically. Coverage is not a strategy.
- An abandoned account is worse than no account: it is public, dated evidence that nobody is home.
- Pick on evidence about your buyers, not on platform size. The biggest platform is rarely the one your buyers use for your category.
- Some platforms cannot carry a text post at all. That is a format constraint, not a preference, and it changes what committing costs.
- Adding a platform is a recurring cost forever. Treat it like hiring, not like ticking a box.
The question almost always arrives in the wrong form. Someone asks which platforms we should be on, and the honest answer is a question back: which ones can you keep doing in six months, on a bad week, when the launch slips and the person who ran them is on leave?
Platform count is the most over-decided variable in marketing and among the least correlated with results. It is easy to decide in a meeting, easy to feel productive about, and almost entirely disconnected from whether anything gets read.
An empty account is worse than no account
A dormant profile actively costs you, which is why breadth is not free. Anyone evaluating you will find it, see the date on the last post, and draw a conclusion about whether the company is healthy. That is a worse first impression than having no presence at all, because absence is neutral and abandonment is evidence.
The asymmetry matters when you are deciding. The upside of an extra platform is some additional reach, discounted by how well you can serve it. The downside is a permanent, public, timestamped artifact of giving up. Those are not symmetrical bets, and most platform-count decisions are made as though they were.
Choose on evidence, not on platform size
Pick where your specific buyers already spend attention, which is frequently not the largest platform. Audience size is a denominator, not a signal: what matters is whether the people who might buy your thing are there in a mode where they would consider it.
Where evidence exists, use it instead of guessing. For a software audience, the 2025 Stack Overflow Developer Survey of 49,019 developers reports the community platforms they actually participate in: Stack Overflow at 84.2%, GitHub at 66.9%, YouTube at 60.5%, Reddit at 53.7% and Discord at 38.9%.
Read that list carefully, because it does not describe a posting schedule. Two of the top four are places you answer questions rather than broadcast, and several of the platforms teams reflexively open accounts on do not appear at all. The same survey found technical documentation is the most-used learning resource overall, at 67.8%, which suggests the highest-return publishing for a technical product may not be social posts in the first place.
- 1
Your own customers told you
Ask in onboarding or a sales call where they heard about tools like yours. Small samples beat large surveys about somebody else's audience.
- 2
Your own attribution says so
A platform already sending clicks that convert. The only evidence here that is specific to you.
- 3
Survey data about your audience type
Real research on the segment you sell to. Useful for ruling options in and out, not for predicting your results.
- 4
A competitor is visibly active there
Weak. It tells you where they chose to spend, not whether it worked, and their audience may not be yours.
- 5
The platform is large
Nearly worthless as a reason. Total users is a denominator; your buyers being there in a buying mindset is the thing that matters.
Some platforms cannot carry your post at all
A few platforms have no text-only post, which is a format constraint rather than a preference. On the visual and video networks a post is an image or a video by definition, so committing to one is committing to produce creative on a schedule, not just to write. That is a different capacity question and it belongs in the decision rather than after it.
This is where breadth quietly gets expensive. A team that can comfortably write three posts a week may have no way to produce three videos a week, and discovering that after opening the account is how a profile becomes dormant. If you cannot name who is making the creative, the platform is not a real option yet.
| What you commit to | What it actually requires | Fails when |
|---|---|---|
| A text feed | Someone who can write in that room's register, weekly | The writing gets delegated to whoever is free |
| A visual or video feed | Recurring creative production, not just copy | Nobody owns making the asset |
| A discussion community | Participating far more than you post | You arrive only on launch day |
| A chat community | Answering people, in something close to real time | It becomes an announcements channel |
How to tell you picked wrong
Judge a platform on whether it produced anything you can bank, not on whether the numbers moved. Reach and engagement will always show something, which is exactly why they are poor evidence for a keep-or-cut decision: a platform can look alive for a year while sending nobody who ever becomes a customer.
Give it a real trial, then decide honestly. A quarter of consistent posting is roughly the floor for a fair read, and less than that mostly measures whether you kept it up. At the end, the question is whether this platform sent anyone who converted, and whether it is worth what it cost to run.
Cutting is a legitimate outcome and should be treated as a success rather than a retreat. If you close an account, say so and point at where you are instead, so the artifact you leave behind is a redirect and not a ghost town.
A selection you can defend
Start with two and make the third earn its place. The sequence below is deliberately slow, because nearly every platform-sprawl problem starts with committing to more surfaces than anyone had agreed to staff.
- 1
Ask ten customers where they look
Not a survey, a question in calls and onboarding. Ten real answers about your product beat any general statistic about your category.
- 2
Shortlist on evidence, then cut to two
Rank candidates by the ladder above. Take the top two. The instinct to keep a third for coverage is the instinct that creates dormant accounts.
- 3
Check you can actually make the format
For each one, name the person producing the content and the format it needs. If a platform requires video and nobody makes video, it is not a candidate yet.
- 4
Run a quarter without adding any
Hold the line long enough to get a fair read. Adding a platform mid-trial guarantees you learn nothing about either.
- 5
Review on conversions, then add or cut one
Ask what each platform sent you that you could bank. Add a third only if the first two are genuinely healthy, and close anything you cannot staff.
How many social platforms should a small team be on?
Two, in most cases, with a third only once those two are consistently healthy. The constraint is not reach, it is sustained native production: each platform needs someone who can write in its register every week and answer the people who reply. Teams that pick more usually end up posting mechanically everywhere, which performs worse than posting well in two places and leaves dormant profiles behind.
Should I claim my brand name on platforms I will not use?
Claiming a handle defensively is reasonable, but leaving it as an empty or abandoned profile is not. Either keep it genuinely blank with a bio pointing to where you actually are, or do not claim it. The thing that damages you is a profile with three posts from two years ago, because that reads as a company that quit rather than one that chose.
Is it worth being on a platform just for SEO or discovery?
Sometimes, but be specific about the mechanism rather than assuming one. A profile that ranks for your brand name gives you another result you control on a search for you, which is genuinely useful. That is a much narrower claim than social activity improving your site's rankings generally, which is not something you should plan around.
How long before I know if a platform is working?
About a quarter of consistent posting, and judge it on conversions rather than engagement. Shorter than that mostly measures whether you kept the schedule, and engagement will show you something encouraging on almost any platform whether or not it is sending you customers. The decision question is whether it produced anything you could bank against what it cost to run.
