A pottery studio I follow posted daily for a year — trend sounds, five platforms, the full "consistency" gospel — and grew by almost nothing. Last spring the owner cut back to two platforms and three posts a week, and her bookings finally moved. When I asked what changed, her answer was the best social strategy summary I've heard: "I stopped performing for the algorithm and started talking to the eleven people who might actually book a class." A real strategy for 2026 isn't a posting schedule — it's a set of decisions about who, where, what and how much. Here's the set, in deciding order.
Decision 1: Who are the eleven people?
Not "my audience" — the specific human who buys: the bride looking for a hen-party activity, the parent hunting a rainy-Saturday option, the office manager booking team events. Write down two or three of these people, because every later decision (platform, tone, content) is just "would this reach and convince them?" Strategies without this step produce content for an imaginary general public, which is why they feel busy and convert nothing.
Decision 2: Where — one platform properly, two at most
Go where your buyers scroll, not where marketing content tells you to be: visual local businesses live on Instagram and increasingly TikTok; trades and community services convert on Facebook and Google Business Profile posts; B2B services on LinkedIn. The pottery studio's five-platform year spread one person's energy into five mediocre presences; two platforms done properly beat five done thinly, every measured time. Cross-post only with native re-cuts — watermarked reposts get buried and look like what they are.
Decision 3: What — the three-bucket content mix
Sustainable small-business content reduces to three buckets: useful (answers to the questions you're asked weekly), human (the process, the mess, the finished piece, the dog in the studio), and selling (openings, offers, booking links) — in roughly a 2:2:1 rhythm. Feeds that are all selling get muted; feeds that are all vibes never convert. AI's proper role is production inside the buckets: a caption tool for drafts, a hashtag generator for searched-and-vetted tags, an assistant to turn one workshop photo into a week of angles. The voice pass stays yours — followers can tell, and the ones who can't aren't buyers.
Decision 4: How much — the sustainable minimum
The schedule that survives a busy month wins, because the algorithm penalty for posting three good things weekly is nothing compared to the human penalty of burning out in week six. The studio's working rhythm: 90 batched minutes on Sunday (five drafts from real inputs — this week's classes, one process video, one answered question), ten minutes daily for replies. Consistency matters, but it's consistency of showing up in DMs and comments — where bookings actually close — not of feeding the feed.
Decision 5: What counts — measure bookings, not applause
Monthly, three numbers: saves/shares (buyers bookmarking), DMs and enquiries (buyers moving), and actual bookings or sales attributed to the platform ("how did you find us?" — just ask). Follower count and likes are the applause metrics; the studio's near-flat follower graph during its best booking quarter settles that argument. AI helps here too — paste a month of post data and ask what your top performers share — but only if the metric it optimises is one connected to money.
Decision 6: What to do when a post flops (and when it takes off)
Flops are data, not verdicts. When a post dies, check which bucket it came from — most businesses discover they flop hardest on polished promotional posts and win on the unpolished human ones, which is uncomfortable and useful to know. When something takes off, resist the urge to chase its format forever; make one direct follow-up ("lots of you asked about X from Tuesday's post") while the interest is real, then return to the rhythm. The studio's best month came from answering, in a plain video, the one question a viral post had raised — not from trying to go viral twice.
The tools list, kept deliberately short
Everything above runs on: a phone camera, the platform's own scheduler (or one free scheduling tool if you must), a caption generator for drafts, a hashtag generator used with in-app checking, an image resizer and compressor for making one photo work everywhere, and a notes file of what worked. Total cost: £0. Every tool beyond this list should have to argue its way in with a number — the pottery studio's entire turnaround used nothing that isn't in this paragraph.
The strategy on one page
Two named buyer types → one or two platforms → 2:2:1 content buckets → 90 Sunday minutes plus daily replies → bookings counted monthly. That's the whole document, and it fits on an index card, which is why it survives contact with a real business week. Trends, tools and algorithms will churn under it; the card barely changes. For the production toolkit that makes the Sunday batch fast, our social media tools guide is the companion piece.
Strategy that survives the algorithm shift — measured, not guessed
The accounts that survived three algorithm shifts did one thing: owned an email list plus one social, not five socials.
| Channel | Own? | Algorithm risk | Use |
|---|---|---|---|
| Email list | Yes | Low | Your direct line |
| YouTube | Medium | Medium | Evergreen search |
| Instagram/TikTok | No | High | Discovery, not vault |
Editor’s note: Post where discovery happens, save where you own. We publish to social, archive to blog — blog is the vault.
This week’s posts
- One ‘how we did it’ with numbers
- One mistake + fix
- One question to your audience (reply to every answer)