A thread is not a blog post chopped into pieces. It's a retention machine. Every tweet has one job: earn the tap to the next one.
Most threads die because the writer optimized for sounding smart instead of being read. The fix is a system — rules you follow every single time so the structure is automatic and your energy goes into the ideas.
The three failure modes
- Weak hook. Nobody opened it. The other 12 tweets never mattered.
- Value gaps. Tweet 5 was filler. Readers bailed there.
- No destination. They finished... and nothing happened. No follow, no reply, no click.
This course fixes all three: hooks (Lesson 3), value density (Lesson 4), and the split CTA (Lesson 5).
Structure is the part you should never improvise. Lock it in so it's muscle memory:
- Tweet 1: starts with 🧵 + your hook. No number. Numbering the hook signals "homework" before anyone is invested.
- Tweets 2 onward: numbered 2/, 3/, 4/... Numbers create progress momentum — readers finish things they can measure.
- Total length: 10–13 tweets. Under 10 feels thin; over 13, completion rates collapse.
- Every tweet must deliver value or build intrigue. If a tweet does neither, delete it. The thread gets stronger.
Formatting inside each tweet
Short sentences. Line breaks between thoughts. One idea per visual block. Emojis for energy and signposting, not decoration — one or two per tweet max.
The hook is 80% of the thread's performance. These five formats cover almost every winning thread:
- Controversy: "[Common belief]. They're wrong. Here's what actually happens 👇"
- Build-in-Public: "I built [X] in [time] with [tools]. Here's everything I learned..."
- Threat + Solution: "[Big platform] is doing [scary thing]. Here's how to protect yourself."
- Resource Drop: "[Number] free tools that replace $[cost]/month..."
- Zero-to-Something: "I built [impressive thing] with zero [requirement]. Full story 🧵"
What they share: a specific claim, a tension (belief vs. reality, threat vs. safety, cost vs. free), and an open loop the reader can only close by reading on.
🧪 Hook Lab — build one right now
Pick a format, fill the blanks, watch your hook assemble itself.
One idea per tweet
Each numbered tweet should teach exactly one thing: one tool, one step, one mistake, one stat. If you need a second sentence to explain the first, fine. If you need a second idea, that's the next tweet.
Real numbers beat adjectives
"Way cheaper" is noise. "$0 vs $20/month" is proof. Specifics build credibility and get screenshotted — and screenshots are free distribution.
Voice rules
- First-person builder perspective. "I ran into this" beats "users may encounter."
- Casual but technically credible. Plain language, correct details.
- No hype language. If you have to say "game-changer," the content isn't carrying its weight. Honest copy can still be punchy — and it compounds, because readers learn your claims check out.
- Never fabricate specifics. Fake numbers get caught, and one caught lie kills the whole account's credibility. If you don't have the number, restructure the tweet so you don't need it.
The educational arc
Sequence the middle tweets like a mini-course: problem → why it happens → the fix → proof → how to start. Readers should finish feeling capable, not impressed. Capable readers follow you. Impressed readers scroll on.
Most writers cram links into their last content tweet. That kills it — the algorithm and the reader both treat link-stuffed tweets as ads. The fix is the Split CTA: two separate tweets with two separate jobs.
Tweet A — the last content tweet
- Strong CTA ("Save this," "Try it tonight")
- A reply-baiting question — specific, easy to answer, slightly opinionated ("What's the one tool you'd never pay for again?")
- A follow ask + 2–3 relevant hashtags
- No links. This tweet's job is engagement.
Tweet B — the promo tweet
- A standardized, reusable block: your community link, your tools, your referral links, your hashtags
- Write it once, refine it over time, append it to every thread
- Because it's separate, the content thread stays clean and the promo is opt-in — readers who got value scroll one more tweet to find where to get more
One honesty rule for promo blocks: only promote things you actually use, and describe them accurately. The promo tweet inherits all the trust the thread just built — don't spend it on something you wouldn't vouch for.
When to post full threads
- Best days: Tuesday and Thursday — mid-week attention without Monday chaos or Friday checkout.
- Best windows: 8–11 AM (commute + coffee scrolling) or 6–9 PM (wind-down scrolling), in your audience's dominant timezone.
After posting
- Reply to every early comment — first-hour engagement decides reach.
- Your reply-bait question should make this easy. If nobody answers it, the question was too generic.
Iterate like a builder
Treat every thread as an experiment. Track which hook format won, where readers dropped off (impressions per tweet tell you), and which questions got replies. Double down on what works; the system above is the result of exactly that loop.
Reuse your machine
Once your rules are stable, encode them as a prompt for an AI assistant: identity, voice, structure rules, hook formats, output format, promo block. Then generating a thread becomes "give it a topic" — and your energy goes into picking great topics and engaging with replies, where it actually compounds.
1. What goes in the very first tweet of a thread?
2. Ideal thread length?
3. Why split the CTA into two tweets?
4. A tweet in your draft is neither valuable nor intriguing. You should:
5. You don't have a real number for a claim. You should: