A
AIOS Wiki
read-only · public mirror
Open AIOS
Wikisaysonotesmarketing-researchsayso/notes/marketing-research/predict-fun-marketing-playbook.md

predict.fun — Marketing Strategy Reverse-Engineered

Hand-authored·8 min read·16 sections·Last edited May 12 by initial import·View history

Last updated: 2026-05-07 Subject: predict.fun — the BNB-native prediction market (originally Blast-native, Sep 2024 → Dec 2025) Sources: live site, news.predict.fun blog (10 posts), @predictdotfun X profile, @dingalingts founder profile, Wayback Machine captures (Sep 2024 – Mar 2026) Companion: mcap/notes/marketing-research/sayso-x-marketing-playbook.md

The honest preamble

The single most important fact about predict.fun's marketing: the founder, Dingaling (@dingalingts), is ex-Binance Head of Research and a PancakeSwap co-founder, with 254K Crypto Twitter followers since 2017. Almost every "marketing move" is downstream of his Rolodex and pre-existing audience.

That doesn't make the playbook uncopyable — it changes which parts you should copy. The mechanical/operational patterns travel; the distribution-by-incumbent endgame requires either a founder profile, capital, or a slower substitute.

Timeline (reconstructed)

DateMove
Jul 2024@predictdotfun X handle created — before the product is public
Sep 5, 2024First Wayback capture; site live as "Predict — the Blast-native prediction market". Tagline: "Bet on future outcomes (and earn rewards) across politics, sports, crypto and more." Points + LP rewards + Polymarket-comparison embedded from day 1
Sep–Nov 2024Rides Blast L2 airdrop wave + the 2024 US election prediction-market boom that mints Polymarket. The UI literally renders "Polymarket, provided for your reference" next to each market — pure arbitrage lure
2024–2025YZi Labs (ex-Binance Labs) backs them via the EASY Residency Season 2 incubator
Oct → Dec 2025Migrate / relaunch on BNB Chain
Dec 18, 2025"BREAKING: PREDICT GOES LIVE" — BNB launch. Smart-wallet (no gas, no approvals, email/Google/X sign-in), Predict Points (PP), 8-network deposit routing, perpetual referral: 10% off invitee fees + 10% of inviter fees AND 10% of their PP forever
Mar 4, 2026Acquire Probable — the only other PancakeSwap+YZi-incubated prediction market. ~11 weeks post-launch they're already at $1.5B cumulative volume, 120K users, 3.3M txs
Mar 31, 2026YZi Labs + Susquehanna Crypto strategic round announced. Volume: $1.8B cumulative. SIG seeds liquidity; YZi unlocks Binance distribution
Apr 9, 2026Official prediction-market provider inside the Binance Wallet mobile app. Distribution unlock
Apr–May 2026One-name-per-week BNB-ecosystem partnerships: UMA (resolution), Chainlink (oracle), Venus (yield), Four.meme (memecoin markets), BNB Chain co-marketing (Lunar New Year $11K prize, X Spaces)
May 2026 (now)@predictdotfun: 50.9K followers, follows 105 (484× ratio), 1,131 posts ≈ 1.7 posts/day sustained

The 8 patterns (what they actually did)

1. Audience before product

Handle live Jul 2024, product Sep 2024. 60–90 days of brand presence before anyone could click "buy." Founder's personal 254K-follower account seeded the early audience. Operational version: register the handle and start posting about the problem domain (not the product) the moment you commit to building.

2. Territorial framing in the og-description

Never "a prediction market" — always "the [chain]-native prediction market." Sep 2024 og-title: "Predict | The Blast-native Prediction Market." Dec 2025 launch claim: "soon-to-be top prediction market on BNB Chain." Pick a territory you can plausibly own and claim it from the meta tags out.

3. The "rewards" hook in the og-description itself

"Bet on future outcomes (and earn rewards) across politics, sports, crypto and more."

The parenthetical is the growth motor. Tells airdrop farmers there's something to farm before they click. Crypto-launch SEO 101.

4. Polymarket-comparison gambit

Early Blast UI rendered "Polymarket, provided for your reference" next to each market. They didn't fight the incumbent — they used it as a price feed + arbitrage lure so traders had a reason to bridge over. Find the obvious incumbent in your category and embed their data next to yours.

5. Perpetual referral, not bonus

Dec 2025 relaunch: invitee gets 10% off USDT fees; inviter earns 10% of the invitee's fees AND 10% of their Predict Points forever. KOL-bait. Single-shot bonuses cap at the bonus; this compounds, so one big referral pays the affiliate indefinitely. Highest-ROI design decision on the page.

6. Acquihire your closest rival pre-scale

Probable (PancakeSwap+YZi-incubated, same ecosystem) was the only credible BNB competitor. They bought it. Conversion design:

  • 2× refund of all USDT fees Probable users had paid
  • Probable Points → Predict Points: 1:1 for early users (weeks 1–6), 1:10 for late users (weeks 7–10) — premium for the latecomers, which is the elegant inversion of the obvious move
  • Result: kill the comp + inherit Asia network + earn goodwill

7. Single-narrative cadence

~1.7 X posts/day, mostly "NEW MARKET — [topic]". Plus one partnership-of-the-week: UMA, Chainlink, Venus, Four.meme, Binance Wallet, BNB Chain co-marketing. The cadence is mechanical; each partnership is a fresh excuse for the ecosystem to retweet. One announcement per week sustained for 6 months outperforms five in a quarter.

8. Distribution-by-incumbent, not bottoms-up

The endgame: don't market — be distributed. Become the default integration inside the Binance Wallet app. The 50K-follower X account is downstream of that, not upstream. Your equivalent: figure out whose embedded surface area would carry your product and trade equity / data / co-marketing for placement.

What you can actually copy (if you're not Dingaling)

Most teams don't have a Binance-Research-Head + PancakeSwap-Founder + 254K-CT-follower CEO. Adjusted:

PatternCopy 1:1?Notes
#1 Handle pre-launch✅ YesFree, 60+ days early. No reason not to.
#2 Territorial framing✅ YesPick a smaller territory you can defend. "The X for [niche]" is buyable when "the [chain]-native X" is taken.
#3 Rewards in og-desc⚠️ If applicableIf you have rewards/points, copy. If not, replace with a hard-to-fake quantitative claim ("settles in 4 seconds").
#4 Incumbent comparison✅ YesCheapest credibility move in the book. Embed the incumbent's data right next to yours.
#5 Perpetual referral✅ YesIf unit economics tolerate it. Single highest-ROI design lever.
#6 Acquihire rival❌ Not yetPost-fundraise only. Skip until you have capital.
#7 Weekly partnership cadence✅ YesOperational. Just do it. 26 partnerships in 26 weeks.
#8 Distribution-by-incumbent🔀 Invert itWithout founder Rolodex, become the default integration of an open-source primitive (Chainlink feed source, Farcaster frame, Telegram mini-app) where the host needs you more than vice versa. Slower; still real.

What NOT to copy

  • Don't fake a founder profile. "Stealth from top crypto firm" / "ex-FAANG" without specifics reads as cope. CT/X audiences detect it instantly.
  • Don't pivot chains for marketing. Predict's Blast→BNB pivot was substantive (Blast dying, Binance distribution available). Cosmetic chain-pivots burn trust.
  • Don't open the corporate blog with a partnership. The first post on news.predict.fun is "BREAKING: PREDICT GOES LIVE" — product before partners. Partnerships only land because a product is already shipping.
  • Don't fundraise-tweet before traction. They announced YZi+SIG after $1.5B+ cumulative volume. Fundraise tweets without traction are the most ignored content on Crypto Twitter.
  • Don't try to out-cadence them with empty posts. 1.7/day works because each post is a new market. "Building in public" posts at the same cadence read as noise.

Adapted 6-month plan (no founder fame)

Month 0 (now)

  • Register the brand X handle. Pin a problem-statement tweet, not a product tweet.
  • Write the og-title and og-description in territorial form: "[X] — the [territory]-native [category]" or "the [category] for [niche]."
  • Bake in a quantitative claim or rewards hook in the og-description.

Month 1

  • Ship private alpha. Embed the incumbent's data in your UI as a comparison feed.
  • Design the perpetual referral into the contract, not as a marketing afterthought.
  • Start posting daily: 1 "new market / new feature / new data point" + 3 substantive replies. (Same cadence as the SaySo playbook — see mcap/notes/marketing-research/sayso-x-marketing-playbook.md.)

Month 2

  • Public launch. First blog post is "BREAKING: [PRODUCT] GOES LIVE" — product hero, not partnership. List your three differentiators with screenshots.
  • Start the one partnership/week habit: integrations, not press releases. "Now using [primitive]" > "Strategic partnership with [logo]."

Months 3–4

  • 8 weeks × 1 partnership/week = 8 ecosystem logos visible on your timeline.
  • One unique data point nobody else has, posted weekly. Volume / unique users / settlement times. Become the citation.

Month 5

  • Either become the default integration somewhere with native distribution (Telegram, Farcaster, Discord bot, browser extension) OR co-author a research drop with a logo whose audience you want.

Month 6

  • Only now do you talk about fundraise / acquihire / institutional partners. The audience earned is the audience that will retweet that announcement.

The 80/20

If you copy three things:

  1. Brand handle 60+ days before launch with daily cadence.
  2. "Rewards" hook in the og-description and a perpetual referral.
  3. One weekly partnership announcement for 26 weeks straight.

That's ~80% of what took predict.fun from $0 → $1.8B in 14 weeks of BNB-era operation, minus the founder leverage. With founder leverage, the playbook compounds; without it, the same operational pattern still produces a credible 6-month flywheel.

Appendix — raw evidence

Sep 5, 2024 og-tags (Wayback)

  • Title: "Predict | The Blast-native Prediction Market"
  • Description: "Bet on future outcomes (and earn rewards) across politics, sports, crypto and more on Predict — the Blast-native prediction market."

Dec 18, 2025 BNB launch post — pillars

  • Seamless UX: Smart Wallets, no gas, no approvals, email/Google/X login
  • Predict Points (PP): leaderboard fast-pass for trading + LPing
  • Smart Routing Address: USDT/USDC deposits from 8 networks auto-credited
  • Invites: 10% off invitee fees + 10% inviter fees + 10% inviter PP forever

Mar 4, 2026 Probable acquisition — terms

  • All Probable users' past USDT trading fees refunded at 2× value
  • Probable Points → Predict Points: 1:1 weeks 1–6, 1:10 weeks 7–10
  • Distribution metrics named in the post: $1.5B cumulative volume, 120K+ users, 3.3M txs in ~11 weeks

Mar 31, 2026 funding round — quote

"Having first backed Predict.fun during EASY Residency Season 2, this follow-on strategic investment reflects our deep conviction in the prediction market sector and the team's edge — solving the opportunity cost of prediction by transforming idle collateral into a productive, yield-bearing asset class." — Ella Zhang, Head of YZi Labs

Volume metric in the post: $1.8B cumulative, 14 weeks post-launch.

Founder profile

@dingalingts — "Building @predictdotfun ｜ Previously: @binance Head of Research, Founder @pancakeswap." 254.2K followers, joined Nov 2017, 4,415 posts.

Brand profile

@predictdotfun — "Powering prediction markets on @binance." 50.9K followers, follows 105, 1,131 posts since Jul 2024 ≈ 1.7/day. Pinned tweet (Apr 9, 2026): the Binance Wallet integration announcement.