The first-trader funnel the brief describes, measured wallet by wallet from public onchain data, and the channel plan that follows from it. The demo runs the funnel live and links every wallet to Arbiscan.
The June 2026 governance vote moved the buyback budget into growth and set a $267k monthly revenue baseline. September cleared it 2.5x. The base is still small: about 10 organic new traders a day on Arbitrum and 238 daily users across chains.
Half the "new traders" are one operator's wallets: funded by a single address that is also the top referral code, opened at 150x, never back. Without them the organic base repeats well (39% at D7). It starts mostly on crypto, and FX and metals starters repeat a little more.
Count only real acquisitions: screen funders before paying referrers, tag every first trade to a source, then spend on what retains: macro-calendar launches, referrers paid on returning traders, integrators and a few high-value trader deals. Call each channel weekly on payback and D7.
The one line: separate operator wallets from traders, then price every channel on the fees its real traders pay in their first 30 days.
A perps venue since 2021: synthetic markets priced by a Chainlink DON and settled against shared gToken vaults. The levers a growth lead pulls, as they stand today.
| Lever | Today | Growth read |
|---|---|---|
| Markets and leverage | 493 pairs configured. Group caps: FX majors 1000x, minors 750x, exotics 500x, gold and silver 250x, indices 100x, BTC and ETH 200x (500x on degen pairs), altcoins 150x. | No crypto-only venue follows Gains into long-tail FX and metals at this leverage. |
| Chains | Arbitrum, Base, Polygon, MegaETH; Solana wallets trade through sol.gains.trade and settle on Arbitrum. Arbitrum carries about 80% of 72h volume. | One chain to instrument first. |
| Fee split | DAO 76%, vault LPs 15%, referrals 5%, trigger keepers 4%, while the buyback is paused. | Most of each fee dollar is available to fund growth. |
| Referrals | Self-serve since 14 Sep 2026: 10% of L1 fees, 5% at L2, paid daily in USDC, lifetime attribution. Higher rates negotiable. | The rails exist. Coverage and quality are the work (★). |
| Loyalty | Gains Credits up to 25% off fees; GNS staking discount up to 50%; Gains Bonus margin credits on Arbitrum; Gains Guardians UGC program. | Retention tools already built. Tie them to the first week. |
| Governance and budget | June 2026 vote: the 55% buyback share and the old operating budget fund a Growth & Marketing Treasury, with monthly public reporting of spend and attribution. | Every dollar is public. Attribution has to hold up in a forum post. |
| Scale | DefiLlama 30d fees $0.84M, revenue $0.63M; TVL $8.7M; 2025 revenue $9.5M against $16.4M in 2024. | Room to grow back to 2024 levels. |
Sources: Gains trading-variables API (pairs and group caps), Gains docs (fees, referrals, loyalty), governance forum (Make Gains Great Again proposal), DefiLlama. Read 6 Oct 2026.
Where a new perps trader can go instead, ranked by 30-day fees (DefiLlama, 6 Oct 2026), and the opening each one leaves.
| Venue | Fees 30d | Model and reach | What it wins on | Gains' opening |
|---|---|---|---|---|
| Hyperliquid (incl. HIP-3) | $86.0M | Order-book L1; HIP-3 deployers such as trade.xyz list stocks, indices, metals, some FX | Liquidity, builder codes, the default venue for crypto traders | Leverage on FX and metals; long-tail FX pairs |
| Jupiter Perps | $7.9M | Pool perps on Solana, crypto majors | Solana distribution | Solana-wallet front end with FX and metals |
| Aster | $5.4M | Order book and pool, multi-chain; stock perps | Airdrops and BNB Chain reach | FX depth and leverage |
| Lighter | $3.2M | ZK order book; zero retail fees | Free trading | Market coverage beyond crypto |
| GMX V2 | $1.1M | Pool perps, crypto | Arbitrum brand, LP yield | Same chain, wider asset menu |
| Veranta (ex-Avantis) | $0.34M | Vault perps on Base, RWA focus; renamed Sep 2026 | Zero-fee, pay-on-profit pitch | Track record since 2021, against a brand that restarted in September |
| Ostium | $0.08M | Vault perps on Arbitrum, RWA focus | Was the RWA-perps story of early 2026 | Fees near zero since its July 2026 vault exploit |
| CEX TradFi (Binance, Bybit MT5) | n/a | KYC venues adding metals, stocks and FX | Distribution and fiat rails | No signup, no custodian |
CoinMarketCap's May 2026 study of RWA perps covers 17 venues, including Ostium, Avantis and GRVT. Gains is absent. RWA perps did $117B in August. Getting into these datasets is cheap earned media.
In that study commodities are 81.9% of RWA perp volume and FX 0.9%. 1000x FX makes a strong headline. Gold and silver at 250x is where the volume sits.
Fees: DefiLlama fees API (the Rivals tab of the demo pulls them live). Ostium exploit: CoinDesk, 15 Jul 2026. RWA data: CoinMarketCap Research (May 2026); Crypto Briefing on a16z data (1 Oct 2026). Rival leverage caps are left out until checked on each venue's docs.
From the demo: every wallet whose first Arbitrum trade fell in the 60 days to 6 Oct 2026 (1,221 wallets), followed through its trade history, referral binding and the source of its first USDC. D7 means another trade 1 to 7 days after the first.
| Finding | Numbers | What I would do |
|---|---|---|
| Half the new wallets trace to one funder | 582 new wallets got their first USDC from one address, which is also the referral code with the most bindings. It sent $2.24M to 1,114 wallets in the window; 52% of them became "new traders" here. Most open at 150x on altcoins with small collateral. None of its 540 measurable wallets traded again within 7 days; they pay 18% of cohort fees. A second address funded 20 more. | Separate operator wallets from acquisition metrics. Ask who runs it: an app creating wallets for users, a farm, or a market maker. Pay referral rates above the floor only on referees who return. |
| Organic traders retain | 619 organic new traders, about 10 a day. D7 repeat 39%, D30 59%, against 19.5% and 29% with the operator wallets counted. | Report the organic line in the monthly treasury update so growth is judged on traders. |
| FX and metals starters repeat a little more | Organic first trades: crypto 75%, D7 37%. FX 6%, D7 48%. Commodities 8%, D7 42%. Small samples (35 and 47 wallets). | Point launches at the edge on the macro calendar (CPI, NFP, FOMC, gold moves) and re-measure as those cohorts grow. |
| Fees are top-heavy | Top 10% of organic new wallets pay 81% of their 30-day fees; top 1% pay 40%. Median $8. | Run two motions: broad acquisition and a short list of high-value trader deals. |
| Attribution has gaps | In the last 72h, 29% of Arbitrum opens carry no attribution code. On Base, 87% of opens come through one builder code, so integrators already move flow. | Tag every open by source before scaling paid spend. Treat integrators as a channel with its own ledger row. |
Method: first trades from the Arbitrum public RPC (TradeStored events with trade index 0); trade histories and referral bindings from Gains' public API; funders from native-USDC transfer logs (sampled, then traced in reverse). Fees are trading fees on trades opened in the first 30 days. Live numbers change daily; the demo shows the current cut, its limits and every wallet.
Each channel in the brief, with the first move and the number that judges it.
| Channel in the brief | First move | Judged on | Detailed in |
|---|---|---|---|
| Launches on X | A listing calendar built on the macro calendar. Each FX, metals or index listing ships with a chart, a trade idea at sane leverage and a tagged link; result posted at D7. | First trades on the launched pair within 7 days | §04 move 2 |
| KOLs and trader sponsorships | Short retainers with a trial month, paid on first-time traders who reach D7. Two or three sponsored traders who post real PnL on FX and metals. | CAC per returning trader; fees of sponsored flow | §04 moves 3, 5 |
| Affiliates and referrals | Screen the current book for funder clusters first. Then recruit 30 referrers: prop-firm traders, gold and FX Telegram groups, MT5 communities. Rates above the floor go to referrers whose traders return. | Share of new traders referred; D7 of referred vs organic | §04 move 3 |
| UGC, content and community | Guardians briefs on one theme per week, tied to the launch calendar. Disclosure on every paid post. | Tagged first trades per creator | §04 move 2 |
| Paid acquisition | Held until attribution covers every open. Then small tests on X and crypto ad networks, geo-fenced to markets where the product is lawful to promote. | Payback under 30 days | §06 |
| Partnerships and integrators | A builder-code deal sheet: terminals, wallets, prop firms. Base already shows one integrator carrying most opens. | Fees routed per integrator; net of rev share | §04 move 4 |
| Numbers | The weekly cohort board in the demo, plus source tags at first trade; scale, modify or kill each Monday. | Net contribution per channel | §06 |
Ranked by expected fees per dollar, each grounded in a mechanic Gains already has.
Attribution codes and source tags already ride on each trade. Close the gap so every open carries one, bind referral codes before the first trade, and join UTM to wallet at connect. This makes the monthly treasury report defensible.
CPI, NFP, FOMC and every big gold move is a launch moment for FX, metals and indices. One post per event with a chart, a trade idea at moderate leverage and a tagged link. Weekend closures get a Friday reminder.
Keep 10% / 5% as the floor. Add a raised rate once a referrer's traders pass a D7 bar. Before any raise, check who funded the referred wallets: the demo already found one referrer whose bound wallets it funded itself.
Builder codes route flow from terminals, wallets and prop firms. A one-page deal sheet, a public integrator leaderboard, and fees routed per integrator in the weekly ledger.
The top 1% of organic new wallets pay 40% of their first-month fees. Fee-tier offers through Gains Credits and the large-trader program, plus two or three sponsored traders who trade FX and metals in public.
Send data to CoinMarketCap Research, a16z and DefiLlama's RWA views so Gains appears in the next study. Publish one consistent lifetime-volume figure: the hiring page, homepage and mobile page show $84B, $110B and $137B.
Specific KOLs, communities and integrators I would approach are held in my application materials.
| Window | Work | Done when |
|---|---|---|
| Days 1 to 30 | Cohort board live internally. Source tag coverage audit with Data and Product. Identify the operator behind the funded cluster. Funder screen on the referral book. Macro-calendar launch plan for Q4. First 10 referrer conversations. | Every channel has a ledger row and a baseline |
| Days 31 to 60 | Three launch moments shipped and measured at D7. First-trade leverage test. Retention-weighted referral rates live. Integrator deal sheet out to the first five. | First scale, modify or kill calls made on real data |
| Days 61 to 90 | Paid tests if attribution coverage holds. Sponsored traders live. First hire for the system: a creator and community operator. Monthly public report includes cohort metrics. | Net contribution per channel reported monthly |
The definitions I would hold every channel to. The demo's Channel ledger applies them.
| Metric | Definition | Source |
|---|---|---|
| First-time trader | Wallet's first trade on a chain (trade index 0), deduplicated across chains, excluding wallets from a funder cluster | TradeStored events |
| D7 / D30 repeat | Another trade opened 1 to 7 (or 30) days after the first; same-session trades excluded | Trade history |
| Fees per trader, 30d | Trading fees on trades opened in the first 30 days | Trade history |
| Contribution per trader | 30d fees × protocol revenue share, less referral and integrator payouts | DefiLlama ratio, payout terms |
| CAC and payback | Channel spend ÷ first-time traders; days of contribution to recover it | Treasury spend log |
| Weekly call | Kill: payback over 90 days or no contribution. Modify: payback 30 to 90 days or D7 below organic. Scale: everything else. | Channel ledger |
Volume is a segment and impressions are a diagnostic. Neither decides spend.
Outside-in, from public data only (public job posting, 2026). Live figures read on 6 Oct 2026. Confirm every figure internally before it drives spend.
Gains: fees and split · referral program · Gains Credits · Gains Guardians · backend API
Governance: Make Gains Great Again (June 2026) · GNS staking
Data: DefiLlama · Arbitrum public RPC · Gains stats API
Market: CoinMarketCap RWA perps, May 2026 · Crypto Briefing on a16z data
Independent homework by Edward Tay for the Gains Head of Growth application. Public data only, no confidential Gains information. The demo is my own prototype.