Crypto exchanges live and die by active traders, not download counts. The trouble is that the two rarely line up. Hundreds of exchanges chase the same pool of real traders, and standard app install campaigns keep landing on the wrong people: crypto-curious browsers, incentive hunters, and bots that inflate cost per install and never place a single trade.
For this exchange, one of the three most trusted platforms in the world, the brief was narrow and unforgiving. Drive efficient installs from verified crypto traders, not just curious audiences, and keep cost per install in check while doing it.
Demographic targeting could not get there. Deciding that someone “looks like” a trader is not the same as knowing they hold a wallet and use it. The exchange needed an engine built on on-chain behavior it could verify, not proxies it had to hope for.
The exchange ran its acquisition through Addressable, targeting verified wallet holders by their actual on-chain activity across two channels at once. Instead of chasing surface metrics like raw click-through or cost per click, the team optimized against the numbers that move the business: cost per install and downstream lifetime value (LTV).
Three things made it work.
01. Precision Targeting: real wallet owners, not lookalikes
Addressable maps wallets to devices, so every campaign reached people with a verified on-chain history rather than an interest tag. That is the core shift, from “this person might be into crypto” to “this person holds a wallet and trades with it.” In a market crowded with bots and tourists, it meant budget flowed to users who could actually become traders. It is the same logic behind cost per wallet (CPW): pay to reach proven on-chain users, not impressions.
02. Multi-Channel Strategy Optimized for CPI and LTV
Addressable maps wallets to devices, so every campaign reached people with a verified on-chain history rather than an interest tag. That is the core shift, from “this person might be into crypto” to “this person holds a wallet and trades with it.” In a market crowded with bots and tourists, it meant budget flowed to users who could actually become traders. It is the same logic behind cost per wallet (CPW): pay to reach proven on-chain users, not impressions.
3. A head-to-head test against the exchange’s own native campaigns
The strongest proof was not a benchmark or an estimate. The exchange was already running its own native (non-Addressable) campaigns on the same platforms at the same time, so the comparison was direct and like for like.
iOS app install performance on X (Twitter):
Android app install performance on X (Twitter):
On iOS, Addressable nearly doubled click-through rate while holding cost per install essentially level with the native campaign. On Android, the gap was wider: Addressable cut cost per install by 49 percent, from $78 down to $40.
Over roughly two and a half months, the campaign delivered:
Channel breakdown:
The headline number is the 49 percent cheaper Android install. The more useful lesson is what happened on Reddit. It cost more per install, $83 against X’s $51, and the team still moved more budget toward it.
Why? Because the backend data showed Reddit users traded more, stayed longer, and kept their accounts active. On a pure CPI scorecard, Reddit looks like the loser. On an LTV scorecard, it was the best spend in the account.
That only becomes visible when you can tie an install back to a real, verified on-chain user and follow what they do next. Cheaper installs are easy to buy. Better users are not. When every impression reaches someone who already holds and uses a wallet, “cheaper” and “better” stop being a trade-off.
We stopped optimizing for the cheapest install and started optimizing for the trader behind it. That is when the math changed.
Head of Growth, top 3 global crypto exchange
By targeting verified wallet holders instead of demographic lookalikes. Addressable maps on-chain wallet activity to devices, so app install ads reach people with a proven trading history rather than anyone who simply fits a “crypto interest” profile.
Wallet-based targeting reaches users based on their real on-chain behavior: the tokens they hold, the swaps and trades they make, and the protocols they use. It replaces guesses drawn from age, location, or interest tags. In short, it is the difference between probably and verified.
CPI tells you what an install costs. LTV tells you what the user is worth. A channel with a higher CPI can still be the better investment when those users trade more and stay longer, which is exactly what happened on Reddit in this campaign.
X (Twitter) and Reddit, run in parallel as a single multi-channel app install strategy spanning both iOS and Android.
