> ## Content Index > Fetch the complete content index at: https://www.web3-advertising.com/llms.txt > Use this file to discover other available public pages before exploring further. # You can't tell a good user from their first trade. - URL: https://www.web3-advertising.com/you-cant-tell-a-good-user-from-their-first-trade/ - Published: 2026-09-11T11:03:53.000Z - Updated: 2026-09-11T11:03:53.000Z - Description: The whale and the tourist look the same on their first trade. Only time tells them apart. - Author: Madusha Thilakarathna Every advertiser we work with eventually gets down to asking us the same question, and it's a valid one. They've run one or two campaigns, they've read the reports, and then it comes up. The volume is lopsided. A small group of users did most of it, while a long tail did almost none. So they ask why they're paying the same for a user who barely traded as they are for one who moved five or six figures. Charge me for the ones who matter, they say, and stop charging me for the rest. I get it, because for a while we'd have said the same thing. When you're on a budget and the value per user swings by such orders of magnitude, a flat fee for all can feel like you're overpaying for the ones who didn't bring much to the table. The dashboard leans the same way. It shows a conversion with a small number sitting next to it, and the first instinct is to draw a line under those as not worth it. Their instinct is fine. The trouble is the number they're judging them on, because that first small trade isn't telling them what they think it is. A first trade is usually a small amount of money with almost no information attached to it. The users who end up mattering rarely jump in with huge trades on day one. Someone finds your product through an ad, does a cautious swap or two with a dollar here and another ten there to see whether it works, then comes back a week, two weeks or even three weeks later and moves some serious size once they trust it. For some of the best users, that's the pattern. A small test, a pause, maybe another test, and then finally the real activity once the product has earned it. Based on that early snapshot, this person and the mercenary trader who tried you once and left can look identical. Both trade a minuscule amount. Both go quiet for a few days. Judge them on that first move and you can't tell them apart. So you risk throwing out the one who was about to become a valuable user alongside the one who was never coming back. It cuts the other way too. The user who shows up big, moves size on day one, and looks like your dream conversion is sometimes the one farming an incentive who's gone by the end of the month. Grade this one on the snapshot and you'd pour more budget into exactly the wrong wallet. The early read flatters some users and hides others, and it does both for the same reason. Don't fall into the trap and take it too soon. Volume is one of the most important numbers to watch, but it can deceive you because it isn't a number that arrives in one go. It takes time to finish happening. So what are you judging here? The user, or the one moment in their history you happened to measure first? A first trade is too early to carry that verdict. You're better off judging the cohort as a whole and reading it over a window long enough for the full picture to show up. [Make sure you don’t have retention backwards.Your first ten users already know.![](https://storage.ghost.io/c/dd/d7/ddd7acdd-b0e2-462c-8132-53172eb6962a/content/images/icon/w3a-929abecb-8b0b-459d-b72b-e1891dc2c4ff.png)Web3-AdvertisingWayne Hattingh![](https://storage.ghost.io/c/dd/d7/ddd7acdd-b0e2-462c-8132-53172eb6962a/content/images/thumbnail/acq-or-ret5-e9dbf976-0475-4e0b-8238-1055ec00a17c.png)](https://www.web3-advertising.com/make-sure-you-dont-have-retention-backwards/) There's a genuine cost to jumping the gun and grading too soon. Act on the first snapshot and you start optimising toward whoever looks good on day one. That can train you to chase fast, shallow activity while walking away from slower users who end up being worth considerably more. Worse, it carries over into the next campaign. Shape this cohort around the wrong behaviour and you shape the next one around it too. None of this means you take whatever you're given. Judge the numbers, but judge them on the whole cohort and over a proper stretch of time, not on the first moves the user makes. Hardly anyone waits that long, but it's the read that tells the real story.