Client under NDA. Market, product and timeline are real; figures are rounded and shared with the client’s consent.
The situation
An Ontario-licensed online casino, eighteen months after launch, running on a platform theme that had been customised with a logo and a colour change. It was visually interchangeable with four competitors that appeared in the same search results — same layout, same jackpot counter, same hero carousel. Paid acquisition was the only growth channel and it was optimised to installs and registrations. Blended CAC had risen for six consecutive months, and the welcome bonus was doing most of the work of getting a first deposit — and none of the work of getting a second.
Day-30 redeposit rate was in the low twenties. The team knew players were leaving after the bonus cleared; they did not have a view of why.
The diagnosis
- No brand a player could name a week later. In player interviews, users who had deposited could not tell the client apart from two competitors. Retention was being fought with bonus spend because there was nothing else to hold on to.
- Onboarding built for the platform, not the player. Registration → verification → deposit → lobby, with a wall of 1,200 game tiles sorted by provider deal rather than by what a new player would actually try first.
- Paid optimised to the wrong event. Campaigns were bidding for registrations, so the algorithm found the cheapest registrations: bonus hunters who deposited once, cleared the wagering and left.
- Interac was buried. The payment method most Canadian players use was fourth in the list.
What we did
- Months 1–2. Rebrand off the template. New identity, a lobby that leads with a curated first-session set for new players, a promo system the marketing team can change without a developer, and a game-tile design that is recognisably the client’s.
- Month 2. Onboarding rebuilt around a second session. Interac first in the cashier, deposit minimums shown up front, verification moved to withdrawal, and a first-14-days journey whose only goal is a second deposit made for a reason other than a bonus.
- Months 2–5. Paid moved to cohort value. Campaigns restructured to optimise on FTD and 30-day redeposit, with source-level cohorts reconciled monthly against the platform. Sources that produced cheap FTDs that churned by day 20 were cut, even where it made the headline CPA look worse for a month.
- Ongoing. Bonus economics modelled by segment; the welcome offer resized so that it recruits players rather than abusers.
The numbers
Five months, measured from the launch of the new front end.
- FTD volume: +186% on a paid budget that grew by 22%.
- Blended CAC: −28%. The fall came from two places: fewer paid registrations wasted on bonus hunters, and organic and direct traffic that started to exist once the brand became memorable.
- Day-30 redeposit: 34%, up from 21%. This is the number the whole engagement was for.
- Registration completion: +52% after cutting the form and moving verification.
What we’d do differently
We launched the rebrand and the new onboarding in the same week to hit a promotional calendar. That made the attribution messy for a month: we can say what the combination did, but we cannot cleanly separate how much of the Reg2Dep gain came from the design and how much from the flow. On the next project of this shape we would ship the cashier and onboarding changes first, measure for four weeks, then launch the brand. It is slower and it would have cost the client one promotional window; it would also have given us a number we could defend line by line.
Visually identical to four competitors in the same SERP. We rebranded off the template, rebuilt onboarding, and moved paid acquisition to optimise on 90-day cohort value.