01Case File 01 · Forex & CFD

Registrations Halved.
Deposits Nearly Tripled.

Fourteen weeks, $164,000, four European markets. The account produced 4,720 fewer registrations and 451 more first-time deposits than the period before it.

VerticalForex & CFD
ChannelMeta
MarketsUK · Croatia · Germany · Spain
Window14 weeks
Spend$164,000

What happened

Quick answer

An account optimised toward registrations produced 9,860 of them and 268 deposits. Rebuilt to optimise toward the deposit, the same budget produced 5,140 registrations and 719 deposits — a 2.8x improvement in cost per deposit.

This file is a single rebuild window, not the full multi-market aggregate shown on the forex service page. Different period, different scope.

The numbers

Before And After,
Same Budget

MeasureBeforeAfterNote
Ad spend$164,000$164,000―Same budget, both periods
Registrations9,8605,140▼48% fewer
First-time deposits268719▲168% more
Registration to deposit2.7%14.0%▲5.2x
Cost per deposit——▲Improved roughly 2.8x
Delivery halted per cycleMost of it25–30%▲Not zero. Nobody gets zero.

Diagnosis

What Was Broken

ISSUE 01The campaign was paid to buy registrationsThe conversion event fired on registration, so delivery spent fourteen weeks finding people who register. It succeeded. Almost none of them funded an account.
ISSUE 02Cost per lead looked healthyOn every metric above the deposit the account read as working. The sales desk saw something different, and nobody reconciled the two views.
ISSUE 03One learning contour, tied to the working accountsWhen an account went down, everything the delivery system had learned went with it. Each restart began from nothing.
ISSUE 04The pre-lander was built for review, not for intentIt existed to pass inspection rather than to turn a stranger into somebody with a reason to deposit. It passed. It did not persuade.

The work

What We Changed

CHANGE 01Moved the conversion event to the depositDelivery began optimising toward money arriving rather than toward a form being submitted. Registration stayed in the account as diagnostics only.
CHANGE 02Separated the learning contour from the working accountsConversion history stopped dying with each disabled account, so a restart resumed rather than began.
CHANGE 03Rebuilt pre-landers per marketFour markets, four builds. Local payment methods, local references, local reasons to trust. Not one page translated four times.
CHANGE 04Accepted lower volume deliberatelyRegistrations fell by nearly half and nobody panicked, because the number being watched was underneath.

The uncomfortable part

What Did Not Work

Published because a case file without this section is an advertisement. These cost time and money before they were abandoned.

⚠Abandoned after testing

Lookalike audiences built from registrationsPredictably, they found more people who register. We had rebuilt the event and then seeded the audience from the old behaviour. Two weeks lost before we noticed.
A single unified creative set across all four marketsIt underperformed everywhere and won nowhere. Croatia and Germany want different proof, and averaging them satisfied neither.
Optimising toward deposit in the smallest market firstDeposit volume there was too low to leave the learning phase, so delivery stayed unstable. We moved that market to a broader event and held the deposit event where volume supported it.
Raising the budget once the deposit rate improvedThe instinct after a 14% conversion rate is to scale immediately. We increased spend 40% in one step and the rate fell back to 9% within nine days, because the added budget pulled delivery into audiences the learning had not covered. Scaling in 15% increments held the rate. The lesson is dull and expensive: a better funnel does not survive being filled faster than it was built to fill.

What it means

The Takeaway

→A cheap registration is expensive if the person does not deposit. This account did not need more traffic, better creative or a bigger budget — it needed the platform to be told where the money actually happens.

Related

Forex marketing, measured on depositsOpen → Why registrations climb and deposits do notOpen → The same rebuild in casinoOpen → What your own budget turns intoOpen →
Discuss your account → All documented files