A deposit screen with a 30% conversion. Support drowning in tickets about the same five steps. Most of the funnel on a phone. Here is how the Deposit flow inside B2Core got rebuilt — what we changed, what we didn't, and what the early numbers say.

Context

B2Core is B2broker's flagship white-label product — one client cabinet sold to 100+ brokers, each of them launches it under their own brand. Deposit is the most important step in the funnel: if a new user can't get money in on the first try, everything downstream stops existing.

Early 2026. My role on this flow — product designer with end-to-end ownership: research, analytics, prototypes, testing, hand-off, production check. Not one of many designers waiting for tickets.

30%
First-deposit conversion
70%
Of the funnel on mobile
5–7
Clicks to reach the address

Three signals we couldn't ignore

Amplitude said only about a third of new users finished the deposit flow on the first try. Support said the same scenario kept showing up under different names. And 70% of the people we were losing were on a phone — exactly where the UI had been built the worst.

What support kept hearing

Tickets settled into four steady patterns. They weren't dramatic. They were dull, repetitive, and obvious — once you read fifty of them in a row.

Most painful

“Where did my money go?”

User funded the account, balance still shows zero, no idea what happened.

Most frequent

“I don't know what to do.”

No instructions inside the flow, so any non-obvious step led straight to support.

Visual fatigue

“Too many buttons and options.”

Cluttered UI, duplicated secondary functions, the primary scenario lost in the noise.

Friction

“Why are there so many steps?”

Five to seven taps for a routine task — every extra fork was a chance to drop off.

How we dug in

Three lanes of research, all running at the same time. They were chosen so each one would fail differently — that way the patterns that survived all three were the ones worth acting on.

Behaviour
Amplitude + live sessions
Walked the funnel through real session recordings, watched exactly where it leaked, step by step.
Voice of customer
Support interviews
Densest ticket database in the company. Repeats and one-offs are easy to tell apart in there.
Market
Competitor audit
How the major players solve this. What they keep one tap away. What they refuse to add.

Three insights everything hinged on

All three were the kind of thing that's obvious in hindsight — and that you can't be sure about without the data.

01
Shorter linear paths win
Fewer clicks and no going back significantly raise the chance the user actually makes it through.
Prototype tests · 3 lengths
02
Don't push unwanted options
The currency-conversion block was almost never used by default — most users didn't even know what it was for.
Amplitude · 95% same-currency
03
FAQ belongs next to the flow
Across major competitors, the FAQ block sits next to the deposit form — not hidden in a help section.
Competitor audit · 6 of 7

Decision 1 — a shorter path

The old flow took five taps to reach the deposit address in the basic case, seven with currency conversion. Most of those steps were dropdowns sized for desktop and dropped onto a phone screen — a form-filling exercise more than a payment flow.

The new flow lands the user on the address in two taps. Dropdowns turned into full-screen pickers with search — more breathing room, faster to find the right wallet or asset, no fishing for tiny tap targets.

Deposit flow — new version, two screens Deposit flow — old version, five screens
Old flow — 5 screens. New flow — 2 screens.
The new deposit flow in motion — Deposit → Address in two taps.

Decision 2 — conversion goes opt-in

Analytics had the answer ready: in 95% of deposits, the currency on the way in is the same as the currency the user just withdrew from another platform. All seven steps with conversion selection existed for the 5% who actually need it.

95%
of users deposit in the same currency they withdrew. The whole conversion block lived for the other 5% — until we pulled it out of the default path and made it a toggle with an explainer.
Payment Details screen — conversion toggle off and on
Payment Details — conversion off (default) and on (with currency picker).

Decision 3 — FAQ at hand

The quietest of the three decisions, and the cheapest to ship. The pattern came straight from the competitor audit: six of seven major players keep an FAQ pinned right next to the deposit form. No one sends the user off to a help center mid-payment.

On mobile the FAQ opens as a bottom sheet — a tap away, scrollable, can be dismissed without losing the form state. On web it sits as a fixed right column. Same content, same questions, two layouts depending on what the screen can give us.

FAQ pinned next to the deposit form — bottom sheet on mobile, right column on web.

Old and new, side by side

Short version. On the left — a cluttered form, dropdowns inside dropdowns, and a sidebar showing other people's deposits, which doesn't help you make your own. On the right — three numbered steps, an FAQ where the noise used to live, and the Deposit button where the eye lands first.

Same product, same brand. The hierarchy is the change.

B2Core deposit page — new desktop version B2Core deposit page — old desktop version
Same product, same brand — the hierarchy is the change.

What the numbers say

Honest framing first. The new flow rolled out to a handful of clients, and it's been under two months since launch. The numbers are real but they read as “where this is heading” rather than “proof”. The full-audience read comes later, on a longer horizon.

30% → 55%
First-deposit conversion
−40%
Support tickets about deposits
5 → 2
Clicks to the deposit address
Sample · timing
Limited cohort, less than two months of data. The full read happens later, on the full audience.

Most of the lift came from things we removed. Five steps became two, a required block became opt-in, and the help moved next to the form instead of into a separate section.

What I carry forward

Three things that travelled out of this case into the next ones.

Lesson 01
Analytics outranks intuition
I'd have kept the conversion block on instinct — “someone might need it.” The 95% figure decided for me.
Lesson 02
Shortest isn't always best
Sometimes an extra step with a clear explainer is cheaper than a hyper-compact screen the user can't read.
Lesson 03
Support is the cheapest insight
They already know what's broken. The job is to listen and turn it into a product decision.