How to Design a Clear Gold Withdrawal Flow When Users Can Only Withdraw in 5-Gram Increments

What Went Wrong with the Gold Withdrawal Feature?

A gold investment app allowed users to buy gold in tiny increments—down to 0.01 grams—but only let them withdraw in fixed 5-gram, 10-gram, 25-gram, 50-gram, or 100-gram bars. The problem? The interface showed no limits. Users could type in any amount—like 1.5 grams—only to be met with a confusing error message: "Withdrawal not possible."

"User gets frustrated. Why? They don’t know what’s allowed. There’s no information about the minimum or the multiples."

This lack of clarity created confusion, frustration, and a sense of failure—despite the user having enough gold in their account. The interface didn’t reflect the real-world constraint: physical gold bars only come in fixed sizes. The design failed because it didn’t communicate the rule, not because the rule was unfair.

Why This Is a UX Failure, Not a Feature Limitation

The issue wasn’t the business rule—it was the user experience. Even if a feature is technically correct, if users can’t understand it, they won’t use it. The app’s withdrawal screen looked like a free-form input field, which implied flexibility. But the backend only accepted whole multiples of 5 grams.

This mismatch between expectation ("I can withdraw any amount") and reality ("Only 5-gram chunks are allowed") is a classic UX failure. The user didn’t know the rule. The system didn’t tell them. And when they tried, it just said "no"—without explanation.

How to Fix It: A Rapid Design Sprint Approach

In a 1-hour workshop, participants used a design sprint to solve this problem. The goal: create a withdrawal flow that clearly communicates the 5-gram rule while keeping the experience simple and intuitive.

Step 1: Understand the Constraints

Before sketching, everyone needed to agree on the real-world limits:

• Gold can only be withdrawn in 5g, 10g, 25g, 50g, or 100g bars.
• Users must have at least 5g to withdraw.
• Withdrawal includes a 5% transaction fee, calculated on the gram amount, not the value.
• Users can’t withdraw partial bars—only full bars.

"If you want to withdraw 100 grams, you need 105 grams in your account."

This rule meant users couldn’t withdraw 45g—they’d need to withdraw 50g instead, leaving 5g unused. The design had to help users understand this trade-off.

Step 2: Sketch Solutions in 5 Minutes

Participants were given 5 minutes to sketch their ideas—no talking, no collaboration. The only rule: the sketch must be clear enough for someone else to understand without explanation.

Key principles for the sketches:

• Use clear visuals (icons, bars, labels).
• Show the available options (5, 10, 25, 50, 100g).
• Show the total and the fee.
• Indicate what happens if the user tries to withdraw an invalid amount.
• Use text to explain rules, not just rely on images.

"If your drawing is messy, add arrows and labels. If it’s good, you don’t need them."

Some ideas included:

• A visual bar chart showing gold bars in 5g increments.
• A dropdown with only valid options.
• A number input with automatic rounding to the nearest valid amount.
• A step-by-step flow that shows the user how many bars they’ll get.

Step 3: Vote with Dot Stickers (Silent Critique)

All sketches were taped to a wall. Participants had 5 minutes to silently read and vote—each person could place up to three dot stickers on their favorite ideas.

"No talking. No questions. Just silent judgment."

This forced people to evaluate ideas based on clarity and impact, not personal preference. The most popular sketches naturally formed a heat map—areas with many dots became the top contenders.

Step 4: Present and Refine

Each team presented their sketch in 1 minute. The presenter explained:

• What the design does.
• How it solves the problem.
• Why it’s better than the original.

After all presentations, participants had a second chance to vote—this time, with more context. They could now change their votes based on the explanations.

"After hearing the idea, you might realize it’s better than you thought."

This second vote helped surface the most viable solution.

The Winning Prototype: Visual Bars with Real-Time Feedback

The top idea used a visual, interactive interface:

• A horizontal row of gold bars (5g, 10g, 25g, 50g, 100g) with clear labels.
• Each bar showed its weight and value.
• Users could tap or drag to select bars—like stacking them.
• A real-time total updated below.
• A "minus" button allowed users to remove a bar.
• A 5% fee was shown next to the total.
• A question mark icon ("?" button) explained the fee when clicked.
• After selection, a confirmation screen showed the total grams and the fee.

"The user just clicks 5, 10, 25—no typing. It’s instant."

This design solved the core problem: it made the 5-gram rule visible, predictable, and easy to use. Instead of guessing, users saw exactly what they could withdraw.

Why This Design Works

• Clarity over flexibility: It removed the illusion of free input. Users can’t type 1.5g because it’s not an option.
• Visual feedback: The bars show what’s available and what’s been selected.
• Real-time calculation: Users see the total and fee as they build their withdrawal.
• Error prevention: Invalid amounts can’t be entered.
• Educational: The fee explanation helps users understand why they need more gold than they withdraw.

"The app is for lazy people. Just click. No thinking."

Key Takeaways

• Constraints are design opportunities: A rigid rule (5g minimum) can be turned into a clear, usable interface.
• Clarity beats flexibility: A simple, predictable flow beats a confusing free-form input.
• Visuals > text: Showing the bars makes the rule obvious. Text alone isn’t enough.
• Test with real users: The final step is to validate the design with actual users.
• Use voting to surface the best idea: Silent critique and dot stickers help teams avoid groupthink and focus on what works.

The winning prototype wasn’t the most complex—it was the clearest. And that’s the real goal of good product design: to make the impossible feel simple.