Case studies How I work About Resume ↓
← Portfolio Case studies Credit Card Marketplace
CASE STUDY · BONIFY / SCHUFA GROUP
Role: Product / UX Designer Platform: Web, Mobile Web, App
bonify.de/kreditkarten
The redesigned bonify credit-card marketplace on desktop — a filter bar (High approval chance, Travel rewards, Low interest), an expanded Mastercard offer card showing annual fee, interest rate, full card information, payment details and benefits at a glance.
01 · THE PROBLEM

Every card looked the same

The marketplace presented every credit card the same way, regardless of who was looking. A user landed on a long list of near-identical offers and was left to do all the work themselves: compare cards manually, decode financial jargon, and somehow figure out which one actually suited them.

The result was cognitive overload. People couldn't tell which card was right for them, lost confidence in their own choice, and disengaged before applying. The offers weren't bad — the experience just gave nobody a reason to pick one over another.

The original mobile-web and web marketplace with critique annotations: no personalization, all offers look identical, key benefits buried in long text, filters hidden far below and lacking clarity, and the comparison table feeling disconnected from the page.
The starting point. I audited the old web and mobile flows and mapped every friction point: no personalization, identical-looking offers, buried benefits, hidden filters, and a comparison table that felt like a different page entirely.

Five problems kept surfacing: no personalization, weak differentiation between offers, limited filtering, important decision criteria buried, and users left unsure which card fit their profile.

02 · RESEARCH

What actually drives the choice

I started outside our own product. I benchmarked five marketplaces people already trust — Check24, NerdWallet, Forbes Advisor, Credit Karma, and Capital One — to see what each did well and where the opening was for us.

Competitive analysis table comparing Check24, NerdWallet, Forbes Advisor, Credit Karma and Capital One across key features, layout structure, strengths, weaknesses and opportunities for us.
Competitive analysis. The pattern was clear: the strongest players win on personalization and approval-odds — exactly the area where bonify's own credit data gave us an unfair advantage.

Across the research, the same handful of factors decided which card people chose — and they weren't evenly weighted.

Top 5 drivers in most studies: rewards & perks (highest influence, often over 80%), cash back / reward type, annual fees & APR, social trust / recommendations, and eligibility / recommendation indicator such as 'good match for your profile' or 'approval chance'.
The top five decision drivers. Rewards lead, but eligibility signals — “good match for your profile,” “approval chance” — were the lever no competitor in our market had pulled.

That gave us four opportunity areas to design against: personalization, smarter comparison, better filtering, and confidence-building.

Opportunities for us: Personalization (recommend 'Best Match' cards with approval chance), Smart Compare (guided comparison with highlighted benefits), Advisor (let users ask which card suits them), and showing reward/cashback points (estimated annual savings).
Four opportunity areas, each tied directly to a research finding.
03 · DESIGN STRATEGY

Designing for the decision, not the card

The instinct was to redesign the card UI. The real job was to improve the decision. So instead of prettier cards, I built three things that reduced the work of choosing — each one drawn straight from the research.

1 · Personalized approval chance

Users didn't know whether they'd even qualify — and that uncertainty quietly killed applications. Using bonify's own credit data, I added a profile-based approval indicator (“90% approval chance”) so people could see, up front, which offers were realistic for them.

Mobile credit-card offers screen — 'We've selected cards that fit your financial profile', filter chips for High approval chance / Travel rewards / Low interest, and offer cards with approval chance, fees and a clear primary CTA.
01 Personalized offers, ranked by fit
“Cards that fit your profile.” The list now opens with relevance — approval chance surfaced on every card, so people prioritise the offers they're actually likely to get and apply with more confidence.
2 · Filtering people actually use

The old filters were hidden, vague, and demanded too many clicks. I rebuilt them around the criteria research said people decide on — and surfaced the most common ones as one-tap chips so narrowing the list took seconds, not a sub-menu hunt.

Filter options grouped into meaningful categories: card type / use case, annual fee, rewards type, eligibility / credit-score requirement, approval chance, APR / interest rate, and card brand.
The filter model — built around real decision criteria (use case, fee, rewards, eligibility, approval chance, APR, brand) rather than database fields.
Mobile 'Filter By' bottom sheet — card type, annual fee, reward type, card system and APR as tappable chips with a sticky 'Apply filter' button.
02 One-tap filters, on a sheet that gets out of the way
Meaningful, tappable, fast. Card type, fee, rewards, brand and APR as chips on a single sheet — the effort of finding a relevant card dropped sharply.
3 · Offer cards with a hierarchy

The old cards had information but no order — the things that drive a decision were buried in body text. I rebuilt the card around a priority hierarchy: welcome bonus, card identity, approval chance, annual fee, interest rate, savings potential, then the CTA. Everything a person needs to decide, in the order they need it.

The offer-card component system — approval-chance status badges, compact and expanded offer cards, special-offer and applied-card variants, all built as reusable components.
The card as a component system — compact, expanded, special-offer and applied states, with approval-chance badges, so the new hierarchy held up across every context in the marketplace.
04 · THE OUTCOME

Leads up 233%

↑233%
Increase in lead conversion after launch
3
Decision-support features: approval chance, smart filters, ranked offers
5
Marketplaces benchmarked to find the personalization gap

The number that matters underneath the 233% is why it moved. People could finally identify a suitable card quickly, understand their approval likelihood, filter offers without friction, and compare the criteria that actually mattered:

  • Quickly identify suitable cards instead of scanning an undifferentiated list.
  • Understand approval likelihood before investing effort in an application.
  • Filter relevant offers faster with criteria built around real decisions.
  • Compare key financial criteria — fee, APR, rewards, savings — at a glance.
  • Decide with confidence, guided toward fit rather than left to self-serve.
05 · WHAT I LEARNED

Decision support beats information

This started as a card redesign and became a marketplace-optimization project. The lesson held all the way through: personalization builds confidence — showing approval probability turned a wall of options into a short list of realistic ones.

"Users responded better when the experience guided them toward relevant offers — not when it presented everything equally."

Simpler comparisons improved decisions. Highlighting only the most important criteria cut cognitive load, and guiding people toward relevant offers beat showing them everything at once. The win wasn't a better-looking card — it was an experience that helped people choose.

PREVIOUS CASE STUDY ← BBQ Nation Back to case studies