22. Case Studies¶
Who should read this page¶
This page is mainly for product teams, solution architects, fraud teams, and client-facing stakeholders who want realistic examples of how liveness fits into different journeys.
Why this page exists¶
Case studies make the playbook easier to apply.
Instead of talking only about theory, this page shows how liveness strategy changes with the use case.
Case study 1: remote account opening¶
Goal¶
Stop common spoof attacks without creating too much onboarding friction.
Typical flow¶
flowchart TB
A[Capture selfie] --> B[Quality check]
B --> C[Liveness]
C --> D[Face match to ID]
D --> E[Risk policy]
E --> F[Pass, retry,<br/>or manual review]
Common concerns¶
- replay attacks with the real person's media
- low-quality first-time captures
- conversion drop from strict policy
Good design pattern¶
- clear capture guidance
- retry band instead of hard fail for ambiguous cases
- stronger policy for risky segments
Case study 2: login step-up verification¶
Goal¶
Confirm that a returning user is genuinely present during a sensitive action.
Common concerns¶
- users expect speed
- mobile app and web may behave differently
- too much friction harms trust
Good design pattern¶
- use liveness only when risk triggers it
- keep latency budget tight
- monitor retry rate carefully
Case study 3: high-value transaction approval¶
Goal¶
Reduce fraud on high-risk money movement or profile-change events.
Common concerns¶
- stricter security needed
- strong audit trail required
- attackers may be more determined
Good design pattern¶
- tighter threshold or stronger challenge
- extra device and session checks
- clear fallback path for genuine users
Case study 4: account recovery¶
Goal¶
Prevent attacker takeover when the user has lost normal credentials.
Common concerns¶
- higher fraud pressure than normal login
- weak fallback design can be abused
- support burden rises if policy is too strict
Good design pattern¶
- stricter policy than standard login
- strong monitoring and review path
- careful session security and logging
Case study 5: browser-first onboarding¶
Goal¶
Support wide reach without forcing app install.
Common concerns¶
- webcam quality varies widely
- browser support differs
- replay exposure may be higher on desktop
Good design pattern¶
- channel-specific thresholding
- browser-aware capture guidance
- clear fallback or assisted path for weak setups
Compare the use cases¶
| Use case | Security pressure | UX pressure | Typical policy strength |
|---|---|---|---|
| account opening | high | medium | balanced to slightly strict |
| login step-up | medium to high | high | balanced |
| high-value transaction | very high | medium | strict |
| account recovery | very high | medium | strict plus fallback |
| browser-first onboarding | medium to high | high | channel-aware |
Final takeaway¶
The right liveness design depends on:
- the risk of the journey
- the tolerance for friction
- the platform constraints
- the attack pressure
That is why a single universal policy rarely works across all journeys.
Need term help?¶
If any technical terms on this page feel dense, use Appendix A1 — Key Terms first and then jump to the relevant appendix page for deeper detail.
Related docs¶
Read next¶
Go to 23. System Architecture.