Back

When Updates Create Problems: UX Lessons from iOS Liquid Glass

4 MINS

# When Updates Create Problems: UX Lessons from iOS Liquid Glass

The Liquid Glass update surprised me—and not in a good way. Despite offering a choice between Classic and Unified design, the changes to core functionality created real-world problems that shouldn't have shipped.

Silent Failures with Real Consequences

The Phone app navigation feels different for the sake of being different. The alarm snooze screen is oversimplified with oversized buttons that aren't the sleek design we've come to expect from Apple.

In India, my sister had calls from unknown numbers automatically silenced after the update. The result? Package delivery disruptionsdrivers couldn't reach her, and she missed time-sensitive communications.

In Qatar, all messages from "unknown" numbersincluding account debit alerts and OTPsautomatically went to Spam. I called my bank and visited my service provider to troubleshoot. The employee's response? "It could be the annoying new update."

The Bigger Questions

Did such a major issue slip through QA? Or was the assumption that all unknown numbers need to be blocked—without giving users an upfront choice—simply not validated?

These aren't edge cases. OTPs and delivery calls are fundamental to how people live and transact, especially in markets like India and the Middle East where communication patterns differ from the US.

What Startups Get Right

This experience made me appreciate how small startupssurviving on caffeine and fighting for bandwidth while drowning in Jira ticketsare able to focus on discovering and solving real pain points.

They don't have the luxury of revamping UI or features that weren't broken. They can't afford to create new pain points, friction, or aesthetic displeasure in the process. Every change has to earn its place.

The Core UX Failure

We eventually found the settings to manage this. But as busy users, we shouldn't have to dig around just to make sure an OTP or package delivery call arrives.

Turning this on by default, without consent or even a heads up, created unnecessary friction.

The principles violated:

No informed consent: Users weren't told their communication patterns would change
No graceful defaults: The assumption didn't match real-world usage
Buried controls: Fixing the problem required hunting through settings

Takeaway for Product Builders

Before shipping changes that affect core user workflows:

Validate assumptions across different markets and use cases
Default to minimal disruption, not maximal "protection"
When in doubt, ask users first rather than apologize later Sometimes the best product decision is leaving things alone.
Background

Tanmayee skipped presentations and built real AI products.

Tanmayee A. was part of the August 2025 cohort at Curious PM, alongside 15 other talented participants.