When Redesigns Go Wrong: Three Cautionary Tales
Scrabble GO, Instacart, and YouTube have all rolled out major app changes that left users unhappy. A redesign isn't inherently a bad idea, but it must be justified by real problems — not just a desire for something new. If you have an established app that users already enjoy, changes should be incremental and validated through A/B testing. If you're facing serious acquisition or retention issues, a redesign may be necessary, but be aware that a poorly executed one can make matters worse.
The Perils of Overhauling a Beloved Interface
Scrabble was long one of the most profitable board games in history, and EA's mobile adaptation was well-received. That goodwill evaporated in early 2020 when Scopely acquired the app and replaced its clean design with something far more chaotic:
The problems are numerous:
- The color scheme is excessive and overwhelming.
- "Start New Game" should be the clear primary action, but competing colored buttons for "Level 5" and "Level 6" dilute its prominence.
- The interface is so crowded that nothing stands out.
- The design offers users no sense of priority or control.
- Navigation includes gated pages, and the iconography is confusing — why show a gems-and-rewards icon when there's already a gem counter in the top banner?
The in-game experience suffered as well. One Twitter user waited over a week for a simple answer about gameplay that could have been documented in an FAQ or Settings page. Scopely acknowledged the backlash and promised fixes, but rather than restoring the original design, it added a new Classic mode buried within Profile Settings — with no indication in the app that it even exists.
Even the Classic mode misses the mark. While the colors are subdued and the top elements have been trimmed, the fundamental complaints about app usability and gameplay remain unresolved. Worse, many users report the app crashing, likely a side effect of cramming a second, overweighted mode into an already-strained application rather than refining what was already there.
The takeaways: don't casually discard a design that faithfully echoed the physical board game. If an app isn't broken, leaving it alone is often the wisest move. And when users expressly ask you to abandon a redesign, ignoring them means ignoring the people who fund your product.
Don't Deceive Users at the Checkout
Instacart's redesign drew objections not from its shoppers — the people ordering groceries — but from the gig workers who fulfill those orders. The grocery-ordering flow itself is standard: users search, add to cart, and choose between pickup or delivery. The trouble begins at the payment screen, which shows a high-level summary:
The cart total, service fee, and sales tax appear together as expected. But the "Delivery Tip" section sits apart from these line items, set automatically at 5% or $2.00. Instacart's shoppers argued this was a deliberate dark pattern. A year earlier, no percentage was displayed — only the dollar amount — which encouraged users to accept the default without questioning it. Spelling out "(5.0%)" invites users to reconsider and tip more.
The tip-editing page compounds the confusion. For small orders, the pre-selected amount sits at the bottom with a random $6 option seemingly inserted for convenience. Users with larger orders see a completely different layout, including an "Other amount" custom field that small-order users don't get. None of this inconsistency serves clarity.
If Instacart wants to avoid alienating both its shoppers and its customers, the checkout flow requires genuine transparency — dark patterns have no legitimate place in commerce. Grubhub demonstrates a better model: the tip line is visible and anchored at the bottom of the page rather than hidden. Tip amounts appear as percentages (which American consumers expect for service) with a custom option for users who want to adjust up or down.
Commit to Your Rollback — or to Your Redesign
When users protest loudly, vacillation only amplifies the problem. If the change is wrong, remove it and keep it removed. If you believe the change is right, stick with it and let users adapt.
YouTube has struggled with the comments section twice. In 2019, the company tested hiding comments behind a button below the "Up next" recommendations; the backlash prompted a rollback. In 2020, YouTube repositioned comments between the Subscribe bar and the suggested videos.
This time, the old location displays an explanatory message with a direct link:
The accompanying tooltip points users to the new location rather than letting them hunt for it — a thoughtful touch. And putting comments back into the main content flow arguably encourages the social interaction YouTube claims to value. There's no credible claim that this layout creates friction or interrupts binge-watching.
The latest complaints come down to users disliking the disruption of their familiar layout. But YouTube is likely acting on data showing that the bottom-of-screen position suppressed engagement or simply annoyed viewers. This may be the right moment for YouTube to ignore Twitter backlash and hold its ground. Rolling back because users hate change — not because the design is worse — teaches users that every redesign will be reversed on demand, and it prevents the longer-term learning that would validate the new layout.
Redesign or Not? Test First
You will never satisfy every user, but the aim is to satisfy most of them. Before committing to a full redesign, the safer path is to validate your ideas with an A/B test. That way, you release only data-backed changes that measurably improve the experience, and you avoid the cost of rolling back a bad release or weathering negative reviews in the press, on social media, or in app store feedback.
