Six products that drowned in their own features

The product

Word, launched 1983. Now Office 365 bundle.

Commands in modern Word

~1,500

Year Ribbon introduced

2007

Active users

1B+

The bloat

Modern Microsoft Word contains approximately 1,500 commands. The majority of users touch fewer than 20 of them in a typical session. This is not a new discovery: Microsoft's own user research in the early 2000s revealed that the top requested “new features” from customers were already in the product, just undiscoverable. Users wanted things that existed. They simply could not find them.

The response was the Office Ribbon, introduced in Word 2007 and widely regarded as Microsoft's most significant UI overhaul. The Ribbon was designed to surface commands that users were not finding. It reorganised the toolbar into tabs and groups, with contextual tabs appearing based on what the user had selected. It was a genuine design achievement. It was also a band-aid on a wound that ran deeper than navigation.

The cost

Over a billion people carry the cognitive load of a product with 1,500 commands. The cascade into PowerPoint and Excel meant the bloat was replicated across the Office suite. Years of design debt accumulated. The Ribbon was controversial at launch because it moved controls that power users had memorised. Training content became obsolete. Enterprise IT departments ran re-training programmes. The cost was not just time and money; it was the trust of users who had invested years in learning the product's conventions.

The lesson

When your feature audit result is “we need better navigation,” that is a bloat signal, not a design brief. The navigation problem is a symptom. The disease is the 1,480 commands that serve few people but must be discovered, maintained, documented, and tested. Microsoft's resources meant they could invest in a Ribbon-level solution. Most teams cannot. The lesson is to treat the discoverability crisis as the alarm it is, not the design challenge it presents as.

Anti-patterns: kitchen sink UX, no sunsetting

Launch year

2008

Peak registered users

225M

CEOs 2007-2024

4

Acquired by

Bending Spoons, 2024

The bloat

Evernote launched in 2008 with a clear proposition: remember everything. It was elegant, fast, and used by millions of people to organise their thinking across devices. The note-taking core was genuinely good. What followed was a systematic departure from it.

Evernote added: business card scanning, Spaces (a collaborative workspace layer), Presentation Mode (Evernote as slide deck software), Tasks (a to-do list integrated into notes), Calendar (a full calendar integration), a marketplace that sold, among other things, physical socks and backpacks branded with the Evernote elephant. The core note-taking was neglected while each new vertical consumed engineering resources. Sync speeds slowed. Mobile performance degraded. Bugs in the core product lingered.

The cost

Phil Libin stepped down as CEO in 2015. His replacement, Chris O'Neill, lasted until 2018. In the same month O'Neill departed, TechCrunch reported that Evernote's CTO, CFO, CPO, and HR head had all left within a single month. TechSpot and other outlets published reports of internal sources describing a “death spiral.” The company announced 15% layoffs. Ian Small became the third CEO, tasked with fixing what had been built. By 2024, Evernote had been acquired by Bending Spoons, an Italian app studio that acquires and revitalises distressed software products. The socks are no longer for sale.

The lesson

Product strategy is a thousand no's. When a CEO pivots every three years, feature bloat is the unintended consequence of uncommitted strategy. Each new CEO brought a new vision; each vision required new features; each new set of features strained an engineering team that was also expected to maintain the previous CEO's vision. The product became a archaeological record of leadership decisions, not a coherent tool.

Sources: TechSpot 2018, TechCrunch 2018 (executive exits), MakeUseOf 2024 (acquisition). Anti-patterns: feature factory, CEO pet feature

The bloat

Zoom launched in 2013 and IPO'd in 2019. The pandemic made it the default word for video calling. What followed was a familiar trajectory: a company that defined a category attempted to expand into the surrounding landscape.

Zoom added: Zoom Mail, Zoom Calendar, Zoom Whiteboard, Zoom Docs, Zoom Team Chat, and Zoom Contact Center. The stated rationale was ending “the toggle tax” of switching between apps. The pitch was a unified workspace. The result, based on Capterra user reviews and community forum discussions, was a product that charged enterprise customers for services they did not use. One representative Capterra review: “We don't use the chat, email or storage functions; this makes it feel like we are paying for a lot of things we don't need.”

The cost

While Zoom's engineering resources were deployed into Mail and Calendar, its core video-conferencing UX received less design attention. Google Meet and Microsoft Teams closed the feature gap. The market for video calling became competitive on the core product while Zoom was also trying to compete in email and document collaboration, markets with entrenched and genuinely excellent incumbents.

The lesson

Adjacent-market expansion is often just feature bloat in a press release. If you are buying one market, the other markets will notice. The “toggle tax” framing is honest as far as it goes; nobody wants to switch between eight apps. But consolidating into one app that does eight things averagely is not obviously better than eight apps that each do one thing excellently.

Anti-patterns: competitor parity paranoia

The bloat

Slack launched publicly in 2014 with a simple proposition: less email, better communication. It was acquired by Salesforce in 2021 for $27.7B. Post-acquisition, the product has expanded significantly. Huddles, launched in 2021 as an audio-first lightweight alternative to Zoom calls, rapidly accrued video, screen-share, captions, draw-on-screen annotations, and recording. Huddles now competes directly with Zoom.

Alongside Huddles: Canvases (a collaborative document layer), Lists (a project management layer), AI summaries, Workflow Builder v2 (a no-code automation layer), and deepening Salesforce integration. Slack now does: messaging, audio, video, screen-share, documents, project management, automation, and CRM integration. Its original pitch was that it would reduce email. It has arguably become email.

The counterpoint (important)

Huddles reported 95% customer satisfaction in Slack's own research. They were adopted quickly and used by many. This is the hardest bloat to resist: the genuinely popular feature that nonetheless represents a strategic commitment you will maintain forever. Satisfaction is a measurement of usage; it is not a measurement of whether the feature belongs in the product.

The lesson

Every popular feature is a strategic commitment you will pay for in maintenance, design attention, and QA coverage, forever. Popular does not mean free. The question is not whether users like Huddles. The question is whether Slack-as-a-communications-suite is more strategically coherent than Slack-as-a-messaging-first-tool. That question requires an explicit strategic decision, not an accumulation of individually popular choices.

Sources: Slack engineering blog, Fast Company, InvGate. Anti-patterns: parity paranoia, no sunsetting

The bloat

Notion launched in 2016 and reached a $10B valuation in 2022. Its core architecture, everything-is-a-block, is genuinely elegant: documents, databases, kanban boards, and wikis are all composed from the same primitive. The architecture enables extraordinary flexibility. It also means the product accrues complexity in proportion to that flexibility.

Notion has added: AI (summarisation, writing, Q&A), Calendar, Mail, Projects, Forms, and integrations that position it as a full SaaS operating system. The product that once competed with Evernote now competes with Google Workspace. The blocks-based architecture, elegant for note-taking, becomes a navigation challenge when it must also organise a team's entire workflow.

The cost

Desktop and mobile apps have become slower as the feature surface grows. The writer Vaughan van Dyk described the experience: “Click a page and there's a couple moments' pause for it to load.” Hacker News threads from 2022 through 2025 document a consistent pattern of performance complaints. Reddit users regularly post about being overwhelmed by onboarding.

The counterpoint

Notion's power users love the maximalism. The ability to build a custom CRM, project tracker, and knowledge base in a single product is genuinely valuable to a specific user. Bloat for one user is a superpower for another. This is the hardest bloat to argue against, because the users who would leave if you simplified are vocal, engaged, and often the ones who write the glowing reviews.

The lesson

Flexibility is a feature. Too much flexibility becomes a job for the user. The question is whether the product's complexity serves the median user or the power user. Most products cannot serve both without explicitly segmenting them.

Sources: Vaughan van Dyk on Medium, HN threads 2022-2025, Appvizer. See also: notioncost.com

The bloat

iOS launched in 2007. The original Settings app contained approximately 15 categories. Modern iOS Settings has more than 40 top-level categories, each with nested sub-menus. Apple's own engineers have publicly joked about struggling to find specific settings. The Settings app has become the archaeological record of every feature Apple has added to iOS since 2007, without any subtraction.

The cost

Apple introduced a Settings search bar. It is useful. It is also the same structural acknowledgement as Microsoft's Ribbon: when you solve discoverability with search, you are admitting that the hierarchy does not work. The hierarchy does not work because it was never designed; it accumulated. Every iOS feature shipped with its settings in Settings. None was removed.

The lesson

Navigation is the canary. When your solution to complexity is “we added search,” you have bloat. Search is a navigation system for content that resists navigation. It is not a design for the settings of a consumer device. The lesson applies to any product: if users cannot find your features without a search bar, the feature hierarchy is broken. The hierarchy is broken because features were added without considering where they fit in the whole.

Sources: Daring Fireball, Apple HIG, r/apple. Anti-patterns: kitchen sink UX, no sunsetting