The Yellow Pages Still Arrives at My Door. Here's What That Says About How Products Die.
The Yellow Pages still lands on doorsteps. In 2026. Uninvited. Unread. Recycled before it gets opened.
Most people see this as a quaint inefficiency. A relic that someone forgot to shut down. It's not. It's a case study in why legacy products refuse to die.
Why dead products keep shipping
A product that has lost its purpose can still have a viable business model behind it. The Yellow Pages still generates ad revenue from local businesses who buy listings out of habit, or because their accountant flagged a line item, or because the salesperson called and they didn't want to say no. The print run is small. The margins are decent. The infrastructure exists. Why shut it down?
Every legacy enterprise product has this same internal logic. The SaaSpocalypse is AI exposing this pattern across an entire industry. Feature flags nobody can remember turning on. Workflows that exist because one customer in 2014 asked for them. UI patterns that survive because removing them would trigger support tickets.
The product is dead. The business model isn't. So the product keeps shipping.
The hidden cost of dead weight
The problem isn't that dead features cost money to maintain, though they do. The problem is that they make the living product worse.
Every legacy feature adds cognitive load. Users have to navigate around it. New designers have to understand it before they can change anything nearby. Engineers have to test against it. Product managers have to explain why it exists to every new hire.
Dead features don't just take up space. They slow down everything around them.
In SaaS products, this compounds over years. Each release adds features. Very few releases remove them. The product gradually accumulates layers of decisions made by people who aren't around to explain them anymore. This is what happens when project work makes you a stranger to every system you designed.
The harder design skill
Designers tend to focus on what to build next. The harder skill is figuring out what to retire. Most teams can't do it. They build on top of the corpse instead.
Retirement is harder than creation for three reasons. First, someone originally championed that feature. Removing it feels like rejecting their judgment. Second, some users, even just a few, might actually use it. The support tickets from removing it are visible; the cost of keeping it is invisible. Third, nobody gets promoted for removing features. The incentive structure rewards addition.
How to find what to retire
Worth checking your product for this. Not the features users complain about. The features nobody mentions at all. Those are usually the ones generating opportunity cost without anyone noticing.
Look at your analytics for features with near-zero usage. Look at your support tickets for features that generate confusion rather than complaints. Look at your onboarding flow for steps that exist because of features most users never touch.
The Yellow Pages will eventually stop being printed. So will most of the dead weight in your product. AI is accelerating this: if a competitor can rebuild your product in a weekend, the dead features aren't just slowing you down. They're making you replaceable. The question is whether you'll choose when, or whether the choice will be made for you.
---
