Photo by Markus Spiske on Unsplash

Last Tuesday, I sat in seat 14B of a Southwest flight that had been delayed for two hours. The original delay was weather. The second delay was mechanical. By hour three, I wasn't angry anymore—I was curious. What would happen when I landed? Would I get compensation? An explanation? Would anyone care that I'd missed my connection and spent $200 on a hotel room?

So I did what millions of frustrated travelers do: I went to the airline's website and clicked the chat button.

Forty-five minutes later, I was still trying to explain my situation to a bot that kept suggesting I "book a new flight" for my cancellation. It never once acknowledged that the delay was the airline's responsibility. It never offered compensation information. It certainly never apologized.

This isn't a bug in the system. It's the feature.

The Deliberate Design of Avoidance

Airlines, hotels, credit card companies, and insurance providers have discovered something remarkable: if you make your customer service chatbots just incompetent enough, most people give up before reaching a human. It's not that these companies can't build better bots. It's that better bots would cost them millions in actual compensation payouts.

I spoke with a former customer service director at a major airline who asked to remain anonymous. "The metrics we tracked weren't about resolution," she told me. "They were about deflection rates. How many customers could we move through the system without actually helping them?" She left the position because the goal wasn't to fix problems—it was to make problems go away by making customers too exhausted to pursue them.

The math is brutal and cynical. Let's say an airline knows that 5% of passengers will complain about delays. If compensation averages $300 per complaint, that's real money. But if they can get 70% of those complainants to abandon the process before reaching a human agent? Suddenly they're looking at a massive profit margin on suffering.

The bots aren't broken. They're working exactly as intended.

Why Humans Are the Last Resort, Not the First

Here's what pisses me off most: most companies still employ actual humans in their customer service departments. These people sit in cubicles waiting to help. But the company has structured the experience so that reaching them requires navigating a frustration gauntlet.

Think about the last time you called a major company. You probably heard: "Your call is important to us." Then came the menu. Press 1 for billing. Press 2 for technical support. Press 3 for something that doesn't match your issue. By the time you finally reach a human, you've already wasted 10-15 minutes and your frustration level has doubled.

Some companies—looking at you, cable providers—have perfected this into an art form. Comcast's phone system is essentially a maze designed to make people hang up and complain on Reddit instead. It's a feature, not a bug, because Reddit complaints don't result in compensation, but conversations with employees sometimes do.

A 2023 study by the American Customer Satisfaction Index found that 67% of consumers would switch providers if they had a better customer service experience. Yet companies continue making it harder to reach service teams. Why? Because the short-term profit from avoidance outweighs the long-term cost of losing some customers. It's mathematically sound if you don't care about human beings.

The Complaint That Falls Into the Void

What makes this worse is the illusion of action. When you finally submit your complaint through the bot—or worse, through an online form—you get a confirmation number. You get an email saying someone will review your case. You feel like something has happened.

Nothing has happened.

Those complaints get routed to complaint management systems that are monitored by different people than the ones who can actually authorize compensation. Your complaint is filed away, categorized, and analyzed for patterns. But unless you're complaint number 1,000 in a category marked "systematic failure," nobody with authority ever sees it.

I tested this theory last month. I submitted complaints through three different methods: the website chat, an online form, and email. I got three different confirmation numbers. When I called the airline to follow up, the agent could find two of the complaints in the system but said the third one was "in a different queue." None of them had been reviewed. None of them even had assigned case numbers.

This is why so many people ultimately find success not by complaining to the company, but by complaining publicly on social media. A thread that gains traction is a liability that actually gets escalated to a human with decision-making power. The system isn't designed to help customers. It's designed to ignore them until they make noise too loud to ignore.

What Actually Works (And It's Depressing)

After weeks of testing different approaches, I found exactly three things that work:

First, be specific about regulations. Airlines are bound by Department of Transportation rules. Hotels are bound by state consumer protection laws. When you reference specific regulations in writing, your complaint stops being optional. It becomes a legal documentation requirement. Suddenly, the complaint lands on a compliance officer's desk instead of disappearing into the void.

Second, escalate through public channels. Call the company's corporate office instead of the customer service line. Send letters to executives whose names appear in SEC filings. These channels bypass the deflection machines entirely because they're monitored by people who report to the board.

Third, and most shamefully: make a complaint on Twitter and tag the CEO. I hate that this works. But it works. A public complaint that might reach investors is categorized as a reputation risk, and those get prioritized. The company will assign a real person to solve it within hours.

None of these methods should be necessary. The fact that they are is a referendum on how we've allowed service industries to treat complaint systems as profit centers rather than problem-solving tools.

Why We Should Be Angry About This

This complaint system design doesn't just inconvenience frustrated customers. It creates a two-tiered system where people with time, knowledge, and persistence get service while everyone else gets exhausted into silence. It's a way of making companies unaccountable to people who can't navigate their bureaucratic defenses.

If you've ever given up on a complaint because it seemed like too much effort, you've experienced this design working perfectly. You were supposed to give up. The system won. And the company kept the money they should have spent on compensation.

For more on how companies are structuring systems to take advantage of consumers, check out "The Subscription Graveyard: How Companies Are Banking on You Forgetting About Forgotten Memberships." It's the same principle applied to a different profit margin.

The next time you can't reach a human to complain about something, remember: that isn't an accident. Someone designed it that way.