Ch9 01: Tesla’s Best Quality Tool: Making Employees Drive Their Own Cars#

I drove a Tesla every day. Not because anyone checked — nobody did. Not as a perk — although it was a great car. I drove it because it was the fastest way to find out what was wrong with our product.

One Tuesday morning, running late for a meeting, I punched a destination into the nav system for the fastest route. The system sent me straight through a construction zone that had been there for three weeks. The map data was stale. I pulled into the office irritated — not at the traffic, but at the product.

That afternoon, I walked over to the navigation team and laid out what happened. Not as a customer complaint routed through a ticket system. Not as a line item from a survey. As a direct, first-person experience: “I used the product this morning, and here’s what went down.” The conversation lasted five minutes. The fix was prioritized that week.

If the same issue had traveled the normal feedback path — customer complaint → support ticket → product manager → engineering backlog — it would’ve taken weeks to surface and months to land on anyone’s priority list. By the time it reached someone with authority, it would’ve been one line among hundreds.

This is what “eating your own dog food” actually looks like in practice. And despite being one of the simplest ideas in business, it’s one of the least consistently followed.

Why it matters so much comes down to information quality. Every feedback channel introduces distortion. Customer surveys suffer from self-selection bias — the people who fill them out aren’t representative of your whole base. Support tickets get filtered through agents who paraphrase and categorize. Focus groups are artificial settings that produce artificial answers. Market research gets interpreted by analysts carrying their own assumptions.

Each of these tools has value. None is a substitute for direct experience. When you use your own product, you meet reality unfiltered — no categorization, no paraphrasing, no spin. The frustration you feel when something misfires is the same frustration your customers feel. The delight when something nails it is the same delight they feel. The feedback loop from experience to decision is zero — because the experiencer and the decision-maker are the same person.

At Tesla, we pushed this further than personal use. We built what I think of as a distributed sensor network — not electronic sensors, but human ones.

Every employee who drove a Tesla was a potential feedback source. But potential isn’t actual. Left alone, most employees would notice an issue, think “someone should fix that,” and move on. The insight would evaporate.

So we created a channel. A simple, direct pipe where any employee could report a product experience — good or bad — and have it land on the relevant product team’s desk within hours. No support tickets. No bug trackers. No middle-management filters. A short message describing what happened, sent straight to the people who could act.

The volume was enormous. The quality was extraordinary — because every report came from someone who knew the product deeply, used it daily, and could articulate exactly what happened and why it mattered.

This network surfaced issues that no standard feedback channel would’ve caught. Minor annoyances customers tolerated but resented. Interaction patterns that sailed through testing but face-planted in real-world use. Edge cases no QA team would’ve dreamed up.

But there’s a trap in self-use that’s easy to stumble into, and I want to flag it head-on. It’s called normalization bias.

When you use the same product day after day, its flaws go invisible. The nav system that occasionally sends you through a construction zone becomes “just how it is.” The touchscreen that takes an extra beat to respond becomes “normal.” The door handle that demands a specific wrist angle becomes “something you get used to.”

Your brain is built to normalize repeated stimuli. Efficient adaptation — it keeps you from getting distracted by unchanging features of your environment. But for product feedback, it’s a liability. The things you’ve stopped noticing are often the things your new customers notice instantly.

The antidote is competitive immersion. Regularly — I recommend monthly — spend real time inside a competitor’s product. Drive a competitor’s car for a week. Use a competitor’s app for your daily routines. Stay at a competitor’s hotel.

The goal isn’t competitive intelligence, though that’s a useful bonus. The goal is to reset your perceptual baseline. After a week in someone else’s product, you come back to your own with fresh eyes. Flaws you’d normalized light up again. Advantages you’d taken for granted feel impressive again. Your baseline has been recalibrated.

I made this a habit at Tesla. Every few months, I’d spend a week with a different car — BMW, Mercedes, Porsche. Not to copy them. To reset my eyes. Every time I climbed back into the Tesla, I caught things I’d stopped seeing. Some were problems. Some were strengths I’d forgotten to appreciate. Both kinds of insight were gold.

Guidance#

Build a three-layer feedback system in your organization:

Layer one: personal use (daily). If you’re a leader, use your own product every single day. Not occasionally — daily. Make it your default. When you notice something — good or bad — act on it immediately. Your direct experience is your highest-fidelity feedback source.

Layer two: team use (weekly). Set up a simple, direct channel for any employee to report product experiences. No forms. No categories. No approval chains. Just a message that hits the product team within hours. Drive usage by acting visibly on the feedback — when employees see their input leading to changes, they contribute more.

Layer three: competitive reset (monthly). Spend time inside a competitor’s product. Not to study them — to recalibrate yourself. Notice what they do better. Notice what you do better. And notice what you’d stopped noticing about your own product entirely.

The priciest market research on Earth can’t replicate the insight you get from using your own product and living through its failures firsthand. The data is free. The insights are instant. The only investment is the discipline to pay attention.

Use what you build. And listen to what it tells you.