The 6 moats of durable brands

Why do some brands survive everything while others with better products die?

After years of running growth, I’ve landed on six things.

Three of these explain why people choose you. The other three explain why they can’t leave or copy you. Great brands have multiple moats.

The 3 pulls: why people choose you#

Pulls are preference engines. They’re especially useful in commodity environments.

1. Utility - It works#

Wispr Flow is a good example. Voice dictation was broken - they fixed it. That’s probably their main moat right now.

In the past, you saw a lot of utility moats in B2B SaaS. Companies paid for features. Zoom is the classic case: in 2020, it just worked when everything else didn’t. Then Microsoft bundled Teams into a product every company already paid for, and “it just works” stopped being enough.

That’s the danger here. In a commodity environment, utility is easy to substitute. The moment someone matches your utility and adds other moats, your customers will switch.

A lot of AI startups today compete on utility alone - better outputs, lower price per token - without building anything else. They’re easily replaceable.

2. Identity - It says who I am#

Supreme is a great example. A box logo tee has no utility edge and a thin story, but wearing it tells the world who you are. That was enough for VF Corp to buy the brand for $2.1 billion in 2020.

Apple sells an identity alongside the hardware: I’m a creative. Spotify’s catalog is basically the same as every other streaming platform, so they built Wrapped - a once-a-year identity engine that millions of people voluntarily broadcast. Identity works in B2B too. Teams pick Linear partly because it says “we care about craft.”

The danger here is that taste cycles. Cool decays, and identity without utility underneath leaves no functional reason to stay. VF Corp sold Supreme four years later for $1.5 billion - a $600 million lesson in how fast cool moves.

3. Story - It means something#

Patagonia ran a Black Friday ad that said “Don’t buy this jacket,” then its founder gave the company away to fight climate change. When you buy a Patagonia jacket, you’re buying into that.

Yeti started with two brothers who were tired of their coolers breaking on fishing trips, so they built one that wouldn’t. Before running a single ad, they got it into the hands of fishing guides and hunting outfitters - people who beat up gear for a living. Once the pros swore by it, that became the story, and it’s why people happily pay $300 for a cooler.

The failure mode: story converts attention once. The Pet Rock sold 1.5 million units in 1975 and was dead within a year. Pure story, pure spike. Story alone is a moment, not a company.

The pulls ladder#

One pull starts a brand. Two make it durable. Three make it Apple: it works, it says who I am, and the “Think Different” story is still paying dividends 30 years later.

The 3 barriers: why you’re hard to replace#

4. Switching costs - Leaving hurts#

Nobody wakes up excited to migrate their CRM. Salesforce holds years of your records, workflows and integrations. Leaving means pain, so leaving doesn’t happen.

Consumer products work the same way. Your social network holds your friend graph, which is why you’ve never switched. iMessage holds years of your group chats, which is why a lot of people will never buy an Android.

Notice the pattern: the strongest switching costs are the ones customers built themselves. They give you their data, their history, their habits.

5. Scale economies - Volume makes you cheaper#

Costco: every additional member lowers unit costs, which lowers prices, which attracts more members. Most of Costco’s profit comes from membership fees, not the stuff on the shelves. You can’t out-cheap Costco without Costco’s volume, and you can’t get the volume without the cheapness. The flywheel is the moat.

TSMC is the B2B example. A leading-edge fab costs over $20 billion and only makes sense at enormous scale, which is exactly why only a handful of companies in the world can make the most advanced chips.

6. Regulated monopolies - The government says only you can#

Some moats are handed to you. A patent, a license, a charter, a spectrum auction. Your local power company has no competitor because the law says it can’t have one.

This is the strongest moat on the list while it lasts. It also has the clearest expiry date.

Regulated moats leak. A NYC taxi medallion was worth about $1.3 million in 2014. Then Uber operated in a gray zone until the rules caught up, and within five years medallions were selling for under $200,000. The taxi companies spent decades defending the license and never built a reason for riders to pick them. When the rule stopped protecting them, nothing else did.

AbbVie saw its patent cliff coming and built a wall of 130+ patents around Humira. That bought years, but the biosimilars still arrived - Europe in 2018, the US in 2023.

The danger here: if the license is the only reason customers are with you, they aren’t really your customers by choice.

The barriers ladder#

One barrier slows competitors down. Two make you expensive to attack. Three make you Microsoft: companies can’t rip out Office and Azure, Microsoft spreads its R&D over billions of users, and compliance certifications like FedRAMP take challengers years to earn.

Pulls without barriers, barriers without pulls#

Pulls without barriers is a brand people love and leave. Customers pick you, but nothing stops them from picking someone else next quarter. Most consumer brands and a lot of AI startups live here.

Barriers without pulls is a brand people are stuck with and resent. Nobody loves their cable company. Nobody loves their legacy ERP. They stay because leaving hurts, and they’re waiting for a way out. When one shows up, they don’t just switch. They switch and tell everyone they know.

The brands that survive everything have both. People choose them, and it’s hard to leave or copy them. Apple is the obvious one: it works, it says who I am, it means something - and all your photos, messages and passwords are already in it.

Applying this to the AI age#

AI is compressing software to utility. A feature that took a team six months to build in 2022 takes a weekend now. If your only moat is “it works,” you’re one model release away from being a commodity.

So I think the question for any founder or growth leader isn’t “is our product better?” It’s “how many of the six do we have, and which one are we building next?”

Count yours. If the answer is one, you have a product. If it’s three or more, you might have a brand.