The term “Moat”, coined by Warren Buffett, refers to the ability of a business to maintain competitive advantages over its competitors in order to protect its long-term profits and market share.

As a self-proclaimed long-term investor, a critical component of my investing process involves spending lots of time thinking, analysing, and forming expectations about A) companies that have the potential to develop a wide-reaching moat or B) companies that can sustain and grow their moat.

The investing process then involves reconciling expectations of what a business can deliver in the future to current stock prices. Because excess returns are derived when tomorrow’s outcomes are different from today’s perception, leading to a revision in expectations.

When I find that my expectations differ from the market’s perception of the future in terms of magnitude or direction, it represents an investable opportunity.

Here’s how I think about moats, courtesy of Pat Dorsey.

Moats – First Principles

Before we delve into the types of moats, let’s take a step back and go back to first principles and define what a moat entails.

A moat by design is structural and sustainable.

Structural in the sense that it is inherent to the business and does not relate to a specific product. Consider Disney, the association of any content with the brand Disney confers it an enchanting spell that magically draws an audience of all ages. Remove the brand Disney from the movie Frozen and we might live in a radically different world where we don’t hear young girls humming to the tune of “Let it go~”.

Sustainable meaning that they are likely to persist well into the future. Think about Microsoft OS, it is so highly ingrained in our daily lives that it is difficult to imagine a world, 5 to 10 years down without it. Today I drafted this post in Microsoft Words and 10 years down the road, I’m highly certain that I will still be drafting my posts on Microsoft Words.

Moat 1: Intangibles Assets

Intangible assets do not have a physical form, but they produce value. Examples include brands, patents and licensing that make it difficult for a competitor to simply replicate and recreate the offering.


Brands are a powerful source of moat but not all brands are cultivated equally. The most powerful brands create an emotional appeal that draws consumers back to them consistently. Rolex for instance, bestows an elevated positional power. Wearing Nike basketball shoes makes you feel like you are capable of pulling off a 360-windmill dunk. And for the ladies, you know how good you feel strutting down the street with a LVMH handbag and Gucci shades.

However, a common misconception that investors have is associating well-known brands with competitive advantage. In the words of Pat Dorsey, a brand can be considered a moat only if it increases the consumer’s willingness to pay, lowers search costs and attracts them to buy its product again.


Patents by design create exclusivity, preventing competitors from duplicating a business’ offering. They are effectively legal monopolies but are subject to either expiration or piracy. The way you want to think about businesses that operate with significant intellectual property is to assess their portfolio of patents. A firm that owns a portfolio of patents regarding has a more robust moat compared to a firm that relies on a single patent to generate revenue.

Qualcomm for instance, owns a patent portfolio that is amongst the largest and most extensive in the industry, with the company reporting more than 300 patents essential to the development of LTE, 3G, 4G and 5G technology. Any semi-conductor looking to develop 5G chips is legally required to pay Qualcomm a licensing fee.

Contrast that with a pharmaceutical company that has 90% of its revenue attributed to a single blockbuster drug that is highly susceptible to irrelevancy once another ground-breaking drug is being developed.

Moat 2: Switching Costs

Businesses with high switching costs are better able to retain existing customers and have significant pricing power.

Twilio, a communications API firm, is a great example of a business with high switching costs. As firms such as Uber built out their communications platform using Twilio’s APIs, Twilio’s offerings become increasing integrated with the firm’s operations. Switching from Twilio to another provider results in significant disruptions to existing operations as well as a requirement to retrain employees. The longer a firm uses Twilio’s software, the greater the inertia to switch. Only a new offering with significantly better features will tip the scales in a cost/benefit analysis to displace Twilio’s software. 

The idea when it comes to switching cost is to find businesses with products that are highly entrenched in their customer’s business. Replacing such products is akin to ripping out an organ from within your body. Other examples include Bloomberg(financial software), Microsoft Office product suite, Adobe Photoshop etc…

The way I think about it, is to classify switching costs into 3 broad categories:

  • Monetary costs: Financial fees associated with switching i.e., exit fees & installation costs
  • Psychological costs: Experience of discomfort or a perceived loss of status associated with a known brand or product.
  • Procedural costs: Disruptions to day-to-day operations and the effort it takes to set up and retrain employees.

The root of switching costs dials down to a simple analysis between the cumulative costs and benefits of switching.

It is also noteworthy that switching costs can be a derivative of branding or network effects. Power brands and strong network effects discourage customers from switching.

Moat 3: Network Effects

To assess whether network effects exist, just ask yourself this question

“Does each additional user that enters the network adds value to existing users?”

If yes – network effects exist.

Network effects are literally all around us. From the likes of messaging apps like WhatsApp, food delivery apps like Doordash, mobility apps like Uber, to marketplace platform such as Amazon, Shopee and Carousell.

However, just like brands, not all network effects are created equal. Some are by design, stronger than the others, just like Airbnb’s global network effects vs Uber’s local network effects.

A16z has a great primer that drills down into the types, strength, and strategies regarding network effects. You can read more about it here: All about Network Effects – Andreessen Horowitz (

If you’re short on time, here’s a quick summary below:

  1. Homogeneous vs Heterogenous network effects
    Homogeneous: Skype, where most network effects are derived from a single class of users, people who want to chat with one another.

    Heterogenous: Marketplaces have a 2-sided network effect, wherein the network becomes more valuable as the number of users on the other side of the marketplace increases. For instance, in a rideshare marketplace, users derive more value when there are more drivers, and vice versa. (N-sided is also possible in food delivery: Restaurant, Riders, User)
  1. Hub and Spoke vs Clique networks
    Hub and Spoke: Here, all nodes connect directly back to the central node. Consider operating systems such as Microsoft OS or the Android / Apple app store

    Clique network:  Nodes form connection with each other. These are your social media platforms such as Instagram and Facebook.
  1. Global vs local network effects
    There’s a big difference between network effects that span global and continental regions, as opposed to city-by-city networks.

    Global network: Airbnb’s global network effect spans regions, since potential guests book lodging outside their home base and hosts expect to receive visitors from all over the world. It is a single, global network, which provides a strong defensible moat that, thus far, competitors have struggled to copy.

    Local network: Doordash, Grabfood, Ubereats etc.. have local networks effects, meaning that transactions within their 2-sided networks exist within specific cities, not across them

Note: having a ubiquitous presence does not necessary translate to network effects. Using another of Pat Dorsey’s example,

Western Union likes to say, ‘We have the most locations in the world to send money from one place to another.’ Well, that’s true, but each of those locations doesn’t add value to all the others.”

Moat 4: Cost advantages

When it comes to Cost advantages, there are typically two sources: Process or Scale

Process Cost Advantage

A cost advantage stems from a superior process that employs cost innovative methods, deviating from conventional methods. The classic example would be Ford, who first introduced the concept of a conveyor belt system in an assembly line. Prior to production of Ford’s Model T, cars were produced in an artisanal fashion which were highly time consuming and expensive. Ford introduced a system where the chassis of a vehicle was towed by a rope that moved it from station to station to allow workers to assemble each part. As a result, the first Model T cost $850, half the price of existing automobiles in 1908!      

Scale Cost Advantage

Econs 101: Economies of scale, bigger businesses enjoy a distinct competitive advantage because they can offer a lower price and still be economically viable.

Wal-Mart is a great example of a business that enjoys scale cost advantage. As a dominant retail player, the volume of goods that Walmart sells, provides it with enormous economies of scale and bargaining power, enabling it to cement its status as a low-cost retailer.

Scale cost advantages are typically present in industries that require high capital investments, (such as semi-conductors) or/and products are commodities in nature. Business in these industries have high operating leverage.


As I have touched upon earlier, investing is all about reconciling your expectations against the market’s expectations.

Identifying moats is the easy part.

Forming expectations about how these moats develop overtime – this is the hard part.

Facebook, for example was not built with the intention of developing a 2-sided global network effect. Mark Zuckerberg created Facebook because he wanted to start a Harvard directory, where students could look up one another – a one-sided local network. However, as Facebook grew internationally and people started connecting with each other, it planted the seeds of a ubiquitous moat arising out of 2-sided global network effects.

Personally, I ask myself these questions to attempt to piece together a view on how a business might play out in the future.

  1. Does the business have a moat or have the potential to develop a moat?
  2. How strong is the moat?
  3. Is the management able to monetize the moat?
  4. What plans do the management have to grow their moats?
  5. Do their capital allocation decisions reflect their strategy?

Obviously one can delve much deeper into the nuances of specific moats and the appropriate strategy. But I believe the above serves as a generic framework to get one thinking about moats.

If you have enjoyed my writings and wish to be updated on future posts, do enter your email address on the “Get Updates” box on the top right hand corner!

Here are some articles that you might enjoy as well!
The tech company behind a retail façade – Stitch Fix
Are you thinking about software companies’ metrics correctly?

You can also reach me at
See you around!

You can also reach me at
See you around!

Related Posts

Leave a Reply

Your email address will not be published.