Moats, and the ones that only look like moats

Here is the uncomfortable default: in a market with no barrier, profit gets competed away. Someone sees your margin, copies the thing, undercuts you, and the whole industry ends up earning roughly the cost of the capital it uses.

That is not a failure of the market. That is the market working exactly as advertised, and it is what happens to most businesses eventually.

Question seven of the nine asks what stops that. The answers are called moats, and there are about four of them.

The four that hold

Network effects. The product gets better for each user as more users join. The rival's identical product is worse on day one, not because it is badly made but because it is empty. This is the strongest of the four and the rarest, and it is frequently claimed by businesses that merely have a lot of customers — which is not the same thing at all.

Switching costs. Leaving costs the customer money, time, risk or retraining. Data lives in the system, four other systems talk to it, and the person who would run the migration is already busy. Note that this moat is paid for by the customer, not by you, which is why it erodes into resentment if nothing else is offered.

Cost advantage at scale. You can make it for less than anyone else can, because of volume, geography, an owned input, or a process nobody has copied. The rival can match your price and lose money doing it.

Intangibles. Brand, in the specific sense of a name that reduces a buyer's perceived risk. Also patents, licences and regulatory approvals — barriers somebody else erected and you got through first.

The four that only look like moats

Being first. Being first buys time, which is valuable, and time is not a barrier. Most first movers in most categories are not the eventual winner, and being first is usually expensive: you pay to teach the market what the category is and a later entrant sells into the educated demand.

Having the better product. Products are copied. Everything visible about yours is visible to them. A quality lead is a lead of months, and it must be re-won every year at full price.

A large and growing market. This attracts competitors. It is a reason to be in a market and an argument against your margins surviving in it.

A brilliant, hard-working team. Genuinely necessary and completely non-durable. Teams disperse. A moat that walks out of the building in twos and threes over four years is a temporary condition, not a structure.

Predict

A food delivery app has more restaurants and more customers in its city than anyone else. Restaurants list on it because that is where the customers are; customers use it because that is where the restaurants are. Is that a network effect that holds?

Choose one answer

Yes — each side makes the other more valuable, which is the definitionIt is a real network effect, but a weak one, because both sides can use several apps at onceNo — it is just a large market shareYes, and it strengthens with scale until it is unassailable

Lock in answer

Not the same thing

A moat

A structural reason a competent, well-funded rival cannot take your customers.

Measured in

Whether returns stay above the cost of capital over many years

For example

A system whose data is in a format three of the customer's other systems depend on.

Chase it alone and

Defend it too comfortably and you stop improving, which is how a moat becomes an obituary written slowly.

A head start

An advantage that exists because the competitor has not got there yet.

Measured in

Months until a determined rival catches up

For example

A better interface, a cleverer feature, an exclusive that expires.

Chase it alone and

Mistake it for a moat and you plan on a margin that has a delivery date on it.

Which is which? Put each one on a side.

• A hospital records system that took four years and a regulator's approval to install

  A moatA head start

• An eighteen-month lead on a feature competitors are visibly building

  A moatA head start

• Owning the only deep-water port on that stretch of coast

  A moatA head start

• A famous founder who attracts press coverage

  A moatA head start

• A payments network where merchants accept it because shoppers carry it

  A moatA head start

Moats decay

The uncomfortable second half. Every one of the four erodes, and they erode through their own success.

Switching costs breed resentment, and resentment funds the competitor's sales pitch. Scale advantages attract regulation. Brand degrades the moment the product stops matching the promise, and it degrades faster than it was built. Network effects hold until the network's purpose changes and everyone leaves for somewhere the conversation has moved to.

The practical version: a moat buys years, not permanence, and the years are only useful if they are spent on something.

Check

Which of these gives the most durable protection, given a well-funded rival?

Choose one answer

A patent on the core mechanism, with nine years left to runCustomers whose critical records are stored in your format and referenced by their other systemsA recognised brand and high customer satisfactionThe lowest prices in the category

Check answer

Worth remembering

• In a market with no barrier, competition drives returns down towards the cost of capital, and that is the normal outcome rather than the exception.
• The four moats that hold are network effects, switching costs, a cost advantage at scale, and intangibles such as brand or a licence.
• Being first, having a better product, a large market and a strong team are advantages, not moats, because a competent rival can neutralise each one.
• A network effect is weak wherever participants can belong to several competing networks at once.
• Every moat erodes, often through its own success, so it buys years rather than permanence.