Switching Costs

Switching costs are the costs of switching. This doesn’t just mean money. Time, learning effort, hassle, and the risk that something might go wrong also count. One example: You’ve filled your phone with photos, apps, and passwords for years. Switching to a device from another manufacturer would work, but it would cost you a weekend and maybe some lost data. This exact effort is the reason why many people stick with what they already have. Companies know this and often deliberately design their products so that switching becomes inconvenient.

Why high switching costs make companies valuable

For a company, high switching costs are a shield. A competitor can build a better product and still fail to win customers. That’s because the customer doesn’t just compare the products, but also factors in the cost of moving. As long as the advantage of the new provider is smaller than the effort of switching, everything stays as it is.

On the stock market, this is a central argument. Analysts speak of a moat: a lasting advantage that keeps competitors at bay. Companies with high switching costs can raise their prices for years without losing many customers. That’s why such business models are often valued higher on the stock market than those with easily interchangeable products.

But there is a downside. Whoever retains customers only through inconvenience retains them involuntarily. As soon as switching becomes easier, for instance through a legal regulation or a new data-migration tool, the customer base can shrink quickly. High switching costs are therefore an advantage, but not an eternal one.

What switching effort consists of

Economists distinguish several types. First, direct monetary costs: a contractual penalty, a notice period, the purchase of new devices. Second, learning costs: staff must master new software, which can take weeks. Third, data costs: customer data, documents, and settings must be transferred to another system. Fourth, psychological costs: you know the old one and don’t trust the new one.

It becomes especially effective when many things are interlinked. In a hospital, the lab, billing, and scheduling all hang off the patient software. A switch then doesn’t mean a new program, but a new nervous system for the entire building. Such transitions take years and cost millions. That’s why some systems keep running even though they are technically long outdated.

A common mistake is to confuse switching costs with network effects. Network effects mean: a service gets better the more people use it. Switching costs work even if you are the only user. Both can occur together, but they are not the same thing.

Switching costs in cloud, AI, and everyday contracts

In the tech sector, this topic is currently very relevant. Anyone who stores their data with a large cloud provider, meaning on remote data centers on the internet, can theoretically withdraw it at any time. In practice, the data transfer and reprogramming cost so much that companies stay put. Experts call this lock-in, meaning being locked into a provider.

With AI services, the picture is different. Anyone who has texts generated by a language model can often switch providers within a few hours. The requests look similar across providers. That’s why providers try to build additional retention: proprietary tools, stored company documents, customized models. Whether this succeeds partly determines how profitable these companies are in the long run.

You also encounter this principle in everyday life. When switching banks, all standing orders have to be set up again. With mobile phone providers, the contract term runs its course. Legislators often intervene here and deliberately lower switching costs, for example by letting you keep your phone number. When you read about switching costs in the news, it’s usually about exactly this dispute: companies want to keep them high, regulators want to lower them.