Switching Costs as a Strategic Tool: How Hard Is It for Your Customers to Leave?

Issue 15

Welcome to Bill Gurley Notes. Together, we’re taking a step back in time to the early days of the internet and into the writings of Bill Gurley, the famed 6’9 venture capitalist and author of his aptly titled blog Above the Crowd. We started with his maiden issue and are working forward in chronological order. To follow along with the un-notated versions, check out my friend KG’s comprehensive collection of Gurley’s writings here.

Switching Costs as a Strategic Tool: How Hard Is It for Your Customers to Leave?

6/12/1995

“Open is a euphemism for commodity”

—ATC

A New Definition of Open

Frankly, we are nauseated by all the products in today's high-tech marketplace that are being marketed as open. It seems that if your product interfaces at all with the product of another company, you are allowed to call it open. We think the misuse of the term "open" has gone too far, and we would like to propose a new definition.

Our new definition of open is as follows: Products with zero switching costs for their customers can be called open, and those that have switching costs can be called proprietary. We also have a simple acid test to help delineate between open and proprietary products -- just ask the customer, "How much would it cost you (financially) to switch from Company A's product to Company B's product?"

We suspect that this new definition will not sit well with most technology executives. This is because it is quite apparent that products with zero switching costs are commodities. We also recognize that we have created a somewhat unfair argument by redefining "open" and then using that definition to imply commoditization. However, we will lose little sleep over this criticism for the following reason: If you make your product such that it easily interfaces with known industry standards (open), you also make it easy to replace.

• A return to a favorite topic of Gurley’s: switching costs. Increasingly relevant in the 1995 PC market with the rise of Intel’s motherboard strategy.

The Importance of Switching Costs

Our favorite strategy book (which applies extremely well to high-tech industries), Competitive Strategy by Michael Porter, has more than 11 references to switching costs in the index. And within the text, Porter states the following:

"In view of the potential importance of switching costs, the impact of all strategic moves on switching costs should be considered. For example, the presence of switching costs means that it is often much cheaper for a customer to upgrade or augment an already purchased product than to replace it altogether with another brand. This consideration may allow a firm with units already in place to earn very high margins on upgrading, as long as upgrading is priced properly in relation to the cost of the competitors' new units."

This matter is of critical importance for technology companies. Not only does product complexity lend itself well to implementing switching costs, but also the frantic pace of technological change creates wonderful opportunities for upgrade cycles. The existence of switching costs for both their true customers and ultimate users is what gives Microsoft and Intel so much power. It is also the one advantage that Apple has over vendors, such as Compaq and Dell. Lastly, switching costs can also be used to explain the unexpected, prolonged life of the mainframe business.

There is another reason why switching costs are of critical importance to high-tech companies. We will call this topic performance insurance. Let's say that you want to enter into a new business in which Company A is well established, and it will cost A's customer $X to move to your new product. That cost will ultimately be borne by you, the new entrant. To gain share, you must offer the customer a price/performance benefit significant enough to overcome the one-time switching cost. The higher the switching cost, the greater the new product's price/performance must be. This is why you continually hear about high-technology products that have dominant market share but are not "technically" the best product.

• This “better mousetrap != market dominance” phenomenon is the subject of his posts on the Microsoft anti-trust litigation.

Before we discuss how to raise switching costs, let us briefly state that we believe high switching costs are rewarded in the financial community by higher valuations, such as price-to-earnings multiples. It is generally accepted that stock prices represent the overall market's best expectation of discounted future cash flows. Therefore, one driver of high valuation is just how far into the future investors are willing to accept that a company can earn above-average returns on invested capital. At CS First Boston, we refer to this time period, as a company's competitive advantage period, or CAP. It should be quite clear that high customer switching costs increase a company's CAP, and therefore its valuation.

How Do You Raise Switching Costs?

So just how do you raise your customer's switching costs? Porter lists the following six sources of switching costs. Anything that can be done to increase one of these sources would obviously be effective.

(1) Costs of modifying products to match a supplier's product.
(2) Costs of testing or certifying a new supplier's product to insure substitutability.
(3) Investments in retraining employees.
(4) Investments in new ancillary equipment necessary to use a new supplier's products. (5) Cost of establishing new logistical arrangements.
(6) Psychic cost of severing a relationship.

Source number 4 is a frequently used switching cost in high-tech products. If you can encourage other companies to build complementary products around your proprietary design, you can go a long way toward raising switching costs.

Another way to raise switching costs is to make your customer dependent on your services. This can be done by taking on some aspect of your customer's business that is typically done in-house. Dell Computer has accomplished this quite nicely. Due to its build-to-order manufacturing system, Dell can custom configure software installations for every unit. For certain customers, Dell pre-installs customer specific software in its factories, so that the machine arrives ready to run. The use of Dell as an internal-software-distribution mechanism significantly raises customers' switching costs.

We also offer one other observation. Companies whose products are used by other companies in the normal execution of their business have extremely high switching costs. To switch products means to stop doing business for some fixed period. These expensive switching costs are partly responsible for the above-average price-to-earnings multiples of database-software companies.

Why Raising Switching Costs Is Difficult

Before you run out and try to raise your customers' switching costs (or invest in companies with potentially high switching costs), we offer a warning. Raising customer switching costs is difficult. In fact, while explaining sound purchasing strategies, Porter warns, "Good purchasing strategy, from a structural standpoint, involves the avoidance of switching costs." To put this in high-tech terms, companies prefer to buy open products; therefore, any attempt to raise switching costs may be met by customer skepticism that could actually hurt new customer sales.

As a result of this obvious Catch-22, it is crucial that companies wishing to establish switching costs be the first to market with innovative technologies. It is imperative that you "shoot ahead of the duck," when planning new products, as first-movers that establish switching costs will be brutally difficult to compete against.

Next time you invest in a high-tech company, why don't you ask the CEO, "What have you been doing to raise your customer's switching costs?" We would love to hear the response.

Takeaway

In this post, Gurley takes a step back from the news of the day and offers an explainer on one of his favorite topics: switching costs. And although not pegged this time to a specific market event, it’s clearly related to his thinking on the increasing commoditization of the PC industry, a topic he’s covered at length (with his posts on Intel’s motherboard strategy and the ubiquitous PC Feature Wars).

This post answers a question that’s lingered in the background of his analysis: what’s a tech operator to do in the face of commoditization? There are two sides to the coin: (1) for new entrants trying to overcome switching costs in the adoption of their products, Gurley offers a rule of thumb—the cost/performance of your solution must overcome the switching costs for customers to leave incumbent players. A high bar.

And (2) for operators looking to increase switching costs, Gurley, courtesy of Michael Porter, offers six methods of doing so, including the left-field approach of encouraging peripheral equipment and software makers to build a range of products that work exclusively with yours. He adds a seventh: offer services that lead your customers to rely on you. The example here is Dell, which can pre-configure computers for customers, obviating the need for those customers to maintain as big of an IT staff as they might otherwise. This engineers dependency.

This last point is remarkably astute—16 years later we see a phalanx of extremely valuable “startups for startups” that have operated on this dependence model—like Stripe, whose IPO, if executed has a chance be the most valuable of any tech company in history. In lowering the barrier to startup entry, these companies’ offerings also create this kind of organizational dependence in customers—and raise switching costs as a result.

But Gurley also touches on the dark side of switching costs: customers are always looking to avoid them.

Therefore, the optimal choice is to be a first mover. That way, customers have no choice but to accept your product, switching costs and all. Reminds me of the great Arthur Rock who, asked for advice for his younger self, replied: “be lucky.”

Be lucky—or make your own luck by entering a market into which you can be the first mover.

Until tomorrow,

DS