← back to paper map
Pass 2 Part I · classical playbook · chapter 04 Switching Costs

Switching Costs

Customers stay because leaving is expensive. Three distinct flavors that defenders confuse at their peril.

1. Definition

Switching Costs exist when a customer who would otherwise prefer a competing product stays with the incumbent because the cost of leaving exceeds the value of the alternative. The benefit is durable revenue at price levels above what a perfectly contestable market would clear. The barrier is that the challenger must compensate the customer for the switching cost — in cash, services, or time — before the customer will even evaluate the alternative on its merits.

Klemperer's 1987 economics paper formalized three distinct flavors that behave differently and respond to different attacks: procedural (the customer's staff must be retrained, processes redesigned), financial (cash outlay for migration, sunk integration costs, contractual penalties), and relational (vendor trust, escalation paths, the rolodex of who-to-call-when-something-breaks). Most products have all three; the dominant flavor determines the defense.

2. Historical deployers

3. The load-bearing assumption

Switching Costs require the cost of leaving to exceed the discounted value of the alternative, calculated honestly by the customer. That arithmetic depends on three sub-assumptions: the cost of leaving is real (not just inertia), the alternative is genuinely better (or the moat is feature parity, not switching), and migration tooling does not collapse the cost. When any of those breaks, the moat thins fast.

The deeper assumption is that the switching cost is paid by the same party that captures the value of staying. When the buyer (CIO) pays the migration cost but the user (employee) gets the value of switching, the moat is structurally unstable — the parties' incentives are misaligned, and the buyer eventually pushes back.

4. How it's deployed and won

  1. Identify which flavor you are running. Procedural switching costs respond to retraining tooling; financial to migration discounts; relational to channel competition. A defender who confuses procedural for relational fights the wrong attacker.
  2. Embed during onboarding. The deepest switching costs are the ones the customer pays for unconsciously — configuration, integrations, customizations — in the first year of use. The vendor builds the cost into the product surface.
  3. Multiply the integration surface. Each additional connector, custom field, or workflow tied to your system raises the migration cost faster than linear. SAP's real moat is the third-party ecosystem of consultants who built the customizations.
  4. Bind the relational layer. Make the vendor staff who own the customer relationship economically locked to your platform — through certification, partnership tiers, and exclusive features.
  5. Resist data portability standards — or, if forced, ensure they expose only the data, not the workflow logic that uses the data.

5. Classical failure modes

Visual: the three flavors of switching cost

Fig. 4.1 — The mix of flavors determines which attack works. Bloomberg Terminal SAP / Oracle ERP QuickBooks SMB Procedural keystroke memory Relational Financial Procedural Relational Financial $5–50M migration Procedural Relational accountant channel Financial Procedural-heavy → 10× UX rebuild collapses retraining time. Financial-heavy → migration tooling + heavy services discounting from challenger. Relational-heavy → capture the channel, not the customer. Confusing one for another is the most common mistake in defending switching-cost moats.

Cross-references

Switching Costs are the moat most often paired with Process Power (chapter 07) — the customer's organizational know-how about how to use your product is itself a switching cost — and with Cornered Resource (ch. 06) when the resource is exclusive integration rights. The hardest defender confusion is with Network Economies (ch. 02): a single user's difficulty leaving (switching) versus the value lost when many users leave together (network).

Sources: Helmer, 7 Powers (2016), ch. 4 · Greenwald & Kahn, Competition Demystified (2005), ch. 5 on customer captivity · Klemperer, "Markets with Consumer Switching Costs," QJE (1987) · Gartner ERP migration cost benchmarks (2024).