How data, workflow, learning, and relationships can make a provider costly to replace—and how that friction can erode.
Retention Has a Price
A customer may stay with a provider because the product is excellent, because a better alternative is unavailable, or because leaving would interrupt work. Those are different sources of retention. Switching costs are the expenses and disruptions of changing: termination fees, migration labour, retraining, new equipment, lost history, and the risk that the replacement will not work as promised.
The cost belongs to the customer, not automatically to the provider’s product. A payroll system may be replaceable in principle while years of employee records, integrations, approvals, and staff knowledge make replacement risky in practice. A hospital, manufacturer, or bank can remain with an imperfect system because a failed migration would interrupt a service that cannot be paused.
Four Different Frictions
Financial costs include termination charges, duplicate subscriptions during transition, new hardware, consultants, and the loss of credits or discounts. They are visible and can sometimes be subsidized by a competitor.
Procedural costs include exporting and cleaning data, redesigning workflows, retraining users, testing integrations, and running two systems during a cutover. A migration plan is evidence that someone has described intended work; it is not proof that every dependency has been discovered.
Relational costs include the loss of account knowledge, trusted service staff, supplier-specific routines, and informal problem-solving channels. A new provider can offer a similar feature set without inheriting the context built over years.
Risk costs are the possibility of downtime, data loss, compliance failure, or a wrong result during the transition. For a consumer app, a failed switch may be annoying. For a clinical, financial, or industrial system, the cost can be much larger than the contract price.
Why Digital Systems Can Still Be Hard to Leave
Software is often described as easy to copy because another customer can receive another instance at low marginal cost. That does not make the customer’s configuration portable. A system may contain proprietary data models, undocumented integrations, custom permissions, and workflows that exist only in the incumbent’s environment.
Portability rules can lower one part of the burden without eliminating the rest. The European Union’s Data Act includes obligations intended to facilitate switching and data transfer between data-processing services. The regulation is evidence of a policy response to switching friction; it does not establish that every workload can be moved without reconfiguration, downtime, or new security work.
Open formats and export tools can reduce technical dependence, while customer-specific integrations and accumulated operating knowledge can preserve it. The investor should inspect what can actually be exported, in what form, with what metadata, and whether the customer has people and time to use the export.
When Lock-In Supports Pricing
A provider with high switching costs may raise prices or reduce service before enough customers find it worthwhile to leave. That is a possible pricing buffer, not an automatic right to charge more. If quality falls, regulatory requirements change, or a credible migration tool appears, the tolerated gap narrows.
The provider may also have to keep funding the system that created the lock-in: compatibility, security, support, data retention, and migration assistance. A customer can be unable to leave and still be able to complain, reduce usage, delay expansion, or seek a substitute at renewal. Retention metrics should therefore be read with churn timing, support cost, price changes, and product investment.
Money and Timing Make the Difference
A smaller customer may recognize that switching would save money over five years but lack the cash to pay two vendors, consultants, and retraining costs this quarter. A large customer may have the money but lack a maintenance window or staff with the authority to change a regulated system. A competitor’s free migration offer changes the financial calculation but not necessarily the risk of data loss or service interruption.
Contracts determine when the customer can act. Auto-renewal, notice periods, data-return terms, service-level credits, and audit rights can make a planned switch reachable or postpone it for another year. The provider’s revenue is therefore partly protected by the customer’s operating calendar, not just by product preference.
How to Measure the Friction
- Retention shows that customers remained; it does not identify why.
- Net revenue retention combines expansion, contraction, churn, and price effects; it does not isolate lock-in.
- Export documentation describes an intended path; test whether it preserves usable data and permissions.
- Contract terms establish rights and dates, not the effort required to execute a migration.
- Competitor feature parity says little about workflow, reliability, and transition risk.
Switching costs become a durable advantage only when the provider remains worth staying with and when alternatives cannot cheaply remove the burden. The careful investor asks what is trapped, who bears the transition work, how quickly that work can be financed, and which standard, regulation, or competitive service could lower the cost next.
Inside CompanyGraph
CompanyGraph tracks one recurring-earnings shape live: net income carried by continuing operations and exceeded by operating cash flow, with depreciation passing through at the scale of a mature installed business.
Recurring Earnings Configuration
Continuing-operations is large or larger than total net income, OCF exceeds net income, and depreciation is large relative to OCF
The shape is consistent with retained, recurring demand. It cannot show renewal rates, contract terms, or the switching costs themselves; those live outside the statements.