What Counts as an Important Business Service

A service counts as important when its disruption would cause serious harm to customers, to market participants, or to the wider system, and that harm is judged from the users point of view rather than from the firms internal priorities. The question is not whether the service is large, profitable, or strategically prized. It is whether people who rely on it would be left unable to pay, to access money, to obtain cover, to settle obligations, or to carry on essential activity, and whether they have a realistic alternative while the service is down. A service can therefore be important even if it earns little, serves a modest number of people, or sits quietly behind a better known product. Conversely, a prominent product line is not automatically an important service if its disruption would cause limited practical harm and users could readily turn elsewhere.

Harm takes several forms, and a service may cause more than one at once. Financial harm includes money delayed, lost, or wrongly moved, and obligations missed because a payment or instruction could not be completed. Practical harm includes being unable to access an account, prove identity, report a loss, or obtain a decision when time matters. Wider harm includes disruption to other firms that depend on the service, disorder in a market that relies on its smooth operation, and loss of confidence that spreads beyond the firms own customers. Vulnerability and timing therefore belong inside the harm judgement, not as an afterthought to it.

Substitutability often decides a borderline case. If customers can use another channel, another provider, or another method without material loss or delay, the harm from one service failing is contained, even if the failure is annoying. If no substitute exists, or if the substitute itself depends on the same failed system, the harm is far greater. Substitutes must be realistic, not theoretical. A branch network that has closed, a telephone line that cannot cope with the volume, or a rival product that requires an application process lasting well beyond the disruption are not genuine alternatives for a customer who needs help now. The same logic applies to market users: an alternative settlement route that is untested, capacity constrained, or itself reliant on the disrupted firm offers little protection. Asking what users would actually do, step by step, during a failure is a better guide than asserting that alternatives exist in principle.

A product is what a firm sells; a service is the outcome a customer experiences when using it. A current account is a product, while access to the money in that account, and the ability to pay from it, are services. An insurance policy is a product, while handling a claim after a loss is a service. Internal processes, such as reconciliations or management reporting, support services but are not themselves services unless their failure directly harms users or markets. Keeping this distinction sharp prevents two errors: treating every product as a service, which inflates the list, and treating every back office process as irrelevant, which misses shared functions that several services depend on. The service definition should be written from the users perspective, in language that describes what the user can do, rather than in the language of internal teams and systems.

Borderline cases deserve the most careful reasoning, because they are where a list is most likely to be challenged. Write down the harm argument in both directions, note the users affected and the substitutes available, and record who decided and why. A decision to exclude a service should be as explicit as a decision to include one, so that a later reviewer can see that the service was considered rather than overlooked. If the evidence is thin, that thinness is itself a finding: gather user journey evidence, incident history, and dependency information before settling the question. Borderline services should also be flagged for early re-examination when circumstances change, since a service that is marginal today can become clearly important after a channel closure, a supplier change, or a shift in customer behaviour. This guide explains the test in general terms only; it is not advice for any particular firm.

Four questions, answered with evidence rather than assertion, usually settle the matter. Who uses the service, and what would they be unable to do if it stopped? How serious would that inability be, and for whom would it be most serious? What realistic alternative exists, and how quickly and completely could users reach it? Would the failure affect other firms, markets, or confidence beyond the firms own customers? If the answers describe serious, practical harm with no ready workaround, the service counts. If they describe limited inconvenience with a genuine alternative, it does not.