What can digital services learn from Singtel's latest outage?

About two weeks ago, Singtel's subscribers were greeted by an outage of fixed broadband services lasting more than 20 hours. This, understandably, caused much consternation, together with relief when the problem was resolved the following Sunday.

There are far more lessons at hand for those who are offering or planning to offer digital services, however. Customers affected by the outage, for instance, would have been unable to access vital business or work apps, whilst Internet of Things (IoT) devices connected to wireless hotspots would have ceased to function in the hours following the event. Such problems can cause thousands of dollars worth of losses, as well as potential liabilities (on the part of digital service providers) from having failed to provide a business-critical service to a customer.

This will depend, in part, on the contractual language and circumstances surrounding an incident. Network outages are typically beyond the control of most, if not all, digital service providers, and should not be provided for as part of a contract for digital services. However a digital service provider may, without knowing, make promises that cannot be reasonably fulfilled. This article will look at the type of obligations digital service providers should avoid, and the terms they should be aware of.

What is a digital service provider?
For the purposes of this article, "digital services" are services which are dependent on a connection to a cloud or server to function, with "digital service providers" being providers for the same. This includes software-as-a-service (SaaS) offerings and connected hardware systems (e.g. drones, fleet management systems), and to a lesser extent platform – infrastructure-as-a-service (PaaS, IaaS) offerings, which may offer service uptimes even where there are network disruptions affecting the servers on their end.

This, of course, should be distinguished from a customer's inability to access digital services through their own network service provider (e.g. Singtel, Starhub, M1), which should be excluded from a service provider's agreements and terms of use. This is obvious, but should be explicitly stated for the avoidance of doubt.

Again, the exact balance between what is offered and what is not will depend on the bargaining power between the parties – for example, a business-to-business (B2B) SaaS provider may be more willing to cave on issues of uptime and service levels than a business-to-consumer (B2C) provider – but for now, it should be noted that demands for higher service standards will vary according to the customer, market segment, and industry.

For instance, a consumer using a free app will be more willing to sit out service disruptions, whereas corporate customers may face losses if important business functions (e.g. sales, transactions) are disrupted. These issues will have to be accounted for when drafting a deal for a given customer, and, where possible, adjusted to a service provider's business needs.

Service level agreements
This brings us to the next step: drafting an agreement. This may be the result of a long negotiation process, or a fairly straightforward consultation with the necessary internal parties (e.g. customer service, finance, engineering). The latter is more common where a standard form is required – for example, the terms of use found on a website – whilst the former is typically reserved for large B2B sales.

Regardless of the method used, a digital service provider will have to decide if the customer requires a service level agreement (SLA) and the scope of the SLA, if any. This is in addition to the terms of use, which describes a more general agreement for services between a service provider and its user(s).

This is unlikely to be an issue where uptime is not a concern, but should be considered where customers have something to lose from the non-availability of a service or an app. For instance, restaurants with cloud-based point-of-sale systems may need guarantees that the service will be available during peak periods, while more system-critical functions (e.g. IaaS services) may require far higher service standards – up to 99.99% availability, barring pre-defined maintenance periods and exceptions.

Again, this will depend on the exact needs and bargaining power of the customer, but digital service providers should, for the most part, expect such terms to be part of the discussion depending on their industry and market segment.

Offline accessibility – a bugbear for cloud-based service providers
Finally, there is also the possibility that a customer's data or access is so important that the customer prefers an on-site solution. In such cases, digital service providers may have to offer on-site solutions, or lose the customer's business altogether.

This is why large enterprise developers (e.g. Oracle, SAP) have not lost their appeal – they continue to provide a vital link between today’s business needs and the needs of tomorrow. This is perhaps the greatest challenge facing smaller suppliers today – they often lack the deep capabilities or trust needed to convince larger corporates to take on their services and move important business functions into the cloud.

Such a challenge goes beyond establishing the right terms in a service level agreement or a contract, and into whether a firm is capable of competing in its given market segment.

Conclusion

To recap, a digital service provider's concerns are not only limited to poor network coverage and/or downtime – issues such as bargaining power and services which go beyond what a cloud-based provider can provide are relevant too. These are the problems which have to be addressed before service providers can expect to deliver vital digital services today.

Disclaimer: This article is for general information purposes only, and does not constitute specific legal advice or a solicitor-client relationship. Please seek the advice of a qualified practitioner in your jurisdiction before taking any action that might affect your rights.