What makes a web service or web site mission-critical in a natural disaster?

This is a surprisingly difficult question to answer.

For one, it depends on the sort of disaster and scale involved. A cyclone might damage roads and bridges and bring power and communication cables down over a large area but typically does not injure or kill a large number of people or does serious damage to solid buildings. A tsunami kills and injures and does significant damage even to well-constructed buildings, but generally over a much smaller area. Earthquakes do so over a wider area, potentially. Volcanic eruptions can have a variety of impacts depending on where they occur and what is there.

It also depends on where it happens: The deadly 2024 earthquake on Japan’s Noto Peninsula killed vastly more people than the 2022 eruption of Hunga Tonga Hunga Ha’apai, but while Japan itself was able to shoulder the vast majority of the relief effort out of its own resources, Tonga was strongly reliant on international support.

This all determines what is needed: Medical services to tend to the injured? Funeral services even? Heavy equipment to remove debris? Road, sea and air transport to evacuate the afflicted, to bring relief supplies and personnel in and ensure local mobility? Temporary shelter or housing? Expertise to deal with knock-on effects such as damage to chemical or nuclear plants?

Likewise, there are phases during and after a disaster when different services are critical. In the immediate aftermath, focus may be on search and rescue, frontline medical treatment, shelter, food and water. In early recovery, access to basic necessities to assist in cleanup and life continuity becomes important: hardware retailers, payment services, routine medical and dental services, access to equipment such as diggers, bulldozers, and materials for temporary structures and shelter. Later, focus shifts to services such as insurance or builders, as well as to economic recovery.

Some services are likely to be critical throughout: Communications, power, water, fuel, payment systems and logistics in general. There are also different types of criticality: Victims, responders on site and emergency management need local situational awareness. They also need to be able to communicate their needs to the outside world (outreach). Outside organisations involved in disaster response need to know what is going on in the disaster area (inreach) and need to be able to manage their resources in the field on site.

While most services in this day and age have some sort of a web presence, the criticality of this presence varies. It would be hard to argue that the official disaster response’s web presence is not critical. But who is likely to look up the web site of a petrol company before heading down to their local petrol station – which may or may not be affected? How important is Internet grocery shopping for survival after a disaster? Restaurants can feed a lot of people – but do they need their website to do so?

Likewise, media play a huge role in disseminating critical information in a disaster. Large broadcasters can reach enormous numbers of people across large areas, so are arguably critical services. But what about a local community newspaper that gets published once or twice a week. How critical is this? If it adds local detail that the big media organisations might miss, one could claim that it is a critical service. But what if it just publishes syndicated “news” of limited value and has no local reporting or editorial staff?

In many ways, it is easier to determine what is not mission-critical during a disaster or its immediate aftermath: porn and gambling are obviously not needed for a reasonable response, as are scented soaps, many wellness and beauty products, movies, tattoo parlours and fine wine retailers. Likewise art galleries, sports clubs (although their facilities can often be put to good use in a response), boutique fashion, luxury goods and classic cars.

But this still leaves a huge grey area. Plastic surgeons could be very useful in a response to provide reconstructive surgery for the injured. But it’s probably not the time for facelifts. Travel agents could be useful to organise evacuations or inbound travel for relief personnel. But package holiday bookings probably aren’t on the must-have list in a disaster. Churches might reasonably be expected to temporarily put evangelising on the back burner while their facilities and networks are often essential in a disaster response.

Any line one draws is bound to be in the wrong place.