Policy Practicum: COVID-19

How does general policy change in the face of a known health emergency? It’s a really tough question that can be fraught with pitfalls and emotions. So, this will be the very first of an intermittent series of posts covering the policy implications of things happening in real life.

First, a disclaimer

I feel the need to be really, super-specific here: I’m not pronouncing policy for systems that I am in charge of watching over. The purpose of this post is merely to give a series of considerations that lead me in a certain direction and belief that the world should work a certain way. So, as with the rest of this blog, these are merely my opinions and you should not use this post to decide that my current employer will treat any particular message in any particular manner.

Not much changes

It’s probably somewhat glib to start this way, but our baseline here is that nothing important changes for “emergencies.” Gmail isn’t suddenly turning off machine learning for its spam filtering solution. It isn’t rolling back it’s algorithms in order to create exemptions for emails mentioning “COVID-19” or “coronavirus.” Opt-in is still opt-in, opt-out is still opt-out, and the general rules that people have to abide by all still apply.

If it were any other way, then there would be a lot more truth to this Twitter post by an intermittently posting, humorous, marketing account:

But, things do change

On the other hand, just about everyone involved understands that in times of public crisis that material allowances need to be made. The question becomes what change is allowable and how long will those changes last?

Mailbox providers understand that they play a very important role in helping to get vital information to their subscribers. In fact, they understand that being perceived as “getting in the way” of the flow of that information could well be detrimental to their interests.

And, the more things change…

From the mailbox provider’s standpoint, however, there are two ways that they could “get in the way” of the flow of important information:

  1. Block information from coming in.
  2. Open the gates so far that information is buried.

The problem with an issue that presents like the current COVID-19 pandemic is that everyone wants to “reassure their customers.” That can lead to non-essential information overload:


There are literally hundreds of more examples that I could include just on Twitter alone. But these should suffice to show that email recipients are already starting to suffer from an overload of non-essential COVID-19-related information.

Policy in the time of Corona

This means that rational policy decisions need to be made based upon the actual importance of the message presented as perceived by the recipient. This is important to the mailbox providers as they play their role of gatekeeper, it is important to the brands sending the messages (as they would likely want to prevent the types of consumer reaction that we see in the Tweets above), and, finally, it has to be important to the email service provider who sits between those entities.

The most often asked question will be “can I send this message as a transactional or relationship message?” And the answer will be that “it depends.” Transactional or relationship messages are actually defined by 15 USC 7702(17)(A) of the CAN-SPAM Act ​(United States 2003)​, and it’s what my own team uses, in general, to inform our policy about what can be considered a transactional or relationship message.

So, what would count? Announcements about:

  • Shipping delays (i, via “facilitation”)
  • Reports of
    • office closures,
    • quarantines,
    • limitations of hours and/or services, or
    • potential exposure to infected persons reports (ii)
  • Changes or cancellations to trip itineraries (v)

And even these announcements need to have a level of sanity applied to them. If someone has not actually purchased anything, sending a notification that their non-existent shipment might be delayed should be out of bounds — the proper time to make that announcement is at the point of sale or if an unexpected delay occurs post-sale. And while it makes a lot of sense to tell cruise line customers that operations have been suspended for 30 days, it does not make a lot of sense to tell that to people who haven’t been on that line’s cruises at all in the last two years.

What probably doesn’t count here at all are the endless notifications that companies “are very concerned about you and our employees” and that they are now making their employees wash their hands and wiping things down with disinfectant. But, both are things that we hope were happening before this outbreak.

There can certainly be exceptions made to policies. Even those normally applied quite stringently. But those need to be extraordinarily exceptional in both cause and content. For instance, a local company may decide to send out a message to its list of contacts to help spread the word about a local disaster declaration and it’s ramifications for the customer — even if the company itself would suffer no harm and there would ordinarily be no logical reason to link the company to the content. The assistance to the greater public good can win the day (especially with excellent targeting and on-point messaging).

And so endth the lesson…

If you would like more information about the dos and don’ts of sending virtue-signaling messages during a time of crisis, let me point you to an excellent post by Augie Ray at Gartner ​(Ray 2020)​. It’s an excellent primer of what to do (and what not to do) from a brand marketer’s perspective.


  1. Ray, Augie. 2020. “Beware of Virtue Signaling or Outright Greed in Brand Communications About COVID-19.” Gartner for Marketers Blog. March 15, 2020. https://web.archive.org/web/20200422061123/https://blogs.gartner.com/augie-ray/2020/03/15/beware-virtue-signaling-outright-greed-brand-communications-covid-19/.
  2. United States. 2003. CAN-SPAM Act of 2003.
Mickey Chandler