The Meeting That Should Have Been an Email — And the Email That Should Not Have Existed

The complaint that a meeting should have been an email is so common it has become a genre of workplace humour. Less commonly discussed is the inverse: the email that should not have existed at all, because what it was attempting to accomplish through asynchronous text was something better handled in a brief synchronous conversation or handled by a system rather than a person. Both failures are expensive, and both are symptoms of the same underlying problem: organisations that have not deliberately designed how communication happens default to the path of least resistance, which is consistently neither the most efficient nor the most effective path.

What meetings actually cost

A one-hour meeting of six people costs six person-hours, not one. In an organisation where person-hours are priced at even a modest rate, a meeting of six senior people costs substantially more than it appears on any individual's calendar. Multiply this by the frequency of meetings that deliver less value than their cost, and the aggregate is significant. Most organisations, if they calculated the actual cost of their meeting load and assessed that cost against the decisions made and work produced, would find the ratio unfavourable. The cost is not only financial. Meetings consume attention as well as time. The preparation for a meeting, the context-switching required to move in and out of deep work, and the time required after a meeting to return to where you were before it — these are real costs that do not appear on anyone's calendar but accumulate across an organisation into a measurable drag on the quality and volume of deep work.

When a meeting is the right medium

Meetings are the right medium for a specific set of purposes: collaborative problem-solving where the problem is ambiguous enough that structured dialogue produces something that no combination of individual thinking would produce. Decision-making where the decision requires the active participation of multiple parties who need to engage with each other's perspectives in real time. Relationship-building where the purpose is to develop the connection between people rather than to transfer information. And conflict resolution where the emotional dimension of the situation requires presence rather than text. A meeting called for any other purpose — to share information that could be read, to provide an update that could be written, to gather input that could be collected asynchronously, to make a decision that one person could make and communicate — is a meeting that is using the wrong medium. The cost of this mismatch is the time of everyone in the meeting, subtracted from the work they could otherwise be doing.

What email actually costs

Email is routinely underestimated as a cost. A message that takes three minutes to write and three minutes for each of five recipients to read has consumed eighteen minutes of collective time. A reply chain that extends to ten exchanges, each involving multiple recipients, has consumed hours. The hidden cost of email is not in any individual message but in the aggregate volume, the constant interruption of checking and responding, and the cognitive overhead of maintaining context across dozens of simultaneous conversations. The emails that should not exist are the ones attempting to handle in text what would be faster, clearer, and less disruptive if handled differently. A clarifying question that generates a back-and-forth chain is better handled in a two-minute conversation. A status update that generates questions is better handled as a dashboard that people can consult when they need the information. An approval request that requires context that cannot be conveyed efficiently in text is better handled in a brief call. The discipline of asking, before writing an email, whether this is the right medium for this communication produces a measurable reduction in volume and a measurable improvement in quality.

How AI changes the equation

AI changes the communication equation in two specific ways. First, it handles a category of communication that currently occupies a large portion of many people's email time: routine information requests, standard updates, and frequently-asked questions that could be answered by a system rather than a person. When this category is handled by an intelligent system that people can query directly, the email that would have requested the information from a colleague or manager simply does not get sent. This is not a marginal efficiency gain — in organisations where internal information requests are a significant portion of email volume, it is substantial. Second, AI assists in drafting communication of sufficient quality that the email or message that does need to be sent is better written, more concise, and less likely to generate clarifying follow-ups. An email that is drafted clearly and completely is less likely to generate three replies asking for the information that was not included. This reduction in reply chains reduces volume for everyone involved. The combination of these two effects — handling what should be handled by systems and improving the quality of what must be handled by people — is where AI creates tangible and measurable improvement in the communication overhead that most organisations carry without examining.

For further reading on this topic, check out our guide on [How to manage ESIC claims and ensure your employees can access benefits](/how-to/how-to-manage-esic-claims-and-ensure-your-employees-can-access-benefits).

For further reading on this topic, check out our guide on How to build a learning and development programme on a limited budget.