In Defense of Email

One of the “first world problems” that I seem to hear most often, but understand the least, is people who complain about email: getting too much of it, how it’s a waste of time, interpreting it, the list of complaints goes on. Usually, the people making this argument are selling something (like, say, an enterprise chat solution) or relying on email for something it’s not as useful for (like a substitution for a verbal discussion). So allow me make the counterargument: email is great. Email was literally the first thing humans used networked computers for and it is the ur-example of how open standards truly power the internet.

Unfortunately, this is also why I am increasingly dubious about the centralization of global email platforms, particularly Gmail, which threaten this model for how email works.

Don’t get me wrong – I like chat too. And when I say “chat,” I’m really talking about Slack. I understand that the word “chat” has been banned at Slack, which prefers being called a “collaboration” tool, which it certainly is. Slack is hands-down one of the most innovative “enterprise” products around right now, and its potential has thus far mostly gone unrealized, especially in large companies. As I wrote about in The Three Types of Enterprise Software:

These tools are often hard to convince executive budget holders to spend on, because the business value they represent can be difficult to neatly articulate in the same way that a product recommendations widget is. But fundamentally, the way companies continually transform into producing higher value in a non-zero-sum fashion is by continually re-imagining their processes exactly like these, which is why each one of these products has taken off in the more innovative corners of our economy.

While there’s always room for improvement, Slack is, in my view, meaningfully superior to any other collaboration tool I’ve used in the workplace (and I’ve used many of them).

That said, it isn’t an “email killer.” Not by a long shot.

Most complaints about email come down to not using it properly. The large majority of email users don’t even use filters, which was probably a big reason why Gmail rolled out their default ones (“Social”; “Promotions” etc.) in 2013. Predictably, lots of people whined, but it also helped make email better for what most users want to do – communicating with direct contacts while also keeping tabs on email lists, social updates, etc. In the workplace, where the amount of email you receive seems to rise geometrically with the size of the organization, filters are a must. I personally have something like 15 set up in my Outlook, which makes my inbox perfectly manageable.

It seems to me that we drastically overlook the value that good email practice can provide, particularly in the workplace. Having worked at mostly bigger companies, this is especially clear to me. This works in a couple of flavors:

  • A clear, thorough, well-written email that takes 15 or 20 minutes to write can easily replace an hour-long meeting. In a business context where everyone’s time is literally money, the “this meeting should’ve been an email” phenomenon is a giant organizational value-leak.
  • Long email chains have a reputation for being a common antipattern that doesn’t lead to clarity or resolutions. The standard advice is that if an answer isn’t clear within a few messages, a verbal discussion is more efficient. This may or may not actually be true. Asynchronous responses may, in fact, be the most efficient use of everyone’s time, or participants’ contributions may not be immediately ready. Moreover, other recipients (those poor bastards on CC) may want the exchange available for reference, but not need to weigh in personally. (This, in fact, is a use case where a “chat” solution should swap in for email.)
  • Many emails really only require a line or two of response, with a minimum of preamble/conclusion. My emails have a reputation for sounding somewhat curt, and this is why. (If you’ve been on the receiving end of that, I promise, I’m not trying to be rude.)
  • Most emails require no response at all, and that’s okay.

Writing and responding to email is easy to mock as an activity unrelated to value creation, but this is an illusion. Knowledge work is tricky that way. Email can be absolutely fundamental to knowing what to create, or how. It can also be a tool of “performative” work that doesn’t add value. Like everything, it depends.

The future of email: open-ish

The open standards that the modern email system work on, like IMAP, SMTP and POP, are close to 30 years old. All major email exchanges operate with these standards, which is why sending an email from the U.S. to Kazakstan, or from Gmail to Shenzhen University, will always work. Open standards are one of the things that makes stuff “just work,” which is why they’re so great. They prevent one company from siphoning off all of the consumer surplus a technology provides and fuel an ecosystem of fundamental innovation. The poor state of the social web today is largely due to the absence of open standards for social media (or a viable way to distribute them).

For a long time, those open email standards are why internet users had a smorgasbord of choice when it came to email clients: Outlook, Thunderbird, Apple’s Mail, Eudora (!). The combination of web-based email and mobile killed a lot of those off, driving a lot of centralization to Gmail, Hotmail, AOL and the like. Even so, email’s open standards still make the encrypted email market possible, and enable collective spam-blocking heroes like SpamHaus to exist.

Gmail in particular has risen to become the dominant email provider today, with a little over 20% market share and far higher than that among the most lucrative user segments. (A common email marketing rule of thumb is that Gmail represents about a third of all recipients.) It is fair to say that Gmail has become a sort of a free public utility across the web, with deep adoption among developers and a whole constellation of third-party apps. This positions Google to do what no company has ever managed to really do before: to own the inbox.

When I put on my Product Manager hat, it’s easy to see the problems with open email standards. Not only are they really old, but they also impose strict limitations on what features you can offer – obviously, I can do whatever I like with my own email client, but who knows what my recipient’s client can support? This is a big problem for everyone except Gmail, who, in addition to controlling the biggest share of the email market, is also the basis for lots of companies’ email in G Suite. When you control both ends of the message, it’s possible to offer all sorts of enhanced features – like, for example, message rights management (i.e. preventing forwards, recalls or time-limited messages). With two-sided email control, the product possibilities are vast, as the booming messaging market bears out.

It’s no surprise, then, that Gmail is launching “Confidential Mode” in its recent update, along with several other product enhancements, including those I mentioned above. There’s no clear way features like that would work with a non-Gmail recipient, and I am dubious that Google will offer one.

I predict that these are the first of several Gmail-only end-to-end features that Google will release. In part, this is to encourage more people to sign up with Gmail, but it’s also to drive more engagement in Gmail itself. My guess is that most people in major markets who don’t use Gmail today aren’t considered that high-value to Google anyway. Gmail’s internal features will make the global email system’s standards somewhat moot. IMAP/POP will continue to be a standard, but they will stagnate as Gmail’s feature set accelerates, requiring more and more third-parties to use Gmail’s proprietary API for email.

This isn’t to say that everyone is going to use Gmail one day. But if Gmail gets to a third of all global internet users, and the closest competitor isn’t really close, isn’t that almost the same thing?

As for me, I still use Gmail. Of course I use Gmail – I don’t really feel like there’s a viable option. I’m as locked-in as anyone. As with social, a set of open standards to facilitate expanded email product features is completely feasible, but obviously unappealing to an industry giant with massive market share. I fear that IMAP/POP are not long for this world.


Related posts:


I send out a semi-regular update with the more interesting stuff I've read and written. Learn more about it and read old issues, or just sign up below. I won't spam you - promise.