Two Tales of Empty Inboxes
I have a friend who runs an investor-backed online education company. He recently made an interesting change to his email setup. When you send a message to his normal address, you now get back an autoresponder that reads (in part):
“I appreciate you reaching out. I’m currently in hermit-mode creating as much value as I can for all of our stakeholders and having fun seeing if I can eliminate email from my life…Of course, if this is important, we’re here to help! Just email <address of a virtual personal assistant> and we’ll use our evolving email-free strategy to communicate.”
This extra step of re-sending your message to the assistant should add, at most, 10 extra seconds to the process of emailing this individual. Rationally speaking, therefore, it should have minimal impact on the number of messages that make it to my friend.
But this is not what happened.
As he reported to me recently, this additional step has “massively” reduced the amount of communication he receives.
Earlier this week, a reader wrote me with a similar tale. A computer programmer by trade, he setup a custom system that responds to incoming emails with a web form in which the sender can describe his or her purpose and needs in a series of text boxes.
Again, the extra effort of re-entering this information is minimal.
But the effect was significant.
His incoming message count reduced by a factor of 40. (He measured.)
The Danger of Zero
These stories caught my attention because they underscore a point I’ve seen gain traction recently: one of the core contributors to email’s nastiest impacts is its zero marginal cost.
Put another way, once we all made the necessary efforts to get setup with this technology, the cost of sending a message to someone became effectively zero.
This is important because we know from a variety of different disciplines that when friction in a system minimizes below a certain threshold, unpredictable and non-linear dynamics often follow.
I’m increasingly convinced that a similar effect is at play with email. Look at my above two examples: when even a small amount of extra friction was injected into the picture, email overwhelm essentially vanished.
The natural follow-up question to ask is how such friction could be added in a more general way. For example:
- A good friend of mine (who is a Silicon Valley venture capitalist) recently quipped to me that if he could charge people a small amount to send him messages, his life would likely significantly improve.
- Under the assumption that micro-payments are too complicated to engineer, consider instead a system that replies to any incoming message with a url and the note: “one hour from now this web site will list an address you can use to contact me”.
- Or, consider a system where a potential correspondent emails a triage address with the high-level purpose of their communication in the subject line, and either you or an assistant eventually responds with the appropriate channel to which to send the full message (if any).
I’m just brainstorming, but the general point is that almost any addition of friction to our digital communication systems might generate massive benefits — so tweaks of this type are worth considering.
(Though a necessary precondition would be overcoming our misguided obsession with “convenience” above all else.)
Bottom Line
It’s easy to lament that the “email problem” is intractable. But these stories give me hope, as they indicate that perhaps communication overload is not necessarily fundamental in a society with digital networks, but is potentially instead just a side effect of the particular and somewhat arbitrary way in which the tools we currently use are configured.
(Image by Bryan Alexander)
Why does Gmail annoy you with the “updates from your favorite sites” link if you manage to reach an empty mailbox? It is almost as if they think read email for fun/distraction..
I’m not convinced by your writings on email, and I think the volume of email being too high to be manageable is actually a problem which affects really very few people. I can think of many occasions where it is far more beneficial to me (as an early career Assistant Professor) to receive an email than it is to the sender to send it – conference speaking invitations, invitations to participate in special issues, invitations to take part in prestigious peer review panels, invitations to co-author grants. You think any of the people writing those messages are going to jump through the hoops you describe? They will just move on to the next person on their list. These don’t usually come from addresses I can pre-screen as being “acceptable” yet I need to receive these messages so that I can do at least some of them, and therefore to demonstrate my standing in the field and fulfil the criteria which I must meet to be given tenure.
You might counter this with your argument from Deep Work, that a tool being occasionally useful is no reason to persist with the tool. But these emails are critical to my future career and they will not come in by another route. Your suggestions in this case would just lead to me never being invited to do anything, and probably irritating a lot of people working in my field.
Surely what is more required is a more sustainable relationship with checking the emails you get, as you have talked about previously, rather than these elaborate hoop-jumping systems to act as barriers to people getting in touch with you, which frankly just seem like productivity time-sucks which are unlikely to actually achieve the desired effect?
I agree with this comment!
I agree… also because many people who are swamped by email are not in a position to drastically reduce it using the measures described. The other day, I was in a queue at a nationwide tool-hire warehouse and overheard an overalled manager complain to his colleague that he had 300 unread emails waiting for him. I can’t see this huge firm allowing their minions to hermitise themselves, or make it hard for customers to reach them. And as a customer, I would find this sort of behaviour annoying at the least and likely to make me turn towards competitors who are more willing to communicate.
A lot of big companies have in fact already applied some alike strategy: many use contact forms to reduce incoming mail (people are less likely to fill in the details of a contact form than write a free email) and reduce the need for a lot of replies (as all the information has to be submitted at first). Further, I see more and more, that company websites try to intelligently “understand” my question and refer to appropriate sections in their FAQ.
I agree with this statement. The average person isn’t in a position where people will be making efforts to offer them chances.
Also, one of the unstated reason that an auto-reply like that would decrease email volume is simply because you’re telling them that you’re not really interested. It’s not that it’s so much more effort, but you’ve already told them that you’d rather not get the email unless it’s important (which is fine).
Also, I send a lot of emails from my phone using GMail app and it’s a pain copy/pasting/forwarding emails through those systems, so I end up not doing it b/c it’s frustrating and those little blocks to one’s flow (figuring out the webpages) decrease my productivity.
+1
I’m curious – do you get these cold-call invitations via email often, presumably from people you seemingly don’t otherwise engage with?
And whilst people are really keen to have someone of your calibre part of their event etc they won’t call you (where you could give them a direct email address)?
This blog post is contextual. It applies to visible, sought-after individuals who get bombarded all day by people 99% of whom are offering little or no value to the recipient, and don’t mind much if they don’t get a response. If that’s not you, then you might not have an email volume value problem.
Definitely, people in high profile positions will confront high volumes of email from strangers.
I wouldn’t discount the distraction from email (or other ‘tools’ that allow others to control or manipulate where are attention is focussed) from people in non high profile positions either. In fact, I think people who aren’t of a high profile are more susceptible to falling victim to someone else’s agenda as they may not have as much pressure on them to act with efficiency in pursuit of their purpose.
Cal’s posts challenge us, and I think draw an initial defensive response – we’d much prefer to attribute these problems to the elite – but I think we benefit from asking ourselves how our attention is consumed, and re-asking, because the answers can be confronting. ‘Attention engineers’ are around every corner unfortunately, whether we want to believe it or not. Challenging ourselves rather than detatching from Cal’s profession is easier, but challenging our own assumptions is for the better.
That’s my experience anyway.
This comment is better than the post itself here, and holds up really well even close to a decade later.
Cal’s advice here is good for a person *who already has a personal assistant*, who is extremely high-profile to the point that the general public routinely harasses them, and whose biggest problem is distraction. But one of my biggest enduring problems with Cal’s work is that it’s hyper-specific to someone with his exact set of problems and often generalizes quite poorly.
Hope you’re crushing it with tenure and an empty inbox, wherever you are.
I think this tip is more suited to help employed, non-management level workers reduce the amount of personal email they get. I know that I would rather get a text with any relevant information from the people close to me. If you are in management, you could always have your assistant filter emails as needed.
I agree with Claire and the follow-up comments and disagree that there is “zero marginal cost” to sending emails. They take time and effort and, speaking for myself, are part of a necessary feedback loop.
For you, the person with less time, they take time and effort, but for the average random person from the public it’s very easy to send a hastily written email requesting something and creating a bigger time/energy request from person receiving the email.
Love it Cal – great ideas. I think another one: have a special email address you give to clients/bosses who you want to be quick in responding to. For everything else you use an email that triggers some kind of autoresponder (I know this isn’t an original idea – but how many people really do this type of thing).
I have tried to use a kind of “evolving email-free strategy to communicate” with my team (they were handling the e-mails for me). But, as far as I have seen, the effect of its implementation on the overall management of the workload as well as its impact on personal relationships was meaningful and far less than predictable. The workplace is a delicate ecosystem.
Regards,
I think your observation is sound—the ease with which email can generally be sent (or signed up for!) has drastically increase the amount we send and receive. But I agree with Claire and others that imposing artificial friction on your incoming messages seems as though it would cause many negative side effects. However, assuming our society will only replace email with something that is “better” (read: more convenient), living with those negatives might be necessary…
Nice idea and applicable for those who do not use email as the primary basis for their daily continual work.
Me, I’m in Technical Support and my work and information comes in primarily through email. Each and every email should be valued and prioritized at the moment it comes in. It may contain an issue that is more important than the one I’m working on now. Thus, I need to constantly switch from my work to my incoming email, even when working on issues: other more important problems may need my immediate attention.
Secondly, when new software is deployed it is emailed to us. This can lead to issues a few minutes later, in other sub-systems. When I don’t read these emails at the moment they come in, I might miss the point of other issues completely. Thus, it is vital to be aware of all updates at all times, due to the possible side-effects of deployments. Missing an email can harm a service (read: a customer’s revenue) for hours while chasing wind mills, only because I missed one email.
Of course deep focus is very hard when you can and will be interrupted at any time. But that’s the reason why I’m at this desk: I can manage this better than others. I was handpicked and I also volunteered for this job, from a batch of developers that in general would mentally explode if they had to multi-task as is required in my job.
So perhaps it’s nice to have a filter on your email, but that is clearly not applicable to me in the setting I’m in. It would be nice if Cal would address this type of work, to see how he would handle this. Any improvements or different ways to handle my work are very welcome.
Kind regards
I have started to implement several ideas from Deep Work, i.e., shifting email checking to the periphery of my workday and focusing more on blocks of time for deep work. It has already benefitted my productivity and adjusted my attitude about work in a positive direction, with no appreciable decline in service to my coworkers. I don’t know if I could implement these gatekeeper ideas that Cal has identified in my role, but just unplugging from the network for blocks of time serves this same purpose for me.
Thanks for the thoughtful post. Not every email control mechanism will suit everyone, but some ideas may have big impacts for some users. Please keep the ideas coming!
I’m a business analyst, so it’s my job to manage communications between several teams. Email is vital — and the bane of my existence. To keep it from consuming my day, I use a combination of filtering (low-priority emails go to folders where I can review them when the time is right), modeling good behavior (meaningful Subject lines with the name of the primary recipient; clear messages & requests), moving topics to wikis/documents/issue tracking tools (information becomes self-service), and gently punishing people who waste my time (refusing to give an answer until the sender provides enough information).
And — most important — I reserve the best hour of each work day, 10:00-11:00 a.m., to do my own Deep Work. Surprisingly, people are reluctant to schedule meetings during that time because they see I’m “unavailable”. I turn off instant messaging and don’t look at email during this time unless I need information for the task at hand. This small barrier has improved my ability to get important work done every day.
Interesting approach. My problem is that email is only part of the problem. Inboxes abound: Whatsapp, Linkedin, Messenger, Twitter, Slack, etc. I’ve been working on a solution for myself for all of them. When I crack it, I’ll come back and share it here : )
Cal:
Loved your book “Deep Work” but I am wondering whether you may want to do your blog less frequently so you don’t end discussing trivial, unrealistic “hacks” like so many of the other productivity hacks(!) out there. That also fits your style of deep work for true quality. Keep with what sets you apart, buddy. You are the best Cal N out there, not Tim F.
I agree.
Hi Cal. I loved Deep Work as well, and have written more in the two months since reading it than I did all of last year. So, thanks for that!
Also, I’d love to hear your thoughts on Peak, the new book by Anders Ericsson, and am sure others would as well. (One thing I found interestng is that you both see humanity more and more as being defined by deep work, with you calling the species Homio sapiens deepensis, and him, Homo exercens.)
After reading your book, I was thinking of similar ideas. I think the best way to manage this type of system is for the auto-responder to be conditioned on contact lists. Let me explain with a few examples:
Scenario 1:
You get an email from a new sender who is not in your contact list. This email gets archived automatically, and the sender gets an auto-response saying “Thank you for your email. I do not find you in my contact lists so I may take some time (or never) respond to your email. I only spend 10 minutes on email a day. You are welcome to post me a letter if you wish to be sure that I will read your message.”
Scenario 2:
You receive an email from someone in your list of urgent contacts. This email makes it to your inbox.
Scenario 3:
You receive an email from an email address categorised under the “friend” list. The email is archived and autoresponder replies “Hey. I am so bad at responding to emails. How about you give me a call after 5pm in the evening? ”
You could have as many lists and autoresponders as you like. What do you think of the idea? Does it already exist?
Email is definitely a problem, but avoiding it isn’t the solution. Being difficult to reach doesn’t mean that only the important stuff will come through, only the most tenacious. I’d rather make the decision for myself whether and how to respond than have someone that I would genuinely like to communicate with give up before they reach me. What is needed is an efficient, automated, easy to implement filtering system.
One potential solution for the average person is the use of “multiple inboxes”. This is a feature offered by Gmail, and likely by other email applications like Outlook. Using filters, you can route only the messages from senders you select (or alternatively, from everyone except those you “deselect” using a negative filter) to the appropriate inbox. I love the multiple inboxes feature because it sorts the possibly important from the mostly junk, and I never lose an important email in all the fluff.