Notes on "A World Without Email", plus my practical implementation

post by Peter Wildeford (Peter_Hurford) · 2022-06-20T15:34:53.335Z · EA · GW · 7 comments

This is a link post for


  The Four Core Problems
  Three Principles for Good Communication Interfaces
    (A) Tiered Reachability Principle – you should only encounter messages when either (a) you want to or (b) when they are truly unignorable and urgent. How and when people can reach you for each should be clearly understood.
    (B) Seamlessness Principle – Your system should be as seamless as possible for those outside the system. It’s on you to make your system work, not others.
    (C) The Minimalism Principle – We need to do less stuff.
  Solving Four Core Problems with Three Principles, in Practice
    Still check your inboxes a few times a day at scheduled intervals.
    Use the “one hand” rule and expect it of others (see below).
  When Communicating with Others: Use “One Hand” Rule
  The Biggest Problem? There is Too Much to Do and Track
  Time to Get an Assistant?
  A good tip - work backwards and design your day
  To Be Honest: I Don’t Actually Want a World Without Email

Cal Newport previously wrote “Deep Work”, which outlined a rule for success - you do your best work when you are in a large uninterrupted distraction-free block of time called a “deep work block”. Newport’s thesis is that a single uninterrupted 3-4 hour block of focused work on your most important tasks would pay far more dividends than 9-12x 20min blocks chunked throughout your day. (See my “Notes and Reflections on ‘Deep Work’ by Cal Newport” for details.)

This book didn’t revolutionize the workplace as much as Newport likely hoped, so the real question has become - why is it so hard to get deep work done?

It’s certainly not lack of awareness given how far and wide Newport’s book has spread, so there must be some sinister culprit. There are many ways to kill a deep work block, but the traditional villain has been meetings (see “Makers Schedule, Managers Schedule” by Paul Graham). But this has now been mostly addressed as well (at least within forward-looking tech and tech-adjacent companies) and reducing meetings doesn’t seem to have solved the problem much. So now what?

Newport has emerged with a new villain - email. And by email, Newport also means ubiquitous office instant messaging tools like Slack that have largely replaced email but not solved the problems. And Newport has also emerged with a new book to fight the villain - “A World Without Email”.

Again, as a reminder, when Newport says “A World Without Email”, you should also take this to mean “A World Without Slack” or whatever office chat you use. And therefore in these notes, you should always think of the word “email” as a shorthand for “email, Slack, whatever office communication platform you use, etc.” with the mostly but maybe not entirely correct assumption that they all share similar problems.

Also another reminder/caveat - in these notes I mainly aim to summarize what I find as the key takeaways of the book, from my understanding and as applied to my personal context (CEO of a non-profit), rather than try to present my all-things-considered view on how best to run a company.

Lastly, one more reminder/caveat that I think one’s personal context likely changes a lot about how you should approach this - managers need to be slightly more plugged in and I think independent contributors should consider dialing even harder into avoiding distraction than I would personally feel comfortable doing. You also may need to change your workflow depending on how collaborative your work is. But the basic problems will be very similar even if the constraints and solutions look different and context-dependent.

So with that in mind, let’s get into it.

The Four Core Problems

Basically, there are four core problems of email. I took the liberty to reorganize Newport’s book into concepts and give names to the concepts myself, rather than repeat the names Newport uses:

(1) the distraction-urgency trade-off – it is often incredibly useful to be able to get a hold of someone else instantaneously for them to respond to you about a question related to work but interrupting their work is often detrimental due to the negative effects of constant context switching. A message can unblock your work, but at the cost of interrupting their deep work. How does one solve this trade-off?

(2) the unavoidable trivialities problem - much of email is very trivial (e.g., random gifs and whatnot) that isn’t even worth interrupting your day, but it is very difficult when processing your email to tell which messages are high priority (and thus worth interrupting your work for) and which messages are low priority (and thus properly batched to a once-a-day scroll session or something).

(3) the FOMO problem - the feeling of not engaging with email triggers a fear of missing out (FOMO). This FOMO drives you to check email often and read everything, even when you really don’t need to.

(4) the overwhelm problem - the amount of work we have to do is just honestly truly overwhelming and we really just need to figure out how to individually do less stuff.

The interesting thing about email is that it is very hard to stop doing.

Newport writes that we’ve developed a hyperactive hive mind where we frequently switch attention between work and email (about work) and back again, and that this workflow has become so intertwined with work such that it cannot be easily removed. I think the premise of Newport’s earlier book “Deep Work” was that it wouldn’t be hard to just step away from email for very long periods of time, but this has proven harder than first suggested.

Much of “A World Without Email” is devoted to establishing the distraction generated from email as a problem, due to large amounts of context switching wrecking concentration, the typical worker getting overwh elmed with messages, and the typical worker being unable to make any “deep work blocks” happen. (When was the last time you truly worked for longer than two hours on something important without any email/Slack/etc-checking?)

However, a lot of this evidence is anecdotal and my thought is that really this book will either instantly resonate with you and your personal experience thus allowing you to know it is indeed a problem without needing any anecdotal evidence to point you in the right direction, or the book will not resonate with you and your personal experience and thus the anecdotes will not be convincing and thus this book is probably not going to contain as much value for you. So I don’t think there is any need to recap the evidence provided and thus I will skip all of this and just assume you and I agree that we know our attention is valuable, we need to protect it, deep work is important, and that there’s something about email that is getting in the way of things.

So, assuming this is a true problem, what should we do?

Newport’s answer is to pursue good interface design. Three principles can guide you in the fight against these three problems.

Three Principles for Good Communication Interfaces

The names of these are also names I gave to them to make for a better summary, not what Newport said.

(A) Tiered Reachability Principle – you should only encounter messages when either (a) you want to or (b) when they are truly unignorable and urgent. How and when people can reach you for each should be clearly understood.

(B) Seamlessness Principle – Your system should be as seamless as possible for those outside the system. It’s on you to make your system work, not others.

(C) The Minimalism Principle – We need to do less stuff.

Solving Four Core Problems with Three Principles, in Practice

So how do we actually do this? The trick is to figure out how to get tiered reachability and to do fewer things, while also being seamless. But current systems are poorly set up for this. For example, when doing email, you only have one inbox and it is hard to ignore new emails.

When doing Slack, things are a bit better - sure, other channels beckon to you but you can ignore them while playing in the channel you actually need. However, Slack gets some revenge through a truly terrible “threads” feature - threads are nice because they declutter channels, but they end up creating a new single inbox accessible by the “Threads” button where you load responses to every thread at once and you have to take action on all of them (either reading + responding, setting a Slack reminder to read it later, or marking it unread) or they will disappear forever, violating the Tiered Reachability Principle. (If any Slack developers are reading this, it would be awesome if the Slack threads buttons were channel specific!)

Here are some ideas on how to design a principles-compliant system out of Gmail + Slack:

Some changes to Slack that I really like:

My Slack workflow:

There may also be some needed mentality shifts:

There are also some more radical solutions to consider. I don’t currently do any of these but they are worth thinking about:

When Communicating with Others: Use “One Hand” Rule

A huge improvement I think could be made is ensuring that messages that are sent are more easily skimmable and more actionable. A big contribution to the overwhelm of messages is all the work you have to do to decide what is relevant and important to you and how to act on this. I think it would be a lot easier if people applied the one hand rule:

Imagine the person you’re emailing going from one meeting to the next and checking their inbox between meetings. Could they reply to your email while waiting for Zoom to launch? Could they send off a response in the time that it takes to walk from the bus or their car to the office? Or will they glance at the email, see that it requires a longer reply than they have time to type, and set it aside for later (which may never come)?"

This involves keeping questions to yes/no or multiple choice, make an initial recommendation / default, make everything clear upfront but provide background at the end as necessary. This link has some examples.

The Biggest Problem? There is Too Much to Do and Track

It’s funny to think of where we ended up. Meetings were the original problem breaking up deep work blocks, so we tried to push meetings as much as possible to emails that could be handled asynchronously. But then emails piled up too much and were hard to process due to only having a single inbox, so we built Slack which separated things into multiple channels and made multi-person communication much more seamless. But then Slack piled up and got too overwhelming, so we’re suggesting pushing things back to email again or actually doing more meetings to go through communication items synchronously rather than be constantly distracted by async communication? There’s something confusing here, like a productivity “whack a mole”. Newport confronts this, but I think he misses the overall irony of it.

Newport does eventually agree that the real problem is basically that we are just overwhelmed because our work is intrinsically overwhelming. Having the internet has allowed us to just be part of many more things and feel the responsibility to keep track of much more than was ever possible. It used to be that, back in the old days, we only had to interact with maybe 20-40 people at most, but now between large companies and larger outer webs of collaborators enabled by the internet, it is not unusual to be interacting with 200 or even 2000 people on a regular basis.

I think if you can properly adjust our roles and responsibilities this way and communicate clearly to those who need to know what is going on, we can just get a lot fewer messages and then the problem disappears? Could it really be that simple?

Time to Get an Assistant?

As Newport mentions, it used to be that professional “thought workers” had a decent amount of support staff that could do the administrative functions while allowing the “thought workers” to do deep work. Now, we expect the “thought workers” to do all the administrative work too. Getting an assistant could be a great way to help you with your workflows and avoid overwhelm.

Though we still should follow the “seamless” principle here - if a lot of assistants / operations staff are issuing a lot of different requests to staff members, it could create a lot of work even if it makes things easier for the requesters. Instead, operations staff should take it on themselves to streamline things and think about how to make their requests really easy to handle, even if it is more work for the operations staff.

If you can’t get an assistant, Newport mentions that you can separate your own workday into a “do phase” and a “support phase”.

A good tip - work backwards and design your day

To avoid spending too much time on communications, it helps to have a good picture of what you should be spending your time on. Take a 40hr work week and carve it up into how you think you ought to spend it and develop time budgets and time management from there. For more, see “How To Craft A Perfect, Productive 40-Hour Workweek”.

You could also consider maintaining a “deep work”-to-”shallow work” ratio, a “deep work” quota, or quotas for meetings or other activities. If you feel pressure to deviate from your quotas, you could explain why you maintain the quotas you do and ask them if they think it is worth changing your quotas. I’ve not personally tried this, but Newport suggests it and thinks that it will go over much better than just saying “no”.

From there, you can proactively plan your day, figure out when you should be accessible and when you should be doing communications, and execute accordingly. If you have things set up according to the principles above, things should go well.

To Be Honest: I Don’t Actually Want a World Without Email

Newport’s title is intentionally hyperbolic and controversial - I guess that’s how you attract attention and sell books[3]. But I don’t think getting rid of email is actually a good idea. I think if I literally did not have email (or Slack etc.), I would miss a lot of the camaraderie and I’d feel lonely. I also think we need the “urgency” side of the urgency-tradeoff - we still need a way to reach people, and it’s hard to do that without having at least one form of constant communication.

  1. Actually now I use a Google sheet, because my to-do list was getting too full of Slack tasks. ↩︎

  2. One thing that email did well that Slack didn’t do was subject lines. Zulip and Twist seem to do well because subject lines are making a comeback. ↩︎

  3. I agree “A World With Better Designed Email/Slack/Etc. Using More Clever Interfaces” is a much less catchy title. ↩︎


Comments sorted by top scores.

comment by david_reinstein · 2022-06-21T14:52:26.029Z · EA(p) · GW(p)

 Remove all Sidebar fluff. I don’t think anyone really uses these. (Preferences -> Sidebar -> Always show in sidebar -> uncheck all. 

Some of these are useful


Replies from: HaukeHillebrandt
comment by Hauke Hillebrandt (HaukeHillebrandt) · 2022-06-21T19:40:49.789Z · EA(p) · GW(p)

Thanks! Similarly, I'm enjoying which simplifies gmails design.

Replies from: Peter_Hurford
comment by Peter Wildeford (Peter_Hurford) · 2022-06-21T22:59:02.302Z · EA(p) · GW(p)

I didn’t mention it but I do use that actually

comment by Guy Raveh · 2022-06-20T21:04:13.364Z · EA(p) · GW(p)

(I am writing this comment after having only read the first ~10 paragraphs of the post)

In your post about "deep work" you wrote:

I came in already convinced that deep work is important so I sped through a lot of his arguments for it.

Having not read his book - is there actually good backing for his theory? Anything measurable that can show one kind of work style is better than the other?

I'm asking here because I think this could be relevant to why the book hasn't revolutionised the workplace.

comment by Hauke Hillebrandt (HaukeHillebrandt) · 2022-06-21T19:39:21.518Z · EA(p) · GW(p)

Newport's upcoming book will be on his ‘slow productivity’  philosophy which is:

  • Do fewer things 
  • work at a natural peace 
  • obsess over quality
Replies from: Peter_Hurford
comment by Peter Wildeford (Peter_Hurford) · 2022-06-21T23:00:02.760Z · EA(p) · GW(p)

Sounds exciting!

I’m curious how that will work for people who aren’t self-employed teams of one?

Replies from: HaukeHillebrandt
comment by Hauke Hillebrandt (HaukeHillebrandt) · 2022-06-22T09:17:01.914Z · EA(p) · GW(p)

You could submit it as question to his podcast!