Don't Suck At Email
markosullivan.ca
markosullivan.ca
his method: click "unread" P times
yours: click "archive" A-P times
A-P >> P
His method is less work.
In triage method A (his method), you first page through all your unread mail (1 click/keystroke to open the first, then A - 1 clicks/keystrokes to advance to the next conversation without returning to the inbox, and 1 final click/keystroke to go back to the inbox.) Then, once you've seen them, you select the mail you want (P clicks requiring dexterity, or A+P keystrokes; it's a compromise) and mark them as read (1 keystroke.) M_a = O(A + P) = O(A). (Because, as you said, A >> P.)
In triage method B (the "naive" method), you page through all your unread mail, and at each step, you either advance to the next conversation, or archive the current conversation and advance to the next conversation. (1 click/keystroke either way.) M_b = O(A).
Seems like to many steps to me.
As I remember it, GTD does suggest having multiple to do lists based on where you would do each type of task. I think having your emails to do list in your email makes sense. If responding to the email requires you to do a lot away from the computer, than it makes sense to put it on your other to do list.
Also, I'm not sure I'd like to have email subjects, etc... on my to do list. I don't usually keep my list that granular since a lot of the tasks I need to do are pretty amorphous until I actually start on them and figure things out. Maybe its not the best way to do it though....
Instead, I flag emails I want to respond to/come back to soon. Also, any emails either a) directly to me (i.e. not to a mailing list I'm on) or b) from my manager(s) gets automatically flagged. Once a day I go through and clean up all my flagged emails.
If you communicate with someone who only bothers to skim your email (tech support and sales people rarely impress here), or with people with limited reading comprehension for any other reason:
State that one question three times throughout the email, each formulated in a different way but with the exact same meaning.
Your comment is by no means a minor addition to #2, but more of a refutation.
I provided a special (but annoyingly common) case where it is useful to diverge from the rule. The case is where there is a low probability that the reader will understand any given sentence.
Given that the reader has an estimated 60% chance to understand any given sentence, if you repeat the content in three different sentences, you get a 94% chance that at least one of them will be understood.
As an actual example, I give a shortened transcript from a recent discussion with PayPal support. My first 3 emails were all variations of:
I keep getting "The card cannot be used, try again
later" when trying to pay. This has been going on
for months. What should I do?
2-3 sentences and one question; they were completely unable to understand any of it (likely because the support people are paid by the number of tickets they reply to and just skim any incoming case). I estimated their reading comprehension hit rate to about 30%, which means I had to fire at least 7 explanations at them to get a >90% chance of success. This is my 4th email to them, which was rewarded with an actual helpful answer: Hi. Thank you for the answer, but since you also
recommend "just try again at a later time" it seems
you are missing a vital point I'm trying to make:
I have been "trying again" for months.
I have been trying again weeks later several times.
This has been going on for months.
As you recommended, I tried repeating the process at
a later date, but this did not help.
[skip 5 more reformulations]
Please let me know typically how long this "security-
related" issue takes to pass through PayPal's bowels.
How much "later time" do I have to wait to try again?
[signature]
When trimming the fat, this email is identical to the first one which has 3 sentences, the last one of which is a question.One other issue with CCing everyone is that you always end up with someone throwing their two cents in about minor issues when they aren't even invested in the project.
To: You want to read this.
CC: You may want to refer to this.
For a tool that has been around for the last 30-odd years, I am amazed that people are as bad at email as they are. And I'm not talking here about secretaries, or non-technical staff. I'm talking about scientists in a research institute!
It's lazy, and I hate it, but it gets the job done for them.
http://www.unicom.com/pw/reply-to-harmful.html
There isn't any way, for a given client, to ensure consistent behavior when replying to all lists, and for the lists there isn't any reasonable expectation that all users will have the same client behavior. Things are a mess all around now.