Knowles Knowledge Centre•Effective Email
An Email Costs Two Minutes
To Send and Forty to Absorb
A practical framework for the only measure of an effective email: what it costs the people who open it.
Effective email, in the sense used here, is not clear email. Clarity is necessary and it is not the constraint. The constraint is that sending costs one person two minutes and absorbing costs everybody on the line their own reading, deciding and replying — a total the sender cannot see and has no reason to reduce. What follows is a question of what this message will cost in aggregate, whose answer actually changes anything, when you genuinely need it, what happens if nobody replies, and who closes the thread when it is over.
About this guide

The conventional response to an email problem is better writing. A sharper subject line, one ask per message, shorter paragraphs, a warmer sign-off. All of that is sound and all of it improves the experience of reading a message that should not have been sent to eleven people. Writing well reduces the cost per reader. It does nothing about the number of readers, the replies the message generates, or the fortnight a thread stays open because nobody was asked to close it.
This guide explains the five stages of the method, how it relates to Singapore’s Critical Core Skills, how email differs from chat, meetings and shared documents, and the case in which using it is the wrong move.
The practical answer
The practical answer at a glance
For practical use, the method can be organised into five stages, applied to one message before it is sent:
Cost the open
What will this message cost, added across its readers?
Whose answer
Whose reply would actually change what happens next?
Date the ask
When is it genuinely needed, as a date rather than a feeling?
If no reply
What will you do if nobody answers at all?
Shut the loop
Who states the outcome and closes the thread?
About this framework
This five-stage structure is the Knowles Cost-to-Open Framework used in this article. It is an editorial synthesis for workplace application—not an official government framework, accreditation standard or universal definition.
Four traditions supply the method, and none of them is a style guide
The account of attention as the genuinely scarce resource comes from the study of organisations and decision-making; the account of why the cost of coordinating determines how work is coordinated comes from economics; the account of saying as much as is required and no more comes from the philosophy of language; and the account of a shared resource being protected by rules a group sets for itself comes from the study of common-pool resources.
Herbert
Simon
Herbert Simon observed that where information is abundant, attention becomes the scarce resource, and that an organisation’s real constraint is therefore what its people can attend to rather than what they can be told. The practical consequence for email is that the sender is spending a resource that belongs to somebody else. Copying four extra people costs the sender nothing and the organisation four units of its scarcest input, and nothing in any mail client reports the total.
Ronald
Coase
Ronald Coase argued that the way work gets coordinated depends on the cost of coordinating it, and that activity moves to whichever arrangement makes the transaction cheaper. Email made reaching people almost free, which is why almost everything moved into it, including the things it serves badly. Coase’s point is not that email is bad; it is that the cheapness is the sender’s and the expense is somebody else’s, so the volume will keep rising until a rule rather than a price stops it.
Paul
Grice
Paul Grice set out the maxims a cooperative speaker is assumed to observe, among them the maxim of quantity: say as much as is required for the purpose and no more. The maxim cuts both ways in email, which is why the usual advice to be brief is incomplete. A message that omits what the reader needs forces a reply asking for it, which costs more than the paragraph would have, and a message that includes everything the sender knows forces the reader to work out which part is the ask.
Elinor
Ostrom
Elinor Ostrom showed, studying communities sharing a finite resource, that depletion is not inevitable and that the groups which avoid it do so through rules they make and monitor themselves rather than through rules imposed from outside. Collective attention behaves like such a resource. The organisations where email is bearable are not the ones with a policy; they are the ones where a team has agreed, among themselves, what gets copied and who closes a thread.
The recurring pattern matters more than the labels. Across the study of organisations, economics, the philosophy of language and the study of shared resources, the same five ideas reappear: attention rather than information is the scarce input; activity flows to whatever is cheapest to coordinate; the right length is the one the purpose requires and no other; a shared resource survives through rules the group sets for itself; and the sender bears a fraction of the cost they create.
The last of those disposes of the usual remedy. An organisation unhappy with its email will reliably respond by circulating guidance on writing better messages. Better messages are easier to read, which lowers the cost per reader and leaves the number of readers, the replies generated and the threads left open exactly where they were. Two teams can write equally well, and the one that agreed who may be copied and who closes a thread is the only one whose inbox is different.
The five stages of the Knowles Cost-to-Open Framework
1. Cost the open
The first stage estimates what the message will cost in total. It takes about ten seconds and almost nobody does it, because nothing in the act of sending reveals the number.
It includes:
- Counting the people who will open it.
- Estimating what reading and deciding takes each of them.
- Counting the replies it is likely to generate.
- Counting the people each of those replies reaches.
- Noticing when the total exceeds your own saving.
- Comparing it with a single conversation.
- Accepting the estimate as rough and still useful.
A sender has completed this stage when they can state a figure. “Eleven people, three minutes each, probably five replies going to all eleven” is about three hours of organisational attention for one question. That number is almost never calculated and is usually the only argument needed for a different approach.
For example, a request for views on a draft goes to eleven people, who each read the draft, form a view and reply to the group. The sender has saved themselves four separate conversations and spent several hours of other people’s attention to do it. Sometimes that is the right trade. It should at least be a trade somebody noticed making.
2. Whose answer
The second stage reduces the line to the people whose reply would actually change what happens next. Everybody else is being informed, and being informed is a different act that should look different.
This is the stage that most reduces total cost, and it is the one most resisted, because a short recipient list looks like a failure of consultation and a long one costs the sender nothing.
They include:
- Naming who you need an answer from.
- Asking what you would do differently with each answer.
- Removing anybody whose answer changes nothing.
- Moving the informed into a separate, later message.
- Saying explicitly who is being asked and who is not.
- Declining to copy upwards as insurance.
- Accepting that the shorter list is more exposed.
- Writing the names in the message, not just the header.
The test is whether you can say what each name is for. “Two people have to approve this and one has to tell me whether the date is possible; the other eight will get the outcome on Friday” is a line that halves the cost of the thread before a word of it is written.
There is a second-order effect worth naming. Copying somebody senior as insurance transfers risk from the sender to the recipient’s attention, and does it invisibly. It is the single commonest reason email volume rises in an organisation under pressure, and no amount of writing guidance touches it, because the motive has nothing to do with communication.
3. Date the ask
The third stage puts a date on the request. “As soon as possible” is not a deadline; it is an instruction to every recipient to guess, and the guesses will not agree.
They include:
- Naming a date and a time, not an urgency.
- Saying what the date is driven by.
- Making it the real date rather than a buffered one.
- Distinguishing a date you need from one you prefer.
- Saying plainly when there is no deadline at all.
- Not marking routine messages urgent.
- Letting a recipient negotiate the date.
A stated date is also the sender’s own discipline. Working out when you genuinely need something frequently establishes that you need it next week rather than today, which changes the message from an interruption into a queue item and costs the recipient a fraction as much.
For example, a request marked urgent arrives at four in the afternoon and is answered by three people that evening. Asked afterwards, the sender needed it by the following Wednesday. The urgency was a habit rather than a fact, and it cost three people an evening that nothing required.
4. If no reply
The fourth stage states what will happen if nobody answers. It is the least common sentence in workplace email and it removes more follow-up traffic than any other single change.
It includes:
- Stating the action you will take by default.
- Making the default a reasonable one.
- Saying which decisions cannot have a default.
- Giving the default the same date as the ask.
- Removing the need for anybody to reply agreeing.
- Not using it to push through something contentious.
- Accepting that silence then genuinely counts as consent.
- Recording the default in the thread, not in a diary.
A stated default can answer:
What happens if nobody replies?
By when does that take effect?
Which decisions have no default?
Who can still object, and how?
The sixth bullet is the one that keeps this honest. A default is a convenience for routine matters: which room, whether the template changes, who attends. Using one to carry a contested decision through an inbox is a way of avoiding a disagreement rather than resolving it, and the people who notice will stop ignoring your messages, which costs everybody more.
5. Shut the loop
The fifth stage closes the thread. An unclosed thread sits in the inboxes of everybody on it as an item that may still need something, and the cost of holding it open is paid daily by people who have nothing left to do.
Shutting the loop includes:
- Saying what was decided, in one line.
- Saying it to everybody who was on the thread.
- Naming who is doing what, where anything follows.
- Saying plainly that nothing further is needed.
- Doing it even when the answer is that nothing changed.
- Deciding in advance whose job this is.
- Accepting that the closing message is worth its own cost.
An open thread is a standing charge on everybody still on it. One line closing it is the only message in the exchange that reduces the total.
This fifth stage connects the other four. The costing establishes the stakes; the short list keeps them low; the date stops the guessing; the default removes the confirmations. All of that produces a thread that is cheap to participate in — and a cheap thread is exactly the kind that nobody bothers to close, because no single person is paying enough to notice.
How the five stages work together
The stages are one sequence applied to a single message, not five rules to recite.
Consider an illustrative Singapore workplace scenario: a team reports that email is unmanageable, a mailbox hygiene campaign is run, and six months later the volume is unchanged.
The sequence runs:
Cost the open:
a fortnight of one team’s sent messages is examined. The median message reaches nine people and generates two replies to all nine. Nobody in the team had ever estimated the aggregate, and the arithmetic surprises everybody, including the heaviest senders.
Whose answer:
recipients are tested against the question of what each answer would change. On the messages reviewed, a median of two names would have changed anything. Of the rest, roughly half are being informed and the other half are copied as insurance, which three people acknowledge without being asked.
Date the ask:
deadlines are examined. Two thirds of the requests carry an urgency rather than a date, and in the cases that can be checked the actual need was several days later than the reply arrived.
If no reply:
defaults are looked for and almost none exist, which accounts for a large share of the replies: most are not answers but confirmations that somebody has seen the message and does not object.
Shut the loop:
open threads are counted. More than half of the threads from the fortnight have no closing message, and several are still being mentally tracked by people with nothing left to contribute to them.
The team’s first instinct was mailbox hygiene. Filing and unsubscribing reduce what an inbox looks like and change nothing about what arrives, which is why the volume was unchanged six months later.
The stages depend on one another in a specific way: a short recipient list with no stated default still produces a reply from every name on it, because each one confirms rather than answers; and a dated ask with no closing message produces a thread that is answered on time and then sits open in nine inboxes for a fortnight, which is the cost the whole method exists to remove.
Four situations should be named before anybody adopts this. Where the message is the record — a decision that must be auditable, an instruction with legal weight, anything a regulator may read — the cost calculation is the wrong frame and the right one is completeness; write it long, copy whoever needs to be on the record, and keep it. Where somebody needs to be told something difficult, the method’s arithmetic will recommend the cheapest channel and the cheapest channel is usually the wrong one; a conversation that costs an hour is correct here and the cost is the point. Where the sender has no authority to shorten the recipient list — a culture in which not copying a head of function is read as concealment — the second stage is unavailable to them and the remedy sits with whoever set that expectation. And where the problem is that the work itself is unclear, email volume is a symptom and a message discipline will reduce the traffic without touching the confusion generating it.
Is email the same as chat?
No. The four overlap, and they differ in what they cost and what they leave behind.
Term
Primary emphasis
Examples
Resolving something small, now, with one or two people
Short turns, presence, fast loops, little record
Settling something that needs several people at once
Synchronous time, real-time disagreement, a decision
Holding something several people change over time
A current version, comments, history, no inbox cost
Asking named people for something, with a record
Addressed asks, a date, a default, a closing line
Researchers have a name for the confusion that a family of overlapping labels creates: the jangle fallacy, the assumption that two things must be different because they carry different names. “Business writing,” “email etiquette,” “written communication” and “correspondence skills” describe substantially the same territory with different emphasis, and nothing in this method depends on preferring one of them.
The practical consequence is twofold. Treating email as a cost rather than a skill invites the question this article asks — what did this message spend, and whose? — which an etiquette guide does not invite, because etiquette is about the experience of the message rather than its price.
Conversely, the vocabulary should not be stretched over messages it cannot carry. This is not a method for a difficult conversation, and it is not a method for a document that several people have to edit.
A record, for example, remains a record rather than a message to be priced. Somebody writing an auditable decision may still name the people who need it and close the thread, and should. They may not shorten it for cost, because the reader they are writing for is somebody in two years who was not there, and the whole value of the message is that it is complete.
For a fuller decision guide, see email vs chat, meetings and shared documents.
How does this relate to Singapore’s Critical Core Skills?
SkillsFuture Singapore identifies 16 Critical Core Skills that support workers’ ability to develop and use technical competencies. They are arranged into three clusters:
Thinking Critically
Creative Thinking, Decision Making, Problem Solving, Sense Making and Transdisciplinary Thinking.
Interacting with Others
Building Inclusivity, Collaboration, Communication, Customer Orientation, Developing People and Influence.
Staying Relevant
Adaptability, Digital Fluency, Global Perspective, Learning Agility and Self Management.
Two things in that mapping are worth stating here rather than leaving to the crosswalk. The first is that If No Reply is reached by no named skill at all. Sixteen skills describe communicating, deciding and collaborating; not one describes stating in advance what you will do when nobody responds, which is the single sentence that removes the largest share of confirming replies. The second is that Communication maps to Date the Ask — and the reason deadlines are vague is almost never expressive. It is that the sender has not worked out when they need the thing, which is a planning act performed before any writing begins. The frameworks should not be presented as interchangeable. SkillsFuture’s Critical Core Skills are an official Singapore framework; the five-stage structure is a simplified editorial model.
A fuller mapping is available in the practical crosswalk between Singapore Critical Core Skills and writing effective emails.
This crosswalk is not a claim that a Knowles Training Institute programme is SSG-accredited, WSQ-approved or SkillsFuture-funded.
Why does the cost matter now?
Reaching eleven people has never been cheaper for the sender and has never cost the organisation more, because the same eleven are now reachable by everybody else at the same price. Volume is not a failure of manners; it is the predictable result of a cost that falls almost entirely on people who did not choose to pay it.
World Economic Forum 2025 employer survey
Employers also gave prominence to:
International employer survey — not a Singapore-specific forecast.
Singapore’s Critical Core Skills similarly combine transferable cognitive, interpersonal, adaptive and digital capabilities with technical competence.
Circulating guidance on better writing will not by itself reduce what arrives, because nothing in it changes the number of recipients, the replies generated or the threads left open.
Can this be learned?
Yes, and the gains come from the list and the default rather than from the prose. A working cycle contains:
A rough figure for what a message will cost in total
A recipient list of people whose answer changes something
A date rather than an urgency, with its reason
A stated default for what happens without a reply
A named person responsible for closing the thread
A team agreement about copying, made by the team
Evidence from Singapore’s Institute for Adult Learning
The Institute for Adult Learning’s Adult Learning Collaboratory examined future-oriented teaching methods across 762 worker-learners, 43 educators and facilitators, and 8 organisations, studying 86 learning interventions.
The work focused on emergent scenarios — which ask learners to respond to situations that evolve, change or remain uncertain — and generative dialogue, which asks learners to challenge assumptions and build understanding collectively rather than absorbing information individually.
The findings do not prove that every course using scenarios will produce the same result. They support contextual practice, dialogue and reflection over information delivery alone.
Which part should be fixed first?
Start with the recipient list—not with the wording.
Whose reply would change what happens next?
What will this cost, added across everybody on it?
Is the obstacle the writing, the copying culture or the work itself?
What happens if nobody replies at all?
Question 3 matters most. Where not copying a head of function is read as concealment, no individual can shorten a list and the remedy belongs to whoever set that expectation; where the work itself is unclear, the traffic is a symptom. A structured approach is set out in how to identify whether email is costing your team more than it saves.
Example: a hygiene campaign and unchanged volume
The cause may be nine-name recipient lists, insurance copying and threads nobody closes—not a shortage of filing discipline.
The response may require an agreement about copying, a stated default, and somebody named to close each thread.
Common mistakes in workplace email
Better prose lowers the cost per reader and leaves the number of readers untouched. A well-written message to eleven people who did not need it is a well-written waste of several hours, and it will be praised, which is what makes this the most durable mistake in the set.
Adding a senior name transfers risk from the sender to somebody else’s attention, invisibly and at no cost to the sender. It is the commonest driver of volume in an organisation under pressure, and it is a question of what gets rewarded rather than of how anybody writes.
“As soon as possible” asks every recipient to guess, and their guesses will not agree. Working out the actual date frequently reveals that the thing is needed next week, which turns an interruption into a queue item at no cost to anybody.
Where silence has no consequence, every recipient replies to confirm they have seen the message and do not object. Those confirmations are a large share of all workplace email and are generated entirely by the absence of one sentence.
An unclosed thread is a standing charge on everybody still on it, paid in small daily amounts by people with nothing left to contribute. Nobody closes it because no single person is paying enough to notice, which is why the closing has to be somebody’s job in advance.
Difficult news, disagreement about judgement and anything likely to be misread are badly served by a medium with no turn-taking. The method’s own arithmetic will recommend the cheap channel, and this is the case where the expensive one is correct and the expense is the point.
Volume is produced by a cost structure rather than by carelessness, which is why campaigns aimed at individual habits fade within a quarter. What holds is a rule a team has made and monitors itself: who gets copied, what the default is, who closes.
Turning a framework into a working practice
A practical development cycle contains six steps:
This cycle prevents the organisation from treating a cost problem as a writing problem. It also creates a clearer basis for deciding whether a message should be an email at all, or a conversation, a document or nothing.
Organisations that decide facilitated development is the right response can review the Writing Effective Emails Training Course in Singapore, a commercial programme option for facilitated development across the five stages.
Course format, facilitator suitability, customisation and expected workplace application should be evaluated against the organisation’s identified needs.
Frequently asked questions
That it costs the organisation less than what it achieves. The method estimates the total cost across readers and the replies they generate, reduces the list to people whose answer changes something, replaces urgency with a date and its reason, states what happens if nobody replies, and names who closes the thread.
Etiquette improves the experience of a message: tone, salutations, how a request is phrased, when a reply is owed. All of that lowers the cost per reader and is worth having. This method asks a prior question — how many readers, and should there be any — and the two answers are independent.
Not necessarily, and the advice to be brief is frequently wrong. A message that omits what the reader needs generates a reply asking for it, which costs more than the missing paragraph. The right length is the one that lets the reader act without writing back, which is sometimes longer than the sender finds comfortable.
Not inbox size. Median recipients per message, replies generated per message and who they reach, the proportion of requests carrying a date rather than an urgency, the proportion carrying a stated default, and the proportion of threads with a closing message.
Where the message is a record that must be complete and auditable; where somebody needs to be told something difficult and the expensive channel is the correct one; where the sender has no standing to shorten a recipient list; or where email volume is a symptom of unclear work rather than a cause. In those cases the arithmetic gives the wrong answer.
Conclusion
Email is usually treated as a writing skill and is better treated as a shared resource that senders spend and recipients pay for.
A sender needs to know roughly what a message will cost in total, whose answer would change anything, when the thing is genuinely needed, what happens if nobody replies, and who closes the thread. Better prose answers none of those, which is why a team can be trained in business writing and find its inbox exactly as it was.
For employers, the first step is not to circulate guidance. It is to take a fortnight of one team’s sent messages, count the median recipients and the replies generated, and find out how many threads were ever closed.

