Start hereCarnegie wrote that research had found that, even in engineering, about 85 percent of financial success is due to skill in human engineering. Yet the 1918 report behind the figure asked engineers what they thought mattered, and measured nobody's earnings. So which of the book's ideas hold up?
About the book
How to Win Friends and Influence People is Dale Carnegie’s 1936 guide to getting along with people. Its title has become, in the words of its 2022 preface, “a household phrase”.
The author, Dale Carnegie, was born Dale Carnagey in Maryville, Missouri, on 24 November 1888, and changed the spelling of his name after speaking at Carnegie Hall. He died on 1 November 1955.
Since 1912 he had run adult courses in New York, first in public speaking and then in getting along with people, and the book grew out of them. His daughter’s preface tells how it began. Leon Shimkin, an editor at Simon & Schuster, attended one of his lectures in 1934 and persuaded him to write it. He started from an outline based on recordings of his lectures.
Carnegie’s own line on it, as she quotes him: “I didn’t really write How to Win Friends. I collected it.”
Simon & Schuster published it in 1936 in a first printing of 5,000 copies. The same preface says it sold a quarter of a million copies in its first three months, and more than 30 million since.
It is a practical self-help book. Each principle is illustrated with stories, many drawn from famous people’s lives, such as Lincoln’s, and from what his students reported back to class.
The current edition has four parts and 30 principles. The 1936 first edition went further: its Part Six was Seven Rules for Making Your Home Life Happier, which is not in the current edition’s contents. The text was revised in 1981 and again in 2022, when his daughter Donna Dale Carnegie and the writer Andrew Postman went back to the 1936 edition.
The 2022 preface calls the book “the first of its kind”. Books of self-help are older: Samuel Smiles’s Self-Help appeared in 1859. A few pages later it says Carnegie “didn’t invent the concepts he wrote about, but he was a pioneer in the way he presented them.”
See it from their side
His daughter’s preface calls one idea the linchpin of the book: seeing things from the point of view of others. Almost every principle is a way of doing that.
In the book’s view, people act on what they want. So it keeps asking what the other person wants before asking for anything.
Try it: before you ask someone for something, write down one thing they want from the same situation.
Don’t criticise; appreciate honestly
Part One opens with three principles: don’t criticise, condemn or complain; give honest, sincere appreciation; and arouse in the other person an eager want. Criticism, the book argues, makes people defend themselves rather than change.
The book separates appreciation from flattery by sincerity: “One is sincere and the other insincere.” It also says flattery “seldom works with discerning people”.
One of its stories fits here: a sharp letter Lincoln wrote to General Meade and never sent. Lincoln wrote to Meade on 14 July 1863, after Lee’s army escaped: “Your golden opportunity is gone”.
Does it hold up? Appreciation helps; flattery can work too, though obvious flattery works less well. A 1996 meta-analysis found ingratiation had a small positive effect on performance evaluations and a stronger one on liking. The more transparent the ingratiation, the smaller the effect: low to moderate transparency did best.
A 2010 study found that even when people discounted flattery with an obvious motive, a favourable implicit reaction remained, and it had more influence than their conscious judgement. Our reading: the line on sincerity is a good ethical rule, but the claim about flattery goes further than the studies.
The Lincoln story checks out. The book says Meade never saw it; the letter survives in Lincoln’s papers, filed with his endorsement “To Gen. Meade, never sent, or signed.”
Try it: make praise specific and true. Appreciation raises liking, and praise that looks calculated is the weaker kind.
Be interested rather than interesting
Part Two gives six ways to make people like you:
- Become genuinely interested in them.
- Smile.
- Remember their names.
- Listen, and encourage them to talk about themselves.
- Talk in terms of their interests.
- Make them feel important, sincerely.
Does it hold up? The advice to get people talking is supported in one team’s experiments; we found no independent replication. In two online experiments and an observational study of speed dates, people who asked more questions, particularly follow-up questions, were better liked. Responsiveness (listening, understanding, validation and care) explained the effect. People did not expect it.
A 2019 reanalysis of the speed-dating data argued that a second date is not the same as liking, and that asking questions is not the same as listening. The original authors replied that the reanalysis tested the wrong variables.
The reanalysis covered only the speed-dating study, not the two experiments. Our reading: the link between asking questions and being liked stands; whether asking counts as listening is the open part.
A 2025 correction reports that an independent audit confirmed all the paper’s conclusions. Separately, a 2012 study found that people were willing to give up money to talk about themselves.
Try it: when someone mentions something, ask a follow-up about it before you change the subject.
Remember their names
One principle in Part Two is about names. In the current wording, a person’s name is “the sweetest and most important sound in any language”. The 1936 wording was about a man’s name.
Does it hold up? Your own name does stand out, and being remembered pleases people. In a 1959 listening experiment, the only thing found to break through an ignored message was the listener’s own name. A 2006 brain-imaging study found activation specific to hearing one’s own name.
We found no study that tests whether a name is the sweetest sound. A 2026 retail study found that staff using a customer’s name could raise satisfaction and sales, but that overusing it lowered satisfaction.
Try it: get the name right the first time, and use it without overdoing it.
Smile
The smile chapter goes further than a smile: “Act as if you were already happy, and that will tend to make you happy.”
Does it hold up? A small effect seems real. The famous 1988 pen-in-mouth study did not replicate in a 17-lab test in 2016 (a rating difference of 0.03 units, against 0.82 in the original).
You can’t win an argument
Part Three starts with this rule: “The only way to get the best of an argument is to avoid it.” The 1936 text gives the reason: “if you lose it, you lose it; and if you win it, you lose it”. In the book’s account, the loser feels beaten and resents the winner.
Does it hold up? Facts do tend to move people; the social cost is the untested part. A 2018 meta-analysis of 65 studies found corrections had a moderate effect on belief in misinformation, weaker in politics and marketing.
Carnegie’s main point is about the relationship: that the loser resents the winner. We found no study that tests that directly.
Try it: separate the fact from the contest. Correct the point, then leave the other person room to agree later.
Persuade and lead without a fight
The rest of Part Three is about persuading without a fight. Its principles include:
- Show respect for the other person’s opinion, and never tell them they are wrong.
- Admit your own mistakes quickly.
- Begin in a friendly way.
- Let them do most of the talking.
- Let them feel the idea is theirs.
Part Four is about leading. Its principles include:
- Begin with praise.
- Point to mistakes indirectly, and mention your own first.
- Ask questions instead of giving orders.
- Let people save face.
- Praise every improvement.
- Give people a fine reputation to live up to.
Practise it like a skill
Carnegie’s Nine Suggestions treat the book as a handbook to reread and apply, not a book to finish. He describes a banker who spent part of each Saturday evening reviewing the week’s meetings and asking himself, “What mistakes did I make that time?”
He also suggests offering a friend a dollar every time they catch you breaking a principle. His daughter adds that he kept a file of his own lapses.
We found no study of the weekly review as such, so it is a habit to try, not a finding.
Try it: once a week, look back at one conversation and note what you would do differently.
What holds up, and where the book overstates
What the sources show for each idea
Most of the book’s advice on asking questions and appreciation lines up with later research, with some conditions. Its famous statistic does not: it comes from a survey of opinions.
Where the 85 percent comes from
In his introduction, Carnegie credits research by the Carnegie Foundation for the Advancement of Teaching, which he says was later confirmed at the Carnegie Institute of Technology. It found, he writes, that even in engineering “about 85 percent is due to skill in human engineering”.
There, about fifteen hundred engineers answered in writing what decides success in engineering. They named personal qualities more than seven times as often as technical knowledge. More than seven thousand then ranked six groups of qualities, with Character first and Technique last. So the top group was character, not dealing with people.
Nobody’s earnings were measured, and no 85/15 split appears on those pages. We could not trace the Carnegie Institute of Technology study.
Who it’s for, and who can skip it
It suits anyone who finds dealing with people harder than the work itself: colleagues, customers, family. It is easy to read and built on stories. Its core (ask about people, appreciate honestly) mostly agrees with later research.
Its statistic and some of its framing are from 1936. Readers who want evidence rather than anecdotes, or who already find conversation easy, can skip it. Readers who worry it teaches manipulation will find that the book keeps insisting on sincerity.
Quick answers
Is the "85 percent of success comes from people skills" statistic true?
It is not a measurement. The Carnegie Foundation says the figures were extrapolated from a 1918 report, which asked engineers what they thought mattered for success. Nobody's earnings were measured.
Does asking questions make people like you?
In one team's 2017 experiments, people who asked more questions, particularly follow-up questions, were better liked. A 2019 reanalysis disputed the speed-dating part, and we found no independent replication.
Is a person's name the sweetest sound to them?
Your own name does stand out: in a 1959 experiment it was the only thing found to break through an ignored message. In 1995 experiments, being remembered by name was taken as a compliment. We found no study that tests whether a name is the sweetest sound.
Did Lincoln really write a letter to Meade and never send it?
Yes. Lincoln wrote to Meade on 14 July 1863, after Lee's army escaped, and filed the letter with the endorsement "To Gen. Meade, never sent, or signed."
Use this
Ask about people, praise what is true, and correct the point without making it a contest.
Am I trying to understand this person, or to win?
Quick quiz
Nothing is scored or saved. Recalling it once now is what makes it last.
1. Where does the book's 85 percent figure come from?
Answer: A 1918 report on engineers' opinions.Read it again ↑
2. Which questions did the 2017 studies single out as raising liking?
Answer: Follow-up questions.Read it again ↑
3. What did a 2026 retail study find about using a customer's name too often?
Answer: It lowered satisfaction.Read it again ↑
How we checked this · Last checked October 2026
- What we read: the publisher's free sample of the 2022 edition (copyright page, contents, preface, introduction and Nine Suggestions), the principle list published by Dale Carnegie Training, and short passages of a 1936 printing through the Internet Archive's search-inside tool. Wording differs slightly between editions.
- For the 85 percent figure we read the Carnegie Foundation's FAQ and pages 106-107 of Mann's 1918 report. We could not trace the Carnegie Institute of Technology study.
- For the studies we read the abstracts, plus the full text of the 1996 ingratiation meta-analysis and the 2025 correction notice.
- In October 2026 we re-checked the 2017 question-asking paper: it carries a 2025 correction and no retraction, and a search of Europe PMC turned up no independent replication.
- Where we say no study tests something (a name as the sweetest sound, resentment after an argument, the weekly review), none turned up in our searches.
Sources
- D. Carnegie, How to Win Friends and Influence People (Simon & Schuster, 1936; revised 1981 and 2022): publisher’s sample of the 2022 edition (Vermilion: copyright page, contents, preface, introduction, Nine Suggestions); short passages of a 1936 printing via the Internet Archive’s search-inside tool.Primary
- Dale Carnegie Training, “Principles To Live By”.Primary
- State Historical Society of Missouri, “Dale Carnegie”, Historic Missourians.Scholarly
- Carnegie Foundation for the Advancement of Teaching, FAQ on the 85 percent figure.Scholarly
- C. R. Mann, A Study of Engineering Education (Carnegie Foundation, 1918), pp. 106-107.Primary
- N. Moray, “Attention in dichotic listening”, Quarterly Journal of Experimental Psychology 11 (1959).Scholarly
- D. P. Carmody and M. Lewis, “Brain activation when hearing one’s own and others’ names”, Brain Research (2006).Scholarly
- D. J. Howard, C. Gengler and A. Jain, “What’s in a name?”, Journal of Consumer Research 22 (1995).Scholarly
- “Call me by my name”, Journal of Retailing (2026).Scholarly
- K. Huang, M. Yeomans, A. W. Brooks, J. Minson and F. Gino, “It doesn’t hurt to ask”, Journal of Personality and Social Psychology 113 (2017), and its correction (2025).Scholarly
- D. I. Tamir and J. P. Mitchell, “Disclosing information about the self is intrinsically rewarding”, PNAS 109 (2012).Scholarly
- A. N. Kluger and T. E. Malloy, “Question asking as a dyadic behavior”, Journal of Personality and Social Psychology 117 (2019).Scholarly
- M. Yeomans et al., “It helps to ask”, Journal of Personality and Social Psychology 117 (2019).Scholarly
- R. A. Gordon, “Impact of ingratiation on judgments and evaluations”, Journal of Personality and Social Psychology 71 (1996).Scholarly
- E. Chan and J. Sengupta, “Insincere flattery actually works”, Journal of Marketing Research 47 (2010).Scholarly
- N. Walter and S. T. Murphy, “How to unring the bell”, Communication Monographs (2018).Scholarly
- B. Swire-Thompson, J. DeGutis and D. Lazer, “Searching for the backfire effect”, Journal of Applied Research in Memory and Cognition 9 (2020).Scholarly
- E.-J. Wagenmakers et al., “Registered Replication Report: Strack, Martin, & Stepper (1988)”, Perspectives on Psychological Science 11 (2016).Scholarly
- N. A. Coles, J. T. Larsen and H. C. Lench, “A meta-analysis of the facial feedback literature”, Psychological Bulletin 145 (2019).Scholarly
- N. A. Coles et al., “A multi-lab test of the facial feedback hypothesis by the Many Smiles Collaboration”, Nature Human Behaviour 6 (2022).Scholarly
- S. Centorrino et al., “Honest signaling in trust interactions”, Evolution and Human Behavior 36 (2015).Scholarly
- American Battlefield Trust, “Lincoln’s Unsent Letter to George Meade”.Primary
- House Divided project, Dickinson College, “Letter to George Meade (July 14, 1863)”.Primary
- S. Smiles, Self-Help (1859), Project Gutenberg.Primary