You have seen this one on a wall, a slide, or a coffee mug:
Insanity is doing the same thing over and over again and expecting different results.
Attributed to Albert Einstein. It is not his.
The trail
Nobody has produced it in anything Einstein wrote, said, or signed. What researchers have produced is a much later and much duller origin: the line shows up in Narcotics Anonymous literature around 1981, in the context of recovery, where it means something specific and useful. From there it travels into self-help, then into business slides, and somewhere on that journey it picks up a physicist.
The novelist Rita Mae Brown puts a version of it in a character's mouth in Sudden Death (1983), attributed in the text to a fictional source. That is roughly the period the line becomes common — three decades after Einstein died.
The attribution is not a lie anyone told. It is a compliment that accumulated: a line that sounds clever gets handed to the cleverest person the sharer can name, and then the name is what gets shared.
Why this keeps happening
Because checking is slower than sharing, and there is no penalty for being wrong.
The same trail exists behind most of the famous ones. The Modern Language Association traces "Be yourself; everyone else is already taken" — universally given to Oscar Wilde — back to a 1999 Usenet post. The earliest known attribution to Mark Twain of "a lie can travel half way around the world while the truth is putting on its shoes" is from 1919, nine years after Twain died.
None of those are scandals. They are what happens when a quotation is treated as decoration rather than as a claim.
How to check one in two minutes
You do not need a library. Three steps catch almost everything:
- Search the exact sentence in quotation marks, plus the word the source would use — the book title, or "letter", or "speech". A real quotation has a work attached to it somewhere. A fake one has a thousand image results and no edition.
- Look for the earliest date, not the most confident one. If every attribution is from the last fifteen years and the person died in 1955, something is wrong. Wikiquote's talk pages and the Quote Investigator archive both do this legwork in public and cite what they found.
- Find the page. If you cannot get to a work, a year and a publisher, you have a saying rather than a quotation. Use it as a saying — unattributed, or "as the old line goes" — and you have not claimed anything untrue.
That third option is the one people forget. You are allowed to use a good sentence without a famous name on it. The name was never what made it good.
A line he did write
Einstein published plenty. From the 1934 collection The World As I See It:
The most beautiful experience we can have is the mysterious.
— Albert Einstein (1879–1955)The World As I See It (1934) · collected essays, English edition
It is a stranger sentence than the fake one, and a better one. It does not instruct you. It just says what he found, which is what a real person's words usually do.
A note on why I can print that one
Short quotations are free regardless of copyright status — that is why quote accounts can post a sentence from anything at all without asking. So the 1934 date is not a problem for the line above.
It would be a problem if I wanted to reprint the whole essay. Works published in the United States between 1931 and 1955 are only in the public domain if their copyright was not renewed, and fewer than fifteen percent of registrations ever were. Good odds are not evidence, and an author's death year says nothing about renewal either way. So my rule here is: a line from that window is fine with a citation; a page from it needs a renewal search, or I reprint from an earlier edition instead.
That is a tedious rule and I am telling you about it on purpose. It is the only reason to believe any of the other quotes on this site.
What this has to do with your business
Everything you publish is a claim, including the decorative parts.
A testimonial with no name. A statistic with no source. A quote with no edition. Individually each is harmless, and none of them is a lie. Together they teach a reader that the things on your page are not checkable — and once someone has learned that, they apply it to your prices and your promises too.
This matters most in the places you are least likely to audit. The founder story on your about page. The "trusted by" row. The round number in your headline that started as an estimate two years ago and has been repeated ever since. None of those were written to mislead. All of them are claims, and any one of them can be the thing a careful customer decides to test.
The fix is not to strip your site of personality. It is to make every claim either sourced or softened — "we have worked with over forty businesses" if you counted, "we have worked with businesses like yours" if you did not. The second sentence costs you nothing and cannot be caught out.
The cheapest trust you will ever buy is a source line under a quotation.