dropoffaudit.me · Articles

“One typo cuts sales in half.” I went looking for that study

September 2026 · Luca

You may have read that a single spelling mistake can cut online sales in half. I went looking for the study behind that number. There isn’t one. What does exist is smaller and more careful, and it still points the same way.

Where “half your sales” comes from

On 14 July 2011, BBC News reported that Charles Duncombe, an online entrepreneur running travel, mobile phone and clothing websites, said an analysis of website figures showed a single spelling mistake could cut online sales in half.1

The evidence given in the article is one website. He measured revenue per visitor on tightsplease.co.uk and found it was twice as high after an error was corrected. The article gives no time period, no traffic figures, and nothing about what else changed on the site between the two measurements.

He may be right, but this is one business owner’s observation about one website. It is still quoted on marketing and copywriting websites as if it were a research finding.

A randomized experiment on health information

In 2020, a team at Brighton and Sussex Medical School published a randomized online experiment in the Journal of Medical Internet Research.2 301 adults recruited in the UK rated short passages about multiple sclerosis, written like posts on a health forum, for how trustworthy they seemed, on a slider from 0 to 100.

Each passage existed in four versions: no errors, five spelling mistakes, five words in inappropriate capital letters, or both. The medical content was the same in every version. Compared with the error-free passages:

• −8.9points of trust with five spelling mistakes
• −6.4points with five words in shouting capitals
• −14.3points with both kinds of error

The penalties added up: two kinds of carelessness cost roughly the sum of each one alone. And the ceiling was already low. The most trusted error-free passage averaged 62.3 out of 100, even though every statement in it was medically correct.

What 1,410 people said hurts credibility

In 2001, the Persuasive Technology Lab at Stanford published a large study on website credibility.3 It was an online questionnaire, not an experiment: 1,410 people in the United States and Finland rated how much a list of website features would make a site more or less believable, on a scale from −3 to +3.

• −1.28“The site has a typographical error.”
• −1.45“The site has a link that doesn’t work.”
• −1.67“The site is rarely updated with new content.”

So in what people report, a broken link or a site that looks abandoned does as much damage as a typo, or more. For a clinic website that may be the more useful finding. My reading, not the study’s: a dead link on a services page, or a footer still showing a copyright year from long ago, sends the same signal.

The study that found less

Not every experiment finds a clear penalty, and this one belongs here too.

In 2017, researchers at Indiana University and Michigan State University showed 272 adults aged 35 to 55, from a national online panel, a customer review of Prilosec, an over the counter drug.4 Each person was randomly assigned one version of the review: written either with no errors, with keystroke typos (“wsa” for “was”), or with phonetic misspellings (“sevral”).

Overall, the errors did not significantly change how trustworthy the reviewer seemed. They did change other judgments: the reviewer with keystroke typos was rated less expert and more careless than the one with no errors. And among participants who generally trust other people, keystroke typos lowered trust more than phonetic misspellings did.

What I take from this for a clinic

• I found no study that measures what a typo costs a clinic in bookings. If you see a precise figure, ask where it comes from. The most quoted one is the 2011 anecdote.
• What has been measured is that readers rate text with errors as less trustworthy, that different errors add up, and that in what people report, broken links and a neglected site weigh as much as typos.
• The effect is not universal. It depends on the reader and on the kind of error.
• The fix costs minutes. When the possible downside is a weaker first impression on someone choosing a clinic, and the repair is a text edit, it’s worth doing. That is why these go into an audit, with the page where each one appears.

Sources

Each source was opened and read before this page was written. Where a study is a survey, or was produced with a company that sells a related product, it says so.

1. Coughlan, S. (2011, 14 July). Spelling mistakes ‘cost millions’ in lost online sales. BBC News. bbc.co.uk/news/education-14130854 Type: news report quoting one business owner about one website. No method published.
2. Witchel, H. J., Thompson, G. A., Jones, C. I., Westling, C. E. I., Romero, J., Nicotra, A., Maag, B., & Critchley, H. D. (2020). Spelling errors and shouting capitalization lead to additive penalties to trustworthiness of online health information: Randomized experiment with laypersons. Journal of Medical Internet Research, 22(6), e15171. doi.org/10.2196/15171 Design: randomized online experiment, peer reviewed. 301 UK adults recruited through Prolific, 2,709 ratings.
3. Fogg, B. J., Marshall, J., Laraki, O., Osipovich, A., Varma, C., Fang, N., Paul, J., Rangnekar, A., Shon, J., Swani, P., & Treinen, M. (2001). What makes Web sites credible? A report on a large quantitative study. Proceedings of CHI 2001, 61 to 68. doi.org/10.1145/365024.365037 Design: online questionnaire, self-reported. 1,410 respondents, 42% United States and 58% Finland.
4. Cox, D., Cox, J. G., & Cox, A. D. (2017). To err is human? How typographical and orthographical errors affect perceptions of online reviewers. Computers in Human Behavior, 75, 245 to 253. doi.org/10.1016/j.chb.2017.05.008 Design: randomized online experiment with three conditions, peer reviewed. 272 adults aged 35 to 55 from a commercial online panel.