GLOSSARY · GHOSTWRITING
Unsourced claim
An unsourced claim is a statement of fact, most often a number or a percentage, presented in content without any indication of where it came from. "Eighty percent of B2B buyers decide before talking to sales" is an unsourced claim unless the post says who measured it, when and how. Unsourced claims are common on LinkedIn because they are easy to generate, they make posts sound authoritative, and most readers do not check. They are also one of the fastest ways to lose a careful reader, and AI drafting tools produce them readily, which is why review steps now look for them specifically.
How it works in practice
A claim needs a source when a reader could reasonably ask "says who?" and the answer matters. Statistics always qualify. So do statements about what a platform rewards, what a regulation requires, what a competitor does and what research shows. Opinions and first-hand observations do not need citations, but they should be labelled as what they are: "in the twenty deals I have worked on this year" is honest and specific, whereas "most deals" asserts knowledge the author does not have. The source should be close enough to the claim that a reader can find it: name the study or the organisation, give the year, and link where the format allows.
Where you cannot find a source, you have three choices. Cut the number. Replace it with your own evidence, clearly scoped. Or soften it to an honest observation without a figure. What you should not do is leave a precise number standing because it sounds good. Build this into review: before any post goes out, every figure is checked against a source that the named author could produce if challenged. Tools exist that scan drafts for statistics with no attribution, including Klype's draft checker, but the discipline is editorial. A single unsourced claim that a reader catches discounts the author's genuine proof points as well.
Example
Draft: "Studies show 73 percent of founders regret their first marketing hire." Reviewed: "Of the roughly thirty founders I have worked with on their first marketing hire, most said they would scope the role differently a second time." The second version is defensible.
Common mistake
Keeping a statistic because it is widely repeated. Many popular LinkedIn numbers trace back to no study at all, and "everyone cites it" is not a source when a buyer asks.