The six to nine months of salary figure is not SHRM's

Provenance: Attributed, not found. The figure is credited to an organisation whose published work, as far as we can find, does not contain it; the nearest statement there says something different. This describes how the figure is sourced, not whether it is true, and it is not a claim that anyone acted in bad faith.

You will see it in this form: according to SHRM, it costs a company six to nine months of an employee's salary to replace them. It usually comes with a worked example, a $60,000 employee costing $30,000 to $45,000, and it appears on recruiting sites, staffing-agency blogs, benefits vendors' pages and financial-wellness pitches, in almost identical wording.

A typical instance is Workable's 2023 article on the cost of replacing an employee: "According to the Society for Human Resource Management (SHRM), it can cost a company 6 to 9 months of an employee's salary to replace them." The sentence links to nothing. That turns out to be the rule rather than the exception.

What SHRM has actually published

SHRM has put two cost-of-turnover ranges into print, and neither is in months.

The substantial one is the SHRM Foundation's 2008 practice guide Retaining Talent, by David Allen, which says that "direct replacement costs can reach as high as 50%-60% of an employee's annual salary, with total costs associated with turnover ranging from 90% to 200% of annual salary". Its endnote for that sentence is worth following too. It cites "Cascio, W.F. 2006. Managing Human Resources: Productivity, Quality of Work Life, Profits (7th ed.)", a textbook, alongside a 2001 Academy of Management Executive article. So even SHRM's own range is a summary of other people's estimates, presented as such.

The recent one is a January 2025 piece from SHRM's Executive Network: "According to SHRM, the cost of replacing an employee can range from 50% to 200% of their annual salary, depending on their level." It cites no study for the range either, though it points to SHRM's turnover cost spreadsheet for readers who want to calculate their own.

We searched SHRM's site and its cost-of-turnover publications for the six-to-nine-months statement and did not find it. We cannot prove it never appeared anywhere under SHRM's name. We can say that the pages crediting SHRM with it do not link to a SHRM document, and that the SHRM documents on the subject say something different.

Where it probably came from

Six months of salary is 50% of it. Nine months is 75%. SHRM's direct-cost range runs from 50% to 60%, and its total-cost range starts at 90%. The circulating figure sits between the two, which is what you would expect from someone converting a percentage range to months from memory and rounding the top end up. That is a guess about the mechanism, and we hold it loosely. What is not a guess is that the figure as quoted matches nothing SHRM printed.

What the figure will and will not carry

As a rough order of magnitude it is not wild. Replacement costs of half to three-quarters of a year's pay are inside the range serious estimates cover for many roles, and a business case that used it as a placeholder would not be far wrong in direction.

It will not survive attribution. Credit it to SHRM in a document that gets checked, and the person checking will find the 50% to 200% range instead and wonder what else was made up. It also will not survive the worked example: a flat six to nine months applied to every salary assumes replacement cost scales with pay, which is the same unsupported step the 33% figure takes.

What to use instead

If you want SHRM's number, quote SHRM's number: 50% to 200% of annual salary depending on level, with the honest note that SHRM itself cites a textbook for it. If you want a defensible figure, build your own from hiring spend, interviewer hours, onboarding and the ramp to full productivity. The turnover cost calculator uses SHRM's published range as a bracket and shows the arithmetic, which is the most that a published multiplier can honestly do.

To be fair to SHRM

None of this is SHRM's doing. Their published ranges are hedged, sourced to the extent the underlying literature allows, and accompanied by a tool for calculating the real figure. A more precise-sounding number was attached to their name by other people, and it spread because a bracket with a brand on it is easier to quote than a range with a caveat.