Every prevention vendor in the country quotes a return figure. Very few of them say which one, from where, over what period, or what it counted. This is a guide to reading the numbers, including ours.
Start with the ones that exist
Here are published Canadian benefit-cost ratios, as their publishers state them.
CA$1.62
Median return per dollar invested in workplace mental health programmes
And here are 3 from Ontario, by sector.
2.14
Benefit cost ratio for Ontario transportation employers, 2013 to 2018
Notice the spread. Transportation returns 2.14 and manufacturing 1.24 on the same methodology in the same province. A single large Canadian employer reported $4.10. The programme median across mental health interventions is $1.62, rising to $2.18 where the programme has run 3 or more years.
Anyone quoting you a single number for "the ROI of workplace prevention" is quoting one row of that table and hoping you do not ask which.
What a ratio of 1.24 actually means
It means that for every dollar spent, $1.24 came back over the study period. Net 24 cents. That is a real return and a modest one, and it is worth sitting with how modest it is, because the difference between 1.24 and the 4.10 figure is not a difference in effort. It is a difference in sector, baseline injury rate, programme maturity and what the study chose to count.
A ratio near 1.2 also means the intervention roughly pays for itself and the argument for doing it has to rest partly on things the ratio does not price - statutory duty, retention, the experience of the person who did not get hurt. That is a legitimate argument. It is just not a financial one, and presenting it as financial is how prevention programmes lose credibility with a CFO.
3 questions to ask any ratio
Over what period? A prevention programme that pays back over 5 years and one that pays back over 18 months can quote the same ratio. Only one of them survives a budget cycle.
Counting which costs? Direct claim cost only, or direct plus indirect? The accepted methodology adds indirect cost of 2 to 4 times the direct figure - a range wide enough that a ratio computed at the top of it is 4 times more flattering than one computed at the bottom.
Compared with what? A ratio against doing nothing is not the same as a ratio against the cheaper intervention you would otherwise have bought. Most published ratios are the former, which is the easier comparison.
Why our calculator refuses to give you a projection
The calculator on the why-wereset page takes your headcount, your claim count, your cost per claim and your budget, multiplies them by a published ratio you select, and stops.
It does not apply an injury-reduction rate. It could - it would be trivial to assume 20 per cent and produce a much more exciting number. We do not, because no Canadian publisher reports a reduction rate that generalises to an arbitrary employer, and an assumed one dressed as an output is the single most common dishonesty in this category.
The honest version of a forecast for your site needs your own baseline first. That is what the assessment produces, and until it exists, arithmetic on published ratios is the most anyone can legitimately give you.
Using one in a business case without embarrassing yourself
State the ratio, name the publisher, name the sector, and name the period. Then present the indirect cost as a range rather than a point, because that is how it was published. A business case built that way survives the question "where did this number come from," which is the only question that matters when it reaches finance.
The 3 questions and the position on reduction rates are ours. Every ratio above is reproduced exactly as its publisher states it, with a link to the source on each.
The full list of figures used anywhere on this site, with publisher, year and page reference, is at the evidence base.
