Who Publishes This Research

Nearly every workforce statistic in circulation comes from an organisation with a position. Knowing which kind you are reading is most of the skill in using them.

The five kinds

Peer-reviewed research. Published in journals, methods described, subject to review. Rare in this field and by far the strongest — the Stanford hybrid-work study in Nature is the example this site leans on hardest. For a related reference, see this overview.

Survey organisations. Gallup and equivalents. Large samples, published methodology, commercial but not selling a solution to the problem they measure.

Professional bodies. SHRM and equivalents. Serve a membership of practitioners, which is a real constraint on saying convenient things.

Vendor research. Employee experience platforms, monitoring companies, HR software firms. Frequently large samples and genuine work, arriving at a product.

And industry analysis and trade press, which mostly repackage the above with varying attribution — a pattern this site describes elsewhere.

How to weight them

Peer-reviewed: use directly, cite the finding.

Survey organisations: use the figure, note the sample and date.

Professional bodies: good for cost benchmarks and practice norms.

Vendors: use the direction, treat the magnitude as theirs. A vendor reporting that poor management raises turnover risk is describing something real; the specific multiple is a number produced by somebody selling manager development.

Trade repackaging: trace it to the original before quoting, because attribution degrades at each hop.

The specific problem in this field

The largest published datasets on monitoring belong to monitoring companies.

Their technical descriptions are accurate — they know how the tools work better than anybody.

Their framing arrives at a purchase, and the same page frequently contains both an accurate account of why invasive monitoring backfires and a product that does it less invasively.

That combination is worth reading carefully rather than dismissing, because the accurate half is genuinely useful.

The number that keeps travelling

Large aggregate cost figures — hundreds of billions annually — circulate widely and are almost always vendor estimates with unpublished methodology.

They are useful for scale and useless for arithmetic. Run your own numbers off a per-person figure from a professional body instead, which is checkable and applies to you.

How this site handles it

Every figure carries its source type, in brackets, in the sentence.

Peer-reviewed findings are identified as such, because there are few of them and they deserve the weight.

Vendor figures are labelled at every use, not once on a method page.

And where sources disagree, both are given rather than the more convenient one. What that leaves unverified is listed separately.

The short version

• Five kinds: peer-reviewed research, survey organisations, professional bodies, vendor research, and trade repackaging
• Peer review is rare in this field and by far the strongest; the Stanford hybrid-work study in Nature is the anchor here
• Use vendor findings for direction and treat the magnitude as theirs, since the number was produced by somebody selling the remedy
• The largest monitoring datasets belong to monitoring companies, whose technical accounts are accurate and whose framing arrives at a purchase
• Aggregate cost figures in the hundreds of billions are almost always vendor estimates — useful for scale, useless for arithmetic
• Every figure on this site carries its source type in the sentence, and vendor figures are labelled at every use

For additional context on this topic, see GameSpot.