Talent analytics is sold as a platform problem and it is mostly an arithmetic problem. A company of eighty people running a talent review produces perhaps six numbers worth acting on, and every one of them can be worked from figures the HR lead already has. The reason talent analytics tools get bought anyway is that nobody wants to maintain the arithmetic in a spreadsheet that only one person understands. This page sets out which six numbers matter, what each one is for, and where the line sits between a worksheet that computes them and a product that keeps them.
The six numbers a talent review produces
The distribution across the nine boxes, as counts and as shares of headcount. Bench strength, meaning the proportion of critical roles with at least one successor ready now. Succession exposure, the number of critical roles whose incumbent is expected to leave within a horizon and who has nobody ready. Regretted attrition, the leavers the company wanted to keep, and what replacing them cost at your own assumption. The total talent mix, employees and contingent workers on one line by headcount and by cost. And the build or buy comparison for a single open role. Six figures, each answering a question somebody will actually ask.
Why a platform is usually the wrong first purchase
Talent analytics tools priced per seat per month are built for companies with enough people that the distribution is statistically interesting and enough systems that the data has to be joined. A company of eighty has neither problem. What it has is a maintenance problem: the arithmetic exists in one spreadsheet, the person who built it is the only one who can change it, and it breaks when someone leaves. That is a real problem and it is worth paying to solve, but it is a record keeping problem rather than an analytics one, and buying an analytics platform to solve it means paying for the modelling and getting the storage as a side effect.
What the free worksheets return, and what they do not
The worksheets on this site take your own inputs and return the six figures above. They publish no benchmark, no industry rate and no threshold, because a figure asserted about other companies is not evidence about yours and this site does not hold data that would make it so. What they do not do is remember. Enter the counts and you get the shares; close the tab and it is gone. That boundary is deliberate and it is where the paid product starts: every person, every rating per cycle, every critical role and every successor slate against the record, branded and exportable.
Getting the inputs right matters more than the tooling
Every one of the six numbers is only as good as a definition somebody wrote down. Critical role has to mean something narrower than important, or bench strength becomes a measure of how many roles you listed. Ready now has to mean ready for this role, this year, not promising. Regretted has to be decided at the time the person leaves rather than reconstructed later, when memory is kind. Fix the definitions first and the arithmetic is trivial. Leave them loose and the most expensive talent analytics tools on the market will return numbers that move for reasons nobody can explain.
Questions people ask about talent analytics
What is the difference between talent analytics and HR analytics?
HR analytics covers the whole employee lifecycle, including payroll cost, absence and recruiting funnels. Talent analytics is the narrower set that comes out of a talent review: distribution, bench strength, exposure, regretted attrition and the workforce mix. This site works the second set.
Do I need a data analyst to do talent analytics?
For a company under a few hundred people, no. The six figures are counts, shares and one cost model each. What you need is a written definition for critical role, ready now and regretted, and a place the numbers live between cycles.
Can the worksheets export their figures?
The free worksheets are calculators and return their results on the page. Keeping the figures across cycles, with the people and ratings behind them, and exporting the branded result is what the paid subscription does.
Why does this site not publish benchmark talent figures?
Because a benchmark drawn from other companies tells you nothing reliable about a company of eighty people in your industry, and quoting one would be asserting a figure we do not hold the data for. Every threshold in the worksheets is yours to set.