Why good talent leaves: what a 34.6% turnover rate shows

Alimentos Andina S.A.S. is the company in the talent dashboard: 1,180 career paths over two years, with their reviews, absences and salaries. These are the patterns it shows and how to look for them in your own company.

average review score of those who resign vs. those who stay (1 to 5 scale)

3.62 vs. 3.48

of resignations were rated 4.0 or higher: nearly three in ten

28.8%

replacement cost of turnover in a single year

COP 5,928M

Finding 1 · The better-rated people are the ones leaving

It is common to assume that the people who leave “weren’t a fit.” In this data the opposite happens: those who resign voluntarily are rated higher than those who stay. The gap in the average is small (3.62 vs. 3.48), but it runs the opposite way from what you would expect, and 28.8% of those who resigned were rated 4.0 or higher. That makes sense, because the market makes offers to strong performers. Involuntary turnover removes low performance and voluntary turnover takes the high. If you only look at the total rate, the two effects offset each other in the average and the problem stays hidden.

The signal to build: match every resignation with its last review and report the average score of voluntary departures every quarter. It is one more column in the report.

Finding 2 · Resignations cluster in the first quarter

39.4% of the resignations in the two years happen between January and March, right after the statutory bonus and incentive payouts. The company pays the bonus to people who have already decided to leave, and that repeats every year.

The signal to build: the monthly distribution of resignations over the last 24 months. If the first quarter concentrates more than 35%, development conversations and adjustments for critical talent should happen in November, before the decision to leave has been made.

Finding 3 · Half of all departures happen in the first year

50.5% of departures happen before the twelve-month mark. For each one the company pays for full recruiting, selection and training without the person reaching full productivity. At the company in the dashboard, early departures cost COP 3,483M in one year.

The signal to build: turnover by hire cohort (survival at 3, 6 and 12 months). If the curve drops sharply before month 6, the problem is usually selection or onboarding. If it drops between months 6 and 12, it is usually the manager or the expectations. The collaboration network helps you see it earlier: the note on organizational networks covers the 90-day signal.

What they have in common

None of these numbers needs machine learning or an expensive platform. What they need is for payroll, reviews and HR records, which already exist, to be connected and queried with specific business questions.

What to do first

1. Cross resignations with reviews for the last 24 months. If the data exists it takes an afternoon, and it shows whether the strong performers are the ones leaving.
2. Plot resignations by month. If there is seasonality, bring career conversations forward so they happen before the peak.
3. Build the first-year survival curve and assign it an owner. It is an indicator the finance team understands without much explanation.