If Messi Were Your Employee, How Would You Evaluate His Performance?

If Lionel Messi joined your company tomorrow, would your performance management system actually recognise his contribution — or only the part of it that fits your scorecard?

Messi's public record contains an unusual combination: prolific goalscoring alongside a long history of assists and chance creation. Documented fact: he has repeatedly finished major competitions leading both goal and assist charts. Most corporate review systems would capture one of those two things beautifully and quietly ignore the other.

That is not a football problem. It is a measurement problem, and it is happening in your organisation right now to someone whose most valuable work is invisible to the metric you chose.

The people question this raises

Every scorecard decides what counts. The important question is not whether your chosen metric is accurate — it usually is — but what happens to the contribution that sits outside it: the colleague you unblocked, the deal you set up for someone else, the incident you prevented, the standard you held when nobody was watching. Measured systems do not merely record behaviour; they shape it. Whatever you count, people will produce more of, including at the expense of things you also needed.

This is Goodhart's law applied to people management: once a measure becomes a target, it stops being a good measure. Steven Kerr's classic paper on rewarding A while hoping for B described the same failure in 1975. It is still the most common design flaw in performance systems.

Where this shows up at work

• The engineer who ships less code because they spend Fridays reviewing everyone else's — and whose review comments are the reason production is stable.
• The account manager whose renewal rate looks ordinary because they inherited the three hardest accounts nobody else could hold.
• The person who onboarded and mentored three hires who are now outperforming their own numbers.
• The support lead whose ticket count fell because they wrote documentation that stopped the tickets being raised.

A four-part scorecard that catches all of it

The weights matter more than the categories. Publish them before the cycle starts. A stated 50/20/20/10 split tells people something clear and lets them plan; an unstated split means the loudest, easiest-to-count metric wins by default, and everyone learns that helping colleagues is charity work.

How to make 'enabling others' credible rather than fuzzy

1. Require named beneficiaries. "I helped the team" does not count; "I paired with Anisha weekly and she now owns the release process" does.
2. Collect two or three structured peer inputs per person with one question: what did this person do that made your work better or easier? Keep it short so people actually complete it.
3. Track the load, not the sentiment. Review counts, mentoring hours, on-call handovers and documentation written are all countable and all currently uncounted in most teams.
4. Calibrate in a group so one generous manager and one stingy manager do not produce two different standards.

A worked scenario

Two salespeople close the same revenue. On a single-metric system they are identical and get identical ratings. Under the four-part scorecard, the first inherited warm accounts, worked alone, and left two juniors unsupported. The second opened a new segment, wrote the discovery script the whole team now uses, and coached a junior who closed their first deal in month five. Same number, different contribution. The four-part view lets you pay and promote differently without inventing an excuse — and, just as importantly, it tells the first person exactly what is missing rather than leaving them to guess why someone else was promoted.

What readers often get wrong about this

• More dimensions is not automatically better. Four is manageable; eleven becomes a compliance exercise that managers rush the night before the deadline.
• Do not let team metrics swallow individual accountability. If the team outcome is weighted at half the score, strong performers on weak teams will leave.
• Soft dimensions need hard evidence. "Collaboration" without named examples becomes a popularity contest and a bias magnet — it is where women, quieter people and remote employees typically lose points unfairly.
• The analogy is imperfect: football has ball-by-ball public data on almost everything. Most companies do not, which is exactly why structured peer evidence matters more here, not less.

Interpretation, not fact: describing Messi as "a creator as much as a finisher" is an analytical reading of his record. The documented part is simply that both goals and assists appear in it at elite volume across many seasons. For HR purposes that is enough — a single number would have described half a career.

A 60-day implementation path

• Week 1–2: run the diagnostic question with every team and list the contribution types your system currently misses.
• Week 3–4: draft the four dimensions in your own language, with weights, and pressure-test them against three real people.
• Week 5–6: define what evidence each dimension accepts, and build the two-question peer input form.
• Week 7–8: brief managers with worked examples, run one calibration dry run, then publish the weights to everyone before the cycle opens.

Over to you

Which contribution does your current review process reliably miss — and who on your team is paying for that blind spot?