Mastering the Pymetrics Balloon Game: A Practical Guide for Candidates

The Pymetrics balloon game is based on a well-known research task, the Balloon Analogue Risk Task (BART). You inflate a series of balloons to earn money, and you decide when to stop and collect before a balloon pops. Pymetrics, now part of Harver, says there is no right or wrong way to complete its games: your choices are used to match you to suitable roles, not graded against an answer key.

How the balloon game works

The core loop comes from the BART research task:

• Pump to earn. Each pump inflates the balloon and adds a little money to that balloon.
• The balloon can pop. Each balloon pops at a point you cannot predict. If it pops, you lose the money from that balloon only.
• Collect to bank it. At any time you can stop and collect, which moves that balloon's money into your total and brings up a new balloon.
• Many balloons in a row. You repeat this across a series of balloons.

The exact controls, amounts and number of balloons depend on the version you get, so follow the instructions on your screen.

A worked example

Here is how a few balloons might go. The numbers are for illustration only.

1. Balloon 1: you pump 10 times, it has not popped, and you collect. That balloon's money is banked.
2. Balloon 2: you pump 18 times and it pops on the 18th pump. You lose that balloon's money, but balloon 1's money is safe.
3. Balloon 3: after the pop you feel cautious and collect after 6 pumps.
4. Balloon 4: you return to your usual pace and collect after 11 pumps.

The point of the example: each decision is about one balloon at a time, a pop only costs the current balloon, and how you react to pops is part of how you naturally handle risk.

What the game measures

In research, the BART is a behavioral measure of risk taking: how far people push for a bigger reward when a loss is possible. Pymetrics uses games like this to build a picture of your behavioral traits and compare it with the profiles of different roles.

Pymetrics does not publish a target number of pumps or a target total for any job. Its own guidance is that there is no right or wrong way to complete the games, and that the more you are yourself, the more helpful your results will be.

How to approach the balloon game

• Read the instructions first. Make sure you know which button pumps and which one collects before you start.
• Set aside uninterrupted time. Pymetrics advises finding a quiet place for about 30 minutes. The games cannot be paused, and each one lasts about 1 to 3 minutes.
• Play the way you would naturally decide. Trying to guess what an employer wants usually produces choices that do not reflect you.
• Treat each balloon on its own. A pop on one balloon does not tell you when the next one will pop.
• Stay focused, not frantic. Random clicking or rushing through balloons adds noise rather than showing anything about you.

FAQs

Is there a right way to play the Pymetrics balloon game?

Pymetrics says there is no right or wrong way to complete its games. Your choices are used to match you to roles, so play the way you would naturally decide.

Can you pause the balloon game?

No. Pymetrics says its games cannot be paused, and each one lasts about 1 to 3 minutes.

What happens when a balloon pops?

You lose the money from that balloon only. Money you already collected from earlier balloons is kept.

How should I prepare?

Pymetrics says no preparation time is required. Read the instructions carefully, find a quiet place for about 30 minutes, and play as yourself. Our guide on what a good score in the Pymetrics balloon game means explains how results are used.

Sources

• pymetrics, candidates page (pymetrics.ai, archived October 2024; pymetrics is now part of Harver): no right or wrong way to complete pymetrics; behavioral data used to match candidates to jobs; about 30 minutes; games cannot be paused and last 1 to 3 minutes each; be yourself.
• Lejuez, C. W., et al. (2002). "Evaluation of a behavioral measure of risk taking: The Balloon Analogue Risk Task (BART)." Journal of Experimental Psychology: Applied, 8(2), 75–84.