Planning poker is a popular tool for estimating effort in projects today.

**Image Source**: Rachmaninoff

In planning poker, every member of the team is given cards with numbers to estimate the effort involved for a particular task. What interested me the most was that these numbers were not linear. Instead, they followed the Fibonacci sequence – 0, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, 144 and so on.

In a Fibonacci sequence (which starts at 0, 1), we get the next term by adding the preceding two terms. It is interesting how this sequence starts slowly, creeps upwards and catches us by surprise. It starts off flat, slower than even a linear series. But by the 14th term (233), it leaves behind the series of squares and by the 23rd term, it overtakes the cubic series. I sketched this graph to understand it better. I was taken aback by how sharply it climbed.

Systems behave similarly to a Fibonacci series. At the beginning, their complexity climbs at a manageable rate. With a few more moving parts thrown in, things can quickly get out of hand.

Linear thinking comes to us naturally, but we are terrible at non-linear thinking. For the same reason, it is easier to multiply a number by 2 in our head than to find its square. This problem lies at the root of why we are terrible at estimating long-term projects. The higher the project’s magnitude and complexity, the bigger our errors in estimation are. It would serve us well to correct these tendencies and develop non-linear approaches while estimating effort.

And using differently numbered card decks is certainly a good start.