The IKEA Effect in Data Engineering

The IKEA effect is where one places greater value on something they built, or partially built themselves. It’s named after the Swedish firm who famously provide their furniture as flat-pack, requiring assembly by the customer.

Data Engineering is just as susceptible to this as anyone! We can often favour building our own solutions, both big and small, when the better option would have been to leverage an existing tool (paid or open source), or asking a more appropriate team to take on that work.

For example, we may feel we can build a data catalog that is perfect for our organisation with not too much effort, rather than using an existing solution. But, we tend to be over optimistic on the costs of that (both upfront and ongoing).

We then compound that by convincing ourselves that it is better, it needs just a little bit more then it will be perfect! And so we keep investing in it, over and over again, taking time away from another project that could deliver greater value for the organisation (see also “sunk cost”).

I’ve been burned by this in the past, and as I’ve become more experienced I’ve been able to spot the times when myself and others are suffering this effect. I now try hard to avoid building things ourselves unless they are truly solving a problem that is unique to us, and/or we are solving a problem in a way that adds greater value (or reduced costs) to the organisation.