There is an XKCD joke about maps that makes the point that if you present data about the activity of humans without adjusting it for population, the map simply shows where people live.
The National registry of Iceland publishes market data on the urban housing market in Iceland. They have charts on their page for the number of sales, but as it uses only the raw numbers, those charts too really only show where people live. Visible in inset chart A., the answer is: In Reykjavík.
Stories in data often only emerge after the data has been normalised to an index, or adjusted to per-capita, or for inflation. But the National registry also publishes the data shown in inset chart B., the number of residencies per municipality.
Combine these to end up with the large chart, the number of sales contracts for each municipality as a proportion of the size of the market. The percentages may be slightly off, but what emerges are at least two stories:
A significantly higher portion of the real-estates in Hafnarfjörður changes hands every year than in the capital. But why?
The difference in Reykjavík and Seltjarnarnes has vanished post-financial crash. Is Reykjavík still to pick up steam or has Seltjarnarnes become a hotter market? Who knows.
What we do know is that without the adjustment, we were not able to see far enough to even ask these questions.