Penguins famously and adorably form huddles to thrive in the Antarctic cold. Bank branches in the UK do the same. I can say this with 100% authority because I have worked with the canonical data for currently trading banks and those that have shut in the last five years. You can check my workings, in fact you can do or commission whatever analyses you want because Geolytix are making the last five years of canonical UK bank branch data fully open under a CC0 license. Maybe you think measuring the penguinness is best measured with a simple median nearest neighbours over time (my preference), or Moran’s I of hours, or maybe Ripley’s K. Without these data we are releasing today you really won’t get far, with it and the statistical world is your oyster.

Access to cash and wider financial services are topics exercising many. The latest government consultation closes on Thursday 23rdsept. We hope making these data open and available to all helps everyone involved make evidence-based commentaries and decisions.

The first immutable law of data analysis is inputs drive everything. No data equals no analysis, no models and no stories. Are banks preferentially closing in less affluent areas? Are Post Offices well positioned to serve the newly bank-branch deprived? How many people now live a mile further away from a branch than five years ago? Are particular brands behaving differently? What characteristics unite towns with no branches? Without knowing where branches are and have been you really have no chance answering even these simple questions, let alone constructing and conveying the rich customer and business story of how networks evolve.

Newly closed banks in Acton - the above row is when the premises were banks, and the below row is what the premises are now Source: Google Street View 

This newly open data has all UK branches (including those now shut) that have traded over the last five years, and comes with lat/longs, trading hours, and closure date where relevant. We have also added distance to nearest Post Office branch (trading as of Sep 2021), Geolytix retail place type (town centre, retail park, urban hub, village centre etc.), and Geolytix shopper town name.

For the record the briefest of answers to those questions above are No, Oh Yes, about a million, a bit but not massively, smaller towns in the shadow of bigger ones.

Below, maps visually show what the data is telling us - and this is where the penguin reference really makes sense. Banks that are in the outlying villages/towns are typically the first to close, with those in a more central location remaining open for longer.


You can download the data, supporting documentation here and ReadMe. Please @Geolytix on LinkedIn & Twitter to let us know how you are using the data.

Please email info@geolytix.co.uk for any more information.


Blair Freebairn, CEO at Geolytix

Photo by Martin Wettstein on Unsplash

The image above shows that typically it is the banks in outlying towns or villages that close, however like the penguin, those huddling in the middle seem to survive 
As with the image above, here is an example of 'penguinness' - where those banks that are in the central location are those that survive, with the outliers closing first