Heatmap to show eta collection times from postcodes
- Thread starter mattyg
- Start date
Making some pretty big assumptions over how you want it to look and how your data is stored in the CSV:
If you are using a real map as a background you'll need to split it up in to grids. You could then use Excel to import the csv coupled with a pair of lookups to assign where in the grid the specific values should go, then plot it as a suitable chart type with your image in the background.
If you are using a real map as a background you'll need to split it up in to grids. You could then use Excel to import the csv coupled with a pair of lookups to assign where in the grid the specific values should go, then plot it as a suitable chart type with your image in the background.
Soldato
How big is the data set?
Excel could be used or python and something like Seaborn.
Excel could be used or python and something like Seaborn.
Soldato
What sort of heatmap are you after - It is tabular style with hubs against (collection) time/slots and essentially the best and worse times for collection at a given hub? Or is it a heatmap over a map (of the country) with hub locations showing worst hubs for collection times?
Former could probably be done in Excel; the latter might be achievable using a third-party like Maptive or, building something using PowerBI + Azure(/Microsoft) maps or with a mapping API like Google Maps or Mapbox.
If you can post a snippet of anonymised data and exactly what you're wanting to achieve, then i'm sure someone here would be able to advise the best/better solution.
Former could probably be done in Excel; the latter might be achievable using a third-party like Maptive or, building something using PowerBI + Azure(/Microsoft) maps or with a mapping API like Google Maps or Mapbox.
If you can post a snippet of anonymised data and exactly what you're wanting to achieve, then i'm sure someone here would be able to advise the best/better solution.
about 1700 postcodes
it doesn't need to be interactive just a visual representation.
We've been asked to produce a heatmap of collection times at our clients customers. So lets say a route goes from Birmingham at 6pm and terminates in milton keynes at midnight. Then Birmingham would be early and milton keynes would be later. But a route that starts and ends in manchester but goes via wakefiled would be later but Wakefield would be earlier. if that makes sense.
Looked at third party Mapline which does heat maps but seems wastefull at £500 for a one off. But doable if cant find a cheaper solution.
The idea is the further from the nearest hub, Collections will be earlier due to geographical locations.
Soldato
I would do this and adjust to suit your situation (its very similar but best example I could find quickly).
Soldato
Stay well away from any kind of geography visualisation. I've coded stuff "by hand" in Python and used Metabase, PowerBI, Looker and Holistics. Putting pins on a map is easy but as soon as you want to see regions (political, postcode, delivery hubs etc.) it's a rabbit hole. At the end are some very impressive and functional dashboards but if it's not your full-time job try not to get drawn in 
If your users already have a mental model of the geography of the Birmingham, Manchester etc. routes then something like gpuerrilla suggested would be my preferred option (if I didn't already have a BI tool configured). You might need several steps of manipulation/calculation to get the CSV into the right "shape" to visualise.
If your users already have a mental model of the geography of the Birmingham, Manchester etc. routes then something like gpuerrilla suggested would be my preferred option (if I didn't already have a BI tool configured). You might need several steps of manipulation/calculation to get the CSV into the right "shape" to visualise.
Last edited: