Postcodes, GIS and open data

Behind the scenes in Community Insight, we’re always improving the system functionality and performance. Generally users notice the difference through new features such as stock types and hotspot mapping. Or we notice the difference in terms of it being easier to do things like update the data or launch a new version of Insight for housing providers in Wales.

We also release continual performance enhancements, which can really improve the user experience, such as faster loading of maps and and property assets (even on IE7 and IE8 …). Based on user-feedback, updating the postcode look-up data is one of those performance improvements that users really notice. Which is why we’re excited to have just updated and strengthened the postcode data we hold in the system.

Postcode data is now published on a quarterly basis by Ordnance Survey as open data, and we are always rolling-in the latest version of this information to ensure that we know about the addresses used by housing organisations. This data is linked to a coordinate (based on the centre of the postcode – unfortunately OS decided the postcode boundary was too valuable to publish as open data), which we use for locating property assets on the map. We can also take in property locations based on latitude-longitude (or easting-northing) which we we will roll into a future release.

Based on the postcode location, we can also identify which standard areas each postcode is in – such as wards, Output Areas, Super Output Areas, Local Authorities etc – and use this to generate community profile reports for Insight users. For this, we use lookup tables published by the Office for National Statistics. But we now also top this up with some GIS jujitsu to match previously unmatched postcode centres to the standard area boundary definitions.

The bottom-line is that our postcode matching is getting close to the magic 100%, with users able to map their properties in less than 1 minute. And as a bonus, Community Insight can also help organisations to identify errors in their underlying property postcode data – as we know some users are already doing.