Faculty Customer Intelligence
Getting Started¶
Pass a sequence of customer postcodes to analyse()
:
>>> from faculty_extras import customerintelligence
>>> postcodes = ['AB10 1AA', 'AB10 1AB', 'AB10 1AF', ...]
>>> result = customerintelligence.analyse(postcodes)
Exploring the Results¶
The analysis result returned from analyse()
includes methods to explore how
similar people in different areas across the UK are to your customers. To
visualise overall similarity by parliamentary constituency, run:
>>> result.similarity_map()
It is also possible to visualise similarity by other aggregations, such as
by local authority or census output area. To discover all the possible options
to similarity_map()
, check its documentation.
similarity_map()
returns a folium Map
object. This can be rendered as
an HTML file with its save()
method:
>>> map = result.similarity_map()
>>> map.save('map.html')
To get the values of similarity to your customers at various levels of aggregation, including by postcode, postcode sector and outcode:
>>> result.similarity_score('postcode')
similarity_score()
also can take an optional sort
argument, allowing
you to presort by similarity score and then retrieve, for example, the 500
most similar postcode sectors to your customers:
>>> result.similarity_score('postcodesector', sort=True).head(500)
Read/Write of Results¶
You can write reports to disk with to_json()
:
>>> with open('report.json', 'w') as fp:
>>> result.dump(fp)
and read with:
>>> with open('report.json') as fp:
>>> result = customerintelligence.AnalysisResult.load(fp)