We design and build

Marketing databases
Single customer view
Prospect pools
Data analysis
Reporting automation tools

We support and maintain

Database hosting
Database maintainence
Campaign planning
Data processing
Business intelligence reporting

We advise and consult

Project management
Data sourcing
Data product feasibility
Loyalty/CRM strategy and implementation
Supplier evaluation
Data planning



Approved Apteco FastStats Partner since 2007

Asthma UK

Targeting financial support from non financial supporters

Introduction

Asthma UK is the charity dedicated to improving the health and well-being of the 5.4 million people in the UK whose lives are affected by asthma.  They work with healthcare professionals and researchers and raise funding in a variety of ways, ranging from one off donations, to weekly lottery and raffle tickets, as well as organising events.

Because of its investment in research it provides detailed content and information via its website and call centre.  These booklets are available free of charge to anyone who requests.  This group is known as “Information Requesters.”

Asthma UK was conscious that they had a disproportionate volume of Information Requesters on their database, who did not donate to or support the charity financially.  They wanted to understand the profiles of this group to see whether it was possible to change their behaviour.

The requirement

The client gave us a very clear brief and formulated a set of tightly defined analysis objectives:

  • To profile Asthma UK’s Information Requesters versus donors:
  1. Demographically
  2. Behaviourally
  • To identify the means of converting Information Requesters to support the organisation financially
  • To build a predictive model to identify those Information Requesters who have the
  • greatest propensity to support Asthma UK financially.
  • To understand whether non-financial supporters eventually become financial supporters

The challenges

When undertaking analysis of this kind Aszent aims to provide a full financial analysis of the value of specific segments, which includes analysing response rates, the cost to serve, cost of operations, etc.  However, this information was not available and so the analysis could not provide insight on how much budget to allocate to the groups we found.

The charity did not hold much common data on its supporters, making profiling and predictive analysis very difficult, so we looked to an external source for enriching the data and making the analysis meaningful.

Whilst undertaking the data preparation for the analysis we uncovered a high non-match rate on over 40% of the database.  On further investigation we discovered that these records related to business addresses, which verified the popularity of Asthma UK as a source of information.

What we did

We divided the Asthma UK database into logical groupings:

  • Group 1: Active Supporters – 45% of the database
  • Group 2 : Active Supporters once Information Requesters – 1% of the database
  • Group 3: Information Requesters – 54% of the database

Predictive modelling was undertaken on Groups 2 and 3, the records matched to Experian’s National Canvasse database, and factoring undertaken to determine the most predictive variables to include in the modelling process. Due to a large non-match we developed two sets of models.  After analysing their key characteristics, we were able to identify potentially predictive variables for the subsequent model to be developed. Predictive models were built using logistic regression, then validated using data set aside from that used in the initial build.

We uncovered some interesting facts on the groups, which we used to devise actionable recommendations for Asthma UK:

Our models showed clear discrimination and that it was possible to identify best IR prospects for conversion to active, financial support.  However, as with all communications, much depends on timing, message and antecedent states.

Best prospects tend to be older, more affluent and are more likely to have enquired via telephone than through the web – motivations for contacting the charity were very clear as they were sufferers or related to someone who suffered from Asthma

Active Supporters were much older and more affluent than Information Requesters, with a higher proportion of males than in the Information Requester groups.

We noted that high do not contact rates were recorded against the those with high predictive scores, which may have been a consequence of the blunt wording on the website.

We scored up the entire database so that Asthma UK could make selections based on the targeting model.

The benefits

Asthma UK is now able to make decisions based on projected ROI, i.e. whether to include all IR individuals in the existing campaign schedule, or to focus only on those where there is a higher propensity to convert to an Active Supporter.

Not only this, but the application of the scores means that Asthma UK can now test message, creative, frequency, selections, etc in a controlled manner

The  we provided through the external data enrichment gave the organisation the ability to review the proposition and creative treatment of communications to reflect the characteristics of the groups – leading to an improvement in response/conversion results.

OH MY GOD! This is so fantastic. Thank you so much for all your work on this. Life will be so much easier now. Very excited… — Rachel Hirons, Insight Analyst, Global Radio


Call Aszent on 07887 653595