As the United States enters the business end of its Presidential elections, we turn the spotlight on how the major parties plan on using market research to deliver them the edge at the ballot box.
Who actually matters
For political parties, the marketing problem to be solved in very simple terms is this: who are the voters who will decide the election, where can you find them, what do they care about and how will you reach them.
A good political campaign often uses a mix of quantitative and qualitative market research to achieve this.
However, in recent years the political market research game has changed substantially.
The rise of digital tools and the explosion of unstructured big data that can be bought and sold privately have changed the nature of research.
The voter file
Parties have historically had access to a voter file (US, UK, NZ and Australia). A voter file is a list of all voters for the electorate a candidate is standing in.
The voter file has been standard in politics for decades. It gives parties access to the basic demographics of their voters such as their name, age, address and household composition.
At a basic level, the voter files allow parties to analyse and target clusters of people that share similar characteristics on a simple demographic level.
Historically, political parties have used insights gleaned from polling and focus group research to target people through the voter file.
However, over the course of the past few years, access to new data sets has enabled political parties to gain even better insights.
More insights into attitudes
One big change has been to source data from external data providers and attempt to correlate it to information on voter files.
In some instances, this has added 30 or more new datasets from which parties can predict future voter behaviour.
Parties can now combine demographic data with a better psychographic understanding of key segments of voters. They can delve deeper into the attitudes that motivate behaviour.
Pin point accuracy
New data has also given better insights on how to reach key segments. For example, during the US elections in 2012, data received on TV set top boxes was correlated against the voter file, revealing to the Democratic Party the viewing habits of their target audiences.
This improved their targeting by knowing precisely which TV shows their segments were watching, allowing them to deliver content directly to these people with minimum wastage on advertising costs.
Digital platforms such as Nation Builder are also being used to gain better integration with a party’s activists. Organisations can upload their voter file and the software can trawl social media to relate a person’s social site with their individual record.
Digital tools have also allowed parties to better measure the best email (EDM) fundraising techniques through A/B testing. By testing different messages on smaller audience sizes (through analysis of open and conversion rates), the parties can estimate the content that will deliver the most ROI.
Room for improvement
Why do people vote a particular way? Political research has suffered from the vagueness of which attribute decisions contribute to the final outcomes at the ballot box.
Most published opinion polls (and private research) ask respondents to rate the issues in the order of importance and which candidate they feel is better at managing them.
However, they don’t demonstrate the trade off voters will make to arrive at their final decisions. Ultimately, voters are weighing up a set of different attributes when arriving at a decision.
With the election hotting up, the market research battle will be one to watch closely.
Written by amarg81