Andrew Kennedy is the Group Agent & Campaign Director in West Kent. He blogs at

With much of the country focused on May’s elections, this is a good time to examine what we do on the doorsteps, what we need to achieve and how the data we collect should be used to maximum effect.

One of the challenges we face in West Kent is to convince our more traditional campaigners that canvassing is no longer about having a detailed discussion on every doorstep with the aim of converting the misguided into changing their ways. Similarly, we need to persuade our newer helpers that canvassing is a simple data-gathering exercise, and they do not need detailed knowledge of every aspect of party policy.

With so many Associations being wholly reliant on volunteers, there is a danger that some may lack people with the training and knowledge now needed to gather and store data and how it should be used to maximise our chances.

This includes:

  • How to choose the targets and why certain groups are important.
  • The questions to ask to maximise accurate and useful responses.
  • The medium used to ask those questions (postal survey, doorstep, phone or email).
  • How to record and store that information legally and in an accessible format.
  • How that data can be used to build future success.

And when I refer to “future success” I don’t just mean electorally – though that is clearly our primary concern – but also progress in terms of recruiting members, donors, activists, postal voters and possible even finding future local government candidates.

In a previous era, when 90 per cent of people voted either Conservative or Labour, a simple “may we count on your support” was probably sufficient; you were either with us or against us. This is no longer an option. Voters, even those who identify as Conservatives, happily switch their votes at local, parliamentary, EU and Police Commissioner elections, and even at the same election – as anyone who witnesses the bizarre vote splitting in multi-member wards will testify.

In this climate, we must be smarter and more accurate. Understanding a voters’ second preference is probably just as important as recording their first, and the ability to target nuanced Get Out The Vote (GOTV) messages to different groups is now vital in delivering victory in marginal council wards.

At West Kent Towers, we are now working on our targeted GOTV material for delivery during polling week. For us, the days of a simple “Don’t forget to vote on Thursday” are long gone. In our target wards, our teams will be delivering at least seven versions of GOTV material to –

  • Strong pledges who always vote
  • Weak pledges who sometimes vote
  • Conservative postal voters
  • UKIP voters who would prefer a Conservative councillor to Labour or LD (Q voters)
  • Labour pledges who prefer a Conservative to Lib Dem or UKIP (T voters)
  • Liberal Democrat pledges who prefer Conservative to Labour or UKIP (M voters)
  • Uncanvassed voters in target mosaic groups

For this to be effective, however, we must have the data in the first place, and this requires our doorstep and telephone teams to have the training and confidence to ask the right questions and enter the correct codes on the canvass sheets.

Other initiatives we are (or have been) working on include:

  • The time of day when people vote: by time-stamping tellers sheets and recording this data, we can build up a picture of when people vote (morning, lunchtime, afternoon evening). This allows our teams to focus GOTV efforts where needed, for example not calling on evening voters in the morning releasing time to target those supporters who we need to turn out before dark.
  • Variable paragraph pledge letters*: by ensuring we ask the same questions by post, email, telephone and doorstep we are able to build a database of the main issues which concern individual voters. This information is then recorded and used to produce personalised letters from the candidate dealing specifically with the issues voters have told us concern them.
  • Street-specific newsletters*: again, by recording specific issues which affect roads or groups of residents, we have been able to produce a generic newsletter with a tailored “lead story and headline” for each road or target group of residents.

(*So far we have only had the resources to do this for by-elections, but where we have the impact has been very considerable.)

Data is key to almost everything we do. Knowing which voters will never consider voting Conservative (or who will vote for whichever party is best placed to defeat us) is just as important as knowing who we need to target. I will always remember Britain’s longest-serving council leader, the late Mark Worrall, telling me “the first step to winning an election is not to do anything which sufficiently irritates your opponents to motivate them to go out and vote against you.” And there is nothing more likely to motivate an apathetic Labour voter than having a constant stream of Conservative literature landing on his or her doormat. It is for this reason I don’t like window posters, street stalls and loud speaker cars – they are indiscriminate in their audience and thereby reduce our advantage.

And after the dust has settled, good quality data is the gift that keeps on giving.

  • Overlaying the pledge base with the marked register, you can identify pledges who did not vote and use peacetime to sign them up for a postal vote.
  • Identifying your strongest and most consistent pledges you have a pool of support in which you can fish for helpers, activists, members and even future local government candidates.
  • Each year you can target one or two well canvassed wards and send every pledge a book of Christmas draw tickets; the response rate may only be five per cent but the exercise covers costs and provides an easy gateway to build your donor base for the future.

All of this, however, relies on good quality data, and that means we have to spend time and effort explaining to our volunteers why accurate data really matters, and providing them with the skills and knowledge that they need to ask the right questions and in the right way.