Cllr James Hellyer is a Conservative councillor in Torridge.
Over the last year, we’ve fought five by-election campaigns for Torridge District Council. In the course of these, we’ve lost our Conservative majority, fallen into no-overall control, before regaining and increasing our majority.
We lost the first two of these by-elections heavily, and one was a Conservative defence. The next were wards our councillors had written off as unwinnable for the Conservatives. We won. The last by-election was in Holsworthy this month and we held the ward.
So, what happened to change us from an association who lost by-election campaigns, to one with a winning streak? It’s not the national mise en scène, which has remained consistently depressing since June last year. Rather it was a complete retooling of our campaigning methods.
Torridge and West Devon has a parliamentary majority of 20,686 nowadays, so there are plenty of Conservative voters out there. They’re reliable at turning out for general elections. Leafleting the constituency was the main campaign activity in 2015 and 2017. But there’s been no real attempt at canvassing since the seat was won back from the Liberal Democrats in 2005.
For the first two of our council by-elections, we followed the playbook we knew: we leafletted the wards and hoped for the best. This, as it turns out, was not a good idea. Conservative voters are much less motivated to vote in local elections than Labour or Liberal Democrat supporters. They turn out with little encouragement. Just doing the leafleting not only didn’t get our supporters out, it probably reminded some opposition supporters to get out and vote against us.
A big parliamentary majority was no help when it came to local elections because our supporters weren’t especially motivated to vote, and we didn’t know who they were or where they lived. That had to change.
Many excuses could be made for our poor performance. A difficult national picture. A popular Liberal Democrat in one seat. Other parties having more boots on the ground. In Torrington, the Liberal Democrats brought in outside help and spoke to every postal voter, whereas we only wrote to them. In another, an independent who was the heart of his village stood outside the polling station all day talking to every voter before they went in. We were defeated because we were out-campaigned. If we didn’t want to go into all out elections in 2019 after a morale sapping series of defeats, things had to change.
We had plenty of notice of the next by-election, which was triggered by the death of a UKIP councillor. Conventional wisdom was that Bideford East would be a tough seat for us to win, despite it only having elected UKIP or Conservatives in the last two councils. I was selected as the candidate and rather than rush to call an election, I decided to do some groundwork. The first thing I did was enter the marked electoral registers for the 2017 General Election and the County Council elections on VoteSource. This showed me who did and didn’t vote, and who was most likely to vote in a district council by-election.
Using this data, it was easy to identify people who don’t vote. If someone didn’t vote in the 2017 County or General elections, they would be unlikely to vote in a district council by-election. Whereas people who voted in the County election, which was very Conservative, would be far more likely to vote – and vote Conservative in our by-election. So, we could target our activities where they were most likely to yield results.
I set a target number of pledges based on the worst-case scenario – a 30 per cent turnout with only two candidates. That gave me something to aim for. We started from a low base on VoteSource, with barely 100 identified Conservative voters. We needed 675 to win. A daunting target.
With the election not called, I identified areas where lots of people voted in local elections and began a programme of knock-and-drop survey canvassing. It turned out that this was a fruitful approach. Conservative voters were generally happy to tell you they voted Conservative, which issues they were concerned about locally, and even provide the gold dust of modern campaigning – e-mails and telephone numbers. Even those who refuse to complete a survey can give you a voting intention by the way they refuse…
The 0 (would never vote for them) to 10 (definitely vote for them) canvassing question also did a much better job of identifying people who were inclined to vote Conservative than the old P or C system ever did. This way your voters self-identify, rather than leaving you reliant on canvassers abilities to discern voting intentions.
This process slowly built up the pledge base and gave me the issues local Conservative voters cared about, so that my election literature could talk exclusively about the issues that my target audience cared about.
By the time the election was called the pledge base was over 300. Nowhere near my target, but still much improved over our starting point. Personalised letters were then sent to every pledge in the days before the election, stressing how important their vote was and the risk of not voting. Given the examples of other recent by-elections that turned on just seventeen votes, or in one case a coin toss, their vote really could decide the election.
The pledge list also meant we had something to do on polling day, which was a first, as we made sure our known supporters turned out. Meanwhile Labour and the Liberal Democrats spent the whole election ploughing around large housing estates, where hardly anyone ever voted. They didn’t canvas. They didn’t knock on doors. They didn’t have a get out the vote operation.
Having won this election, I acted as agent and campaign manager in Hartland and Bradworthy. In this case a Liberal Democrat who stood down due to ill-health. Again, our councillors deemed the seat unwinnable, when in fact we’d simply never contested it before. It turns out that if you give Conservative voters a Conservative candidate, they turn out and vote for them rather than fall into the habit of voting Liberal Democrat or Independent. A lesson for next year – contest every seat.
This election was in the immediate aftermath of Chequers. The Liberal Democrats were campaigning on an Exit from Brexit. We stuck solely to local issues in our literature. The only time we discussed national issues were in some doorstep conversations when Conservative supporters expressed their disquiet at the Chequers proposals. Rather than debate the merits of the Prime Minister’s plan, I simply told them that the Liberal Democrats were using by-election victories in areas that had voted Leave to claim that the people had changed their mind and wanted to Remain. By not voting for us, they were simply giving Vince Cable a boost. On polling day all the people who had been told this, turned out to vote, but voted Conservative. Again, a pledge target, survey canvassing, 0-10 canvassing, and staying local worked. A Conservative gain from the Liberal Democrats.
Finally, we had another by-election in Holsworthy. We followed the same game-plan and stayed local with our talking points. Labour and the Liberal Democrats again campaigned on national issues. We talked about recycling and the sorting out the local football pitch. They talked about Brexit. Once again there was an exit from Liberal Democrats being elected.
Over the summer I helped in council by-elections all over Devon and Cornwall. In the elections where we put out leaflet after leaflet, but didn’t talk to anyone, we lost and lost heavily. In the elections where we used survey canvassing and 0-10 model, we either won, or left that ward in a much stronger starting position for the 2019. There’s a lesson in there somewhere.
Building up a database of voting intentions on VoteSource really does work. A lot of people don’t like going canvassing, but survey canvassing is much less daunting and gets back lots of ground level information, even if they don’t say they’ll vote for you. I would encourage everyone facing all-out elections in 2019 to start canvassing now if you haven’t already. Build up your pledge base and identify the issues your voters care about so you can target literature next year. It will make things a lot less stressful next May.