Learning the lessons of the 2010 general election

I’m currently working with Shaun Roberts on updating the old ‘how to win your ward’ ALDC publication. The publication has a long heritage going back through the Liberal Party and there is an interesting indication of how the party has developed as you follow the literary thread from the old Community Politics pamphlet by John Smithson via Chris Rennard’s Winning Local Elections manual and other publications and ending up with the three volume Winning Local Elections series by various people including myself.

As the publications have got longer there has been a steadily increasing emphasis on the technicalities of election campaigning with the local campaigning – and community politics – getting rather overshadowed.

The draft table of contents Shaun and I are working on deliberately tries to take a big step away from that, returning to the local community politics roots. I’m sure some of that clarity will get worn away at the edges as the writing process commences but if nothing else I hope we can get over the spirit behind the original invention of Focus newsletters – that is, you do campaigning and therefore you have news you want to impart and feedback you want to get, so you do a newsletter to let people know.

That’s not to say that there isn’t plenty to think about and improve on the election campaigning side too, and one aspect of that I looked at in my most recent column for ALDC’s Campaigner:

Learning the constituency campaigning lessons after the 1997 and 2001 general elections was fairly easy. Almost without exception, those seats which followed the key seat best practice plan did well and those which did not, didn’t. Most strikingly, the two incumbent MPs in 2001 who insisted on doing something different both lost even as seats elsewhere which followed the template were gained.

However, many people’s immediate reaction after the May 2010 election is one of bafflement. Those simple patterns, which also mostly held in 2005, seem to be shattered. Some intensive ‘traditional’ key seat campaigns scored spectacular swings or saw incumbents re-elected with increased majorities, but a good number of other campaigns which superficially followed the same template did not.

So what was going on?

A significant part of the answer, I suspect, comes from campaigns that were so hard-pressed to meet the necessary scale of a winning campaign that the quality of the campaign suffered. Perhaps the clearest example of this is with direct mail.

The volume of direct mail which key seats aim to do has hugely increased in the last ten years – and rightly so, given the successful experiences of direct mail intensive campaigns in particular seats in 1997 and 2001 in particular. Direct mail gets better readership and response rates than leaflets and allows for the variation in message that an increasingly fragmented electorate requires.

Moreover, the tradition used to be that volumes of direct mail were primarily driven by the volumes of canvass data and a very limited amount of demographic data held in EARS. So a third party squeeze letter would go to the identified third party voters from canvass returns and no-one else – even if that might mean the letters went to only a small fraction of the third party vote.

In order to deal with this mismatch between volume of data and in EARS and volume of letters needed to bring about the right political outcome, campaigners increasingly looked to other ways to increase the reach of letters. Mosaic has been a valuable part of this, as has been categorising areas as Labour or Conservative.

Many of these concepts and lessons were brought together and then more widely publicised following the party’s review of key seat campaigns in 2005. It was in the summer of that year that I coined the phrase “the pool” and “pool mailings” – words which are now uttered repeatedly by campaign organisers as they stand cursing over jamming laser printers or misfiring stuffing machines.

The idea behind the pool was simple: rather than let direct mail be driven by the volume and type of data immediately to hand, have direct mail driven by the political needs (particularly in terms of volume) and then work backwards to figure out what data to use. It meant that many campaigns started to be really stretched to do far more direct mail than they had done before.

It was the right lesson based on successful experiences such as Harrogate in 1997, Chesterfield in 2001 and Hornsey and Wood Green in 2005.

It has also come at a cost. Often campaigns have been so stretched trying to hit the necessary volumes that the quality and variation of the direct mail suffered. That has been both because of the increase in volume of direct mail and also because of the increase in the range of our key seats, with the growth in winnable seats taking in seats with far lower levels of organisation than would have previously been considered necessary.

So whilst in Chesterfield in 2001 what we would come to call a pool mailing might have included letters to new voters, letters to pensioners, letters to Tory areas, letters to supporters and letters to petition signers all in the one delivery, by 2010 in many campaigns a pool mailing meant pretty much everyone in the pool got the same letter.

Similarly, whilst Chesterfield in many ways pioneered widespread double-delivery, that in origin was driven by necessity – too many targeted messages to get out in direct mail and not enough delivery slots. Over time double-delivery has mutated from an enforced economy of delivery in order to ensure all the different niche mailings go out into an almost automatic doubling up where letters are done because letters are done.

One clear challenge for the future therefore looks to be learning how to maintain the volume whilst increasing the quality – both in terms of better and cleverer use of data to make-up the pool and also in more varied and more carefully targeted letters sent out to the pool.

We have won a string of constituencies assisted by large scale, high quality direct mail campaigns. When the letters are good and varied, people are happy to receive far more than our deliverers and bank balances can put out. Where the letters are formulaic, repetitive or produced in a rush, people’s tolerance drops swiftly.

I’m a great believer in the power of direct mail, because I’ve had the joy of hearing the Returning Officer read out results in many elections won with the help of direct mail. Collectively though having cracked the quantity issue in recent years, we now need to crack the quality one too.

8 responses to “Learning the lessons of the 2010 general election”

  1. I think any new manual should emphasis the build up of data in the interviening years between elections. We appear to be going as a party for volume over quality of literature in some cases in election campaign. The numbers of ordinary people saying they do not want any more literature from us ever again is on the increase. Canvassing is something of a luxury during an election campaign.

    The failure to get a proper campaign team in place early enough appears also to be a problem. I will spare the blushes of the constituency in particular, but a broad spread of competent volunteers that are able to take the constituency forwards is crucial. It cannot be done by paid staff alone.

  2. My take on the election is rather more straightforward. Our national campaign gained a high degree of momentum during the second week, began to lose it at an even faster rate during the third week, and the electorate had our leadership pretty well sussed by the fourth week and expressed their views clearly on this matter at the ballot box. Labour did much better than anticipated and the two party leaders – who both did much worse than the reasonable expectations of their members – formed a losers’ coalition, not least, perhaps, because the alternative to a coalition would have been their rapid replacement by their respective parties.

    So far, we have had an “emergency” budget to forestall a non-existent crisis in the bond market which is designed specifically to suck demand out of the economy as the Tories engage in a slash and burn exercise in social engineering which puts Margaret Thatcher’s meagre efforts to shame.

    The best path that we Liberal Democrats can follow if we wish to improve (or, more likely, at best match) our performance in the next election is for our MPs to reject the budget and the vainglorious attempt at constitutional change in the increasingly vain hope of getting a yes vote in any referendum on AV, Certainly, the coalition offers no obvious prospect of improving the representative nature of the House of Commons, nor strengthening the position of MPs the better for them to curb the activities of a know-nothing/cut everything typically Tory mutation of a coalition government.

    Oh…. and we might as well have a leadership election so that we can, at the very least, be certain that our Leader, and the team which surrounds him, does, in fact, effectively represent the views, values, beliefs and attitudes of our membership as a whole.

  3. Canvassing should not be regarded as ‘a luxury during an election campaign’. We need to build canvass data during ‘peactime’ / local elections. Then, with the guidance of EARS/Mosaic data, focus canvassing on specific wards in the last few weeks/months. But of course, as implied above, this is impossible without sufficient volunteers. Strong branches and steady recruitment are therefore essential. Campaigns for parliamentary seats have to be regarded as an ongoing activity. Apologies if this seems to be stating the obvious.

  4. The seat where I campaigned did the full-on literature campaign. It didn’t work. Despite Labour’s pathetic efforts on the literature front, they seem to have had much better voter ID and won by over 3000 votes. The comments you make about improving the quality of our direct mail miss this fundamental point.

    Labour were whizzing backwards and forwards across the constituency for 18 months with their phone banks. Come polling day they knew precisely where their voters were and were able to target them effectively. Targeted mailings by Unite and Unison to their members didn’t help either.

    Because of the shear volume of literature we were pushing through the doors, it was impossible to carry out a proper canvass of the constituency. The LibDem phone banks were next to useless by the way; canvassing known supporters as antis for example. We KNEW that the lack of canvass data was one of the reasons why Durham City was not won in 2005, so why was this not addressed before the 2010 election???

    I have to say bluntly Mark that it will be difficult for the local party where I was working to ever again trust advice from Cowley Street on general election campaigning. I suspect the temptation will be to fund our own self contained campaign rather than flogging our guts out by following plans that failed so miserably in 2010.

    Sorry if this sounds angry and bitter; that’s because we followed our campaign plan to the letter and still got beat.

  5. James: having played a small role in the Durham campaign in the run-up to 2005, I share your frustration at not winning and seeing Carol in the Commons – though I appreciate it’s much more frustrating when it’s your own constituency, where you live.

    A few observations on the specifics:

    a) I agree with you about the need for more canvass data overall in the party’s campaigning and it’s a lesson many people saw in 2005. That’s one of the reasons why the party’s campaign targets for key seats placed a greater emphasis on canvass data when they were revised after the 2005 election.

    b) And that point highlights one of the frustrations to me of these sorts of debates, where a mythical “Cowley Street” campaign plan is so often dismissed as only being about lots of leaflets. In fact, the activity targets included plenty on doing residents survey, collecting canvass data and building a wider team for example. Perhaps that wasn’t always reflected in the discussions in your local party (I don’t know, so can’t comment on that), but if it wasn’t then the real question it seems to me to ask is why these wider aspects got lost somewhere in translation.

    c) Both Labour and the Conservatives have a much higher core level of support than we do. So whilst it is sensible to see what we can learn from their techniques, I don’t think a campaign plan that requires us to work harder than them is obviously flawed; it may well just be a reflection of the fact that we usually start from further away from the winning post than the other parties.

Leave a Reply

Your email address will not be published. Required fields are marked *

All comments and data you submit with them will be handled in line with the privacy and moderation policies.