Yesterday, I recounted the problems in the national Conservative machine, particularly at CCHQ, during the General Election. Those issues affected the strategy, messaging and targeting at the highest level of the campaign. This second piece in our mini-series looks at the campaign machine on the ground.
Drawing on interviews and information from scores of Conservatives around the country, ranging from MPs and party officials to candidates and activists, ConservativeHome is able to give an insight into the aspects of the ground campaign that worked and those that didn’t.
Yesterday’s exploration of the national campaign revealed some issues which had knock-on effects on the ground – not least in terms of targeting. The decision to call a snap election, made at short notice and kept secret even from most of the Cabinet until the day itself, necessarily plunged all levels of the Conservative Party into frantic action to make ready for an election that was already upon them. The pressure of the campaign would reveal a series of other flaws in the strategy and its implementation.
Getting started
Many parts of the country were already active on the doorstep when May delivered her speech, given the approaching local elections. For some, that proved a positive – much of the work of gathering their campaign teams, planning out delivery and canvassing routes and so on – was already done. But there were downsides, too. After two years of what felt like relentless campaigning, from the 2015 election through the EU referendum, the news of yet another national ballot was met with widespread weariness.
The question was how to start the election campaign. The electorate all knew that it was underway, and were therefore expecting to hear from parties and candidates about what they were offering. But, given that almost no-one had any advance warning at all, there were precious few ways to start communicating.
Nobody had candidates in place, for target seats or otherwise (indeed, some of those who had been provisionally working some seats as the association’s intended candidate ended up not being selected after all). Nor did most seats yet know what their place would be in the campaign – were they targets, or were they going to be expected to travel to fight in battlegrounds elsewhere, as many had under the Mutual Aid scheme in 2015?
The message was therefore limited to what was coming out centrally. CCHQ immediately sought to commission huge mailshots to its newly-identified target voters, kicking off the campaign with a splash.
Unfortunately, this also fell victim to the genuine surprise of the snap election. The relatively few companies capable of printing such large quantities at such short notice would be more than happy to take the job on, they said, except for one thing. There wasn’t sufficient paper stockpiled in the country to print what the Conservative campaign bosses were asking for – more would have to be shipped in. This was a reply that nobody had foreseen. They ordered materials for when they could, pushed back many of the mailshots, and and relied for the rest of their initial reach on their digital team. “Digital was the only source of contact for a lot of voters in the first couple of weeks to explain what the election was about, but the data left a lot to be desired,” one planner told me, which further contributed to the early concerns about the given reasons for holding an election in the first place, which we reported yesterday.
By the time the local elections came round, two weeks on, there were encouraging signs that they had got a handle on the direct communications. On the morning of the vote, Thursday 4th May, local newspapers in several areas were emblazoned with wrap-around adverts promoting “Theresa May for Britain” and “strong and stable leadership in the national interest”, exhibiting all the core messages that were to become notorious. The adverts were a canny move – selectively placed in a way that would assist in key battleground seats, but counting as national spending, and boosting local election campaigns, to the satisfaction of activists who were already feeling a little weary of pounding the pavements.
The adverts also offered an early insight into the potential list of target seats which the Conservatives – or, rather, “Theresa May’s team” – were hoping to gain. Buzzfeed’s tally of the adverts suggested that Mansfield, Hove, Chorley, Stockport, Exeter, City of Chester, Barrow and Furness, North East Derbyshire, Hyndburn, Bury South and Scunthorpe were in their sights. That suggested some serious ambitions to expand into areas rarely if ever before won by the Conservatives – and was based, of course, on those swift analyses by Messina, Textor et al of the combined polling and consumer data back in CCHQ. But where else were they looking?
Identifying the target seats
The decision on where to target had to be made swiftly. The initial list was drawn up somewhere between the Tuesday on which the Prime Minister announced the election, and the end of that same week. By the Wednesday, ConservativeHome was able to identify a rough-and-ready way to discern which seats were targets and which were not: the special candidate selection rules explicitly set out different processes for targets and non-targets, allowing us to divine which broad category each constituency fell into. By the Saturday, we had a confirmation of the approach from an email sent to candidates, and were able to reveal that lists of proposed targets circulating among regional officers that May’s ambitions stretched to targeting some seats with Labour majorities of over 8,000 votes.
That spoke of the sheer strength of their poll lead and their confidence about a massive victory. Not that CCHQ was alone in that expectation – I’m told that Labour’s internal polling at this stage of the campaign was worse for them than even the public polls. That the Conservatives were hoping to overturn Labour majorities of over 8,000 suggested the target list could stretch to 80 or perhaps more Opposition-held seats – an estimate that was borne out later by studying the different selection processes.
“The targeting was very ambitious, looking at…Bolsover [Dennis Skinner’s seat],” concedes one member of May’s team. Those on the ground in some of the chosen constituencies were flabbergasted to learn that they were no longer viewed as “hopeless” but were now getting target seat treatment. For example, an experienced local campaigner described to me his feelings as follows:
“Why the heck was [redacted] targeted? A Remain constituency, 10,000 Labour majority, no conservative councillors. No evidence of a revival there…we were being cocky.”
Around the country, as the results came in, numerous experienced campaigners in Tory seats with large majorities realised to their horror that while they had been travelling often long distances to give mutual aid to supposed target seats where Labour won convincingly, Tory-held seats far closer to them had been lost. In one instance, a well-resourced association saw the Labour majority in their allotted target seat increase, while a Conservative seat which they drove through regularly to get to the target was lost.
This mismatch got worse as time went on, too. Positive early canvassing returns (pre-manifesto) and the encouraging local election results led CCHQ’s strategists to start not only treating Tory-held marginals as safe, but to divert resources away from the more marginal Labour-held target seats and towards target seats further down the list, ie those with bigger majorities. A candidate in what was supposed to be a top target – Leave-voting, and narrowly held by Labour in 2015 – tells me that:
“CCHQ’s eyes were definitely bigger than their belly. The limited resource we had in [our area] was originally due to be directed to us. That would have already been a stretch, but once canvass returns came in we lost it all together as 5-8,000 majorities were targeted instead. It left us very exposed.”
Few stories can underscore quite how wrong Tory strategists got it than this account of money and people being switched out of a top target in favour of more ambitious gains. In the end, Labour held onto that seat, as well as the seats with larger majorities to which campaigning efforts had been shifted. A senior MP suggests this overconfidence allowed Conservative-held seats to be lost, when compared with the previous 40/40 approach of targeting seats to defend as well as seats to gain: “We were spread too thin. A 40/40 strategy might have picked up the problems.”
Selecting the candidates
Having identified their targets, they needed candidates to fight them. Regular readers of ConservativeHome will recall our coverage of the candidate selection process at the time. My full account from 9th May provides the detail of the selections saga, but it’s worth revisiting the experience, as it set the tone for the rest of the campaign. Covering it was frenetic work for us, because the process we were covering was equally frenetic
Ordinarily, there might be a couple of selections a week over the course of two years. The aim is pretty obvious: to ensure that there are candidates in place at minimum by the start of a campaign, and ideally long before in target seats in order to build up name recognition. But this time no candidates had been selected in advance. From the first email going out to the candidates’ list on 20th April to the deadline for submitting nominations on 11th May, there were now three weeks in which to select over 300 candidates.
Some Conservative associations had been pressing to select early in the preceding months, with a view to establishing their candidate for a good run-up to the expected 2020 election, but CCHQ had refused to allow them to do so. Instead, CCHQ was biding its time, having drastically culled the candidates’ list after 2015 – some suggest in order to sweat the maximum amount of work out of would-be candidates by dangling the hope of being approved for the list later on. So not only were there no candidates in place, but the approved list didn’t contain sufficient names to run a full slate.
The immediate problem of needing to select hundreds of candidates in a very tight window was answered with special rules to override the usual, more drawn-out, process. These divided seats into two groups, with a distinct process for each. The first, Conservative-held seats where the sitting MP was standing down and Opposition-held target seats, would be presented with a shortlist of three candidates, decided by CCHQ, from which the Association had to select. The second, non-target seats, would simply have a candidate imposed on them.
In effect this was a huge centralisation of power over selections – one of the most valued powers which normally sits with Associations and members, but which had already been eroded somewhat in the Cameron era. The special rules supposedly provided for consultation with senior officers of the local Association when shortlisting or imposing a candidate, but experience proved this to be lip-service at best. When Associations made clear requests for someone to be shortlisted – most notably when Aldershot asked for Daniel Hannan – their request was “noted”, but then ignored. Those who threatened to reject all three of CCHQ’s suggestions were first pressured subtly, being told they would have to do so with the candidates present, on the night, and then bluntly, being told that if they did then CCHQ would impose one of them anyway.
Having established these rules, CCHQ set itself a rolling timetable of selections. With a candidate shortage, they planned to recycle some of those defeated in early selections back onto the shortlists for less plum seats later on. As the hectic fortnight went by, I started to notice that shortlists that had been confirmed to me days earlier were being changed at the last minute as they struggled to manage the ever tighter process of arrangement and rearrangement. It was the wedding seating plan from hell: who was available and good? Who was available but not good, and were the other two on that shortlist strong enough to ensure they could safely be included without risk of them winning? Who was coming through the newly-instituted pipeline for fast-tracking new candidates onto the list?
As I wrote at the time:
‘some associations were receiving their shortlists only hours before the selection meeting was due to take place. For the lucky shortlistees, this sometimes involved a mad dash out of work, with a hurried apology, across long distances to speak in the hope of selection, in some cases addressing activists they had never met in a seat they did not know. On 2nd May, Andrew Lewer MEP had just arrived in Brussels when he was informed he was up for selection in Northampton South that night – he immediately jumped onto a train back to London, and went on to win the selection.’
In Scotland and Wales, the processes varied. Ruth Davidson’s Scottish Conservatives were allowed to set their own rules, and opted straight away for direct imposition in almost all seats – something that seems to have gone mostly unchallenged, given the threat of the SNP. The Welsh Conservatives were supposedly abiding by the same rules as England, but deviated from them repeatedly. In various targets, shortlists of fewer than the official three were put forward, and in others a single candidate was imposed despite the rules. I’m told that senior figures in the Welsh Conservative Party justified this on the grounds that “there has to be at least one female MP returned in Wales this election”.
The overall effect of this dash to select was bruising, for candidates and associations. Those would-be MPs who had been left hanging on the phone without even an answer, never mind a shot at selection, felt “sick and betrayed”, angry at “a mockery of meritocracy””, and “doubting the very foundations of the organisation we have supported for decades”. Members understood the need for haste but nonetheless felt frustrated by effectively being stripped of what they saw as their most valuable democratic right. That harm undoubtedly dented local campaigning efforts in some places, and has had a lasting effect – many members remain disillusioned, some experienced candidates I have spoken to have withdrawn entirely from politics, and even some of those still involved vocally protested their mistreatment by heckling the co-chair of the candidates committee, Amanda Sater, recently.
Getting the message out
Bruises aside, the ruthless selection process did fulfil its primary aim: they had a candidate in place, everywhere, in time. As each association chose – or grudgingly accepted – its candidate, the work of getting the message out on the doorstep began.
The campaign was certainly well-resourced – financed, we now know, by record levels of fundraising. Each target seat was guaranteed at least one mailshot, paid for centrally, and almost all received a local campaign manager. Particularly for those associations which were accustomed to being neglected, this was a welcome surprise.
There were strings attached. A number of candidates and associations had people in mind for the roles – people with local knowledge, or whom they had worked with before, whom they valued for their campaigning experience. Only rarely, if ever, were these suggestions accepted – the central Party was paying for the positions, so it insisted on deciding who filled them.
Some associations got people they were happy with, others did not. Many felt frustrated to be starting again from scratch, after putting in the hard work of training up and building relationships with the 120 campaign managers employed before the 2015 election only to see most of them let go shortly after the majority was won – “a bloody catastrophe”, as one official put it.
With candidates and staff in place, the next task was getting the literature together. Here, too, the decision was taken to exercise an unusual level of central control. In the past, the central Party would handle national mailshots, and local teams would write and design leaflets before submitting them to CCHQ for fact-checking and approval. Sometimes there were disputes over policy or particular messages of which the centre disapproved, but it broadly worked.
For the snap election, though, the system was reversed. Blueprint, the online ordering system for Tory leaflets, offered a limited range of templates based on the type of seat – Tory-held, target, non-target – and the bulk of the content for each leaflet was pre-written centrally on the national message. All the target seat candidates were rushed to London to take part in a marathon photoshoot to ensure each had a picture with the Prime Minister, but there was normally only a small space for local messaging or even information about who the Conservative candidate “standing with Theresa May” was. “We were only allowed about 20 words per leaflet about local issues, rest taken up with Theresa May, Brexit, and Corbyn,” complains an officer in a target seat who used the system.
The dangers ought to have been obvious. Plenty of activists harboured doubts about the centre’s messaging – one tells me of a voter shouting “get that communist trash out of here” then slamming the door, having assumed a leaflet with Corbyn’s face on the front came from Labour, while others found Leave voters annoyed to see the risk of a Corbyn government equated to Brexit.
The same centralised approach applied online. “CCHQ took admin rights to our Facebook pages, but everything they posted was “we’re better than Jeremy Corbyn/Jeremy Corbyn will lead to chaos”, as one candidate puts it.
Even if the designs and messages were right nationally, they still might not be well-suited to local circumstances. Activists keen to deliver in Remain-voting constituencies in London feared the Brexit-focused leaflets they were given would do more harm than good, while in Wirral West “…leaflets were delivered with quotes from The Sun prominent. We’re just over the water from Liverpool! Many LFC supporters live on the Wirral. Some of the Hillsborough dead have families here. [They] may as well have sent leaflets round with ‘We couldn’t give a toss about dead Scousers’ on them.”
In some cases CCHQ sent out mailshots without even notifying, still less consulting, the local operation. In Wales, an association officer was surprised to receive an email notifying them of the cost of a recent mailing for their local campaign spending return. On asking to see the leaflet that had been sent out, in order to check what it said should voters mention it on the doorstep, the central Party was only able to provide them with a lorem ipsum template – the half of the leaflet that contained political messaging read only “GENERIC COPY”, and no-one at the centre could find a copy of the text that had been supplied to the printers. The same thing happened elsewhere: “CCHQ sent national leaflets…to targeted people but we weren’t told, which made things very awkward when we knocked on doors and we’re asked about a leaflet that had been received that we’d not seen and didn’t know anything about,” recalls a defeated candidate in a target seat.
Immediately that they were selected, candidates were able to contribute to the text of an election address, which would be sent out en masse at the expense of the Party. It was a welcome piece of financial assistance which is not normally available. But as the weeks went by, there was no sign of it actually arriving through voters’ letterboxes.
Fighting a snap campaign, with a strategy of micro-targeting a relatively small proportion of the electorate, and mostly without prior name-recognition due to the late selection process, they had hoped such a mass-mailing would help to establish their presence, and free them up to target the minority whom the modelling said were the likely swing voters. But instead, they started receiving angry messages from voters asking why they had been “ignored” by the Conservative campaign. Few candidates who had time to enquire about them got an answer, but behind the scenes CCHQ had decided to delay their delivery.
In addition to leaving voters feeling ignored, this had two damaging effects. First, some of the mailings to postal voters arrived late, after postal votes had been sent out, thus missing a crucial window to influence a sizeable minority of the electorate.
Second, by the time many of them were delivered, in the final week of the campaign, “the content was very out of date – all ‘strong and stable’ and so on”, conceded a senior adviser. This vast numbers of leaflets had been written, designed and printed at the outset of the campaign, but the decision to delay sending them meant they were actively counter-productive by the time they were posted. Some on the ground had assumed that they had been delayed in order to update the messaging, but were horrified to find that when they did go out they were exactly as they had been drafted in week one. They had no answer to questions about the manifesto, and they emphasised the Party leader at a time when even the national operation had started, belatedly, to try to pivot away from a damaged brand.
Deploying the troops to the right places
The most obvious campaigning problem for the Conservative Party has long been its shortage of activists. While membership numbers had recovered slightly in the preceding couple of years, the Corbyn-led surge of new members into the Labour Party had accentuated the imbalance in terms of boots on the ground.
Cameron’s majority was in part won by a careful and deliberate strategy of ensuring that Tory activists were cajoled, incentivised, and bussed into target seats, making the most of overall weakness by turning it into very concentrated strength. This operation was Team2015, which had begun by simply passing on people’s details to local associations. That had proved an inefficient approach – people might turn up once, but few ongoing connections were being forged. So they changed tack, putting £300,000 into a series of techniques – gamification by which activists could climb a scoreboard by taking part, seemingly endless chivvying by phonecall, text message and email – to actively encourage people to travel to target seats.
It worked, delivering tens of thousands of days of campaigning work in the places where they were most needed. But there was no sign of the organisation by the start of 2017. As I reported in February, the campaign in the Copeland by-election displayed concerning signs of poor co-ordination of volunteers, which often repeated teething difficulties that Team2015 had also faced but learned how to overcome two years before. What had become of it?
Much of the answer lies in the fateful decision to co-opt, alongside Team2015, RoadTrip 2015: an initiative which organised bus trips of young members to key seats, run by a controversial former candidate, Mark Clarke.
The bullying scandal that engulfed the RoadTrip after the tragic death of Elliott Johnson in September 2015 has been the subject of several investigations. Publicly, the Conservative Party responded by introducing new procedures to protect young people and to make it easier to report instances of bullying. Privately, the impact on the internal attitudes of the Conservative Party was to tar not only Clarke’s organisation but also the wider Team2015 effort with an air of scandal.
That was unfair to Team2015 – the two operations had similar names but were run by different people, in very different ways, one with a formal CCHQ machine and the other with an operation effectively outsourced to Clarke, albeit with CCHQ funding – but it happened nonetheless. In June 2015 I published a CCHQ organogram that showed “Team2020” at the heart of the Party’s operations. To my knowledge, the organisation held one campaign day in early September of that year, shortly before the RoadTrip scandal broke. Then it vanished without a trace.
The effects of the controversy were felt beyond the closing down of Team2020. The entire youth wing of the Conservative Party had been scrapped, pending a review and relaunch (a process which is still underway). Benefiting from more spare time, and more energy, than older activists, Conservative Future had made a disproportionately large contribution to the bread and butter work of door-knocking and leaflet delivery. But, despite calls from this site a year ago for it to be relaunched, the Conservative Party went into the General Election without a youth organisation. CCHQ did have a Youth Outreach officer, Calum Neilson, of whom volunteers and candidates speak positively. But he had been “hamstrung”, I’m told, by a Party “freaked out” about youth campaigning, and was effectively forbidden from social media engagement during the campaign – leaving the Tory operation even further behind in online engagement.
To make matters worse, another controversy then contaminated the wider concept of centrally directing activists to target seats. Concerns about the costs – particularly the risk that they might need to be allocated as local, not national, spending – had led the one-constituency RoadTrips to be discontinued by the time the 2015 campaign formally began. CCHQ’s advice on electoral law led them to believe they had found a way to avoid that problem: regional battle buses, that toured multiple seats, would, they believed, constitute national spending, avoiding the danger that a bus could wipe out a large chunk of a candidate’s campaign budget in a single day.
As the Electoral Commission and CCHQ have now discovered, that belief was wrong. The investigation that ensued into allegations that MPs had failed to properly declare their local spending as a result of the tactic proved damaging. Individuals and the party as a whole were derided as committing electoral fraud (mostly wrongly, it turned out) and MPs found themselves subjected to police interviews and negative media coverage. CCHQ was reluctant, and therefore shamefully slow, to put its hands up and clearly admit that candidates were innocent of any intentional offence precisely because it had issued them with incorrect advice.
The harm was lasting. The tactic of bussing activists in to target seats was ruled out, for obvious reasons, but MPs and local officers were also left furious with and distrusting of the national Party. The sight of their colleagues being hung out to dry, through no fault of their own, mean that the affair harmed the relationship between the centre and the MPs well beyond those directly affected. Where central help was offered to those defending marginal seats, it was often turned down by MPs who no longer trusted the centre not to plunge them into a career-ending controversy.
That left the Tory ground operation in 2017 reliant on the old system that Team2015 had replaced. Activists were emailed intermittently by the centre to urge them to attend a target seat, but there was no “chivvying team”, no equivalent of the focused operation to press shock troops into the key seats. Instead, they fell back on the local associations, asking the chairmen of non-target seats to gather their activists and bring them to where they were needed. Some did, but this was a system that CCHQ had known by 2014 was not sufficient to the task.
When people did turn up to help, on their own initiative, local operations inevitably didn’t have the necessary advance warning to make the most of their help. In the face of an unexpectedly large Labour grassroots presence, numbering in some places hundreds of volunteers, incumbents found themselves fielding teams of a couple of dozen.
When I wrote in 2015 that “Team2015 was a success in its own terms…[but] it is essential not to have to start from scratch again in 2017”, I was concerned that the Conservative Party would rest on its laurels while its opponents learned how to copy and develop what worked. In the event, something far worse happened – the grassroots operation had gone backwards.
Doubts grow about the targeting strategy
To begin with, activists on the ground were cock-a-hoop. Like CCHQ, they were buoyed up by the national polling and the local election results – and their initial findings on the doorstep seemed to bolster the impression that they were on track for a famous victory. “Before the manifesto launched, I’ve never seen canvass returns like it,” said one veteran councillor in a Tory-held marginal.
Still, some harboured doubts, and worried that the findings of the centre’s data modelling simply didn’t match up with what they knew to be true in their local area.
One MP who was defending a small majority found himself conflicted. “For the first few days, I thought maybe it’s just me…the Whips asked me ‘Where are you going to work?’ and I said I’d be working my own patch. For the first two weeks, we followed their data – going to council estates first that we’d normally do last. It was only ok, not amazing. People were pleasant, but they were undecided. I reported back after two weeks that it looked the same [as 2015] not a new, big switch. They told me I was wrong.”
By that time, the pressure was on from the centre for him and his local activists to leave their constituency and help in an ambitiously targeted Labour-held seat nearby. The national targeting strategy was being rigorously enforced by Lizzie Lumley – who had been promoted from London Campaign Director, much to the annoyance of those volunteers in the capital who were still smarting from the failure of Zac Goldsmith’s mayoral campaign.
“I didn’t hear candidates say that this data was wrong,” protests one of those in the upper echelons of CCHQ. But that’s unsurprising, given the internal culture. Candidates, reliant on CCHQ patronage for their futures and understandably fearful of getting in its bad books, mostly followed orders and kept their concerns to themselves – “[CCHQ] thinks if you call them, you’re a panicking candidate”. Some MPs, more secure in their position, disobeyed and were eventually left to their own devices – though that didn’t stop CCHQ trying to persuade their local activists to abandon them and work elsewhere.
The MP quoted above started ignoring the centre entirely. He drafted and produced his own literature, disregarding the Blueprint formats: “I thought I’d put her on my literature…but after a week or so I didn’t like being linked to ‘Corbyn’s a lunatic’.” And he started ignoring the TVT target lists CCHQ was sending down to him. Eventually, the national party stopped badgering him, but it still wanted his activists. The Saturday before polling day, engaged in what he believed was a fight for his seat, he found several of his most experienced volunteers had been persuaded to go to the nearby target instead. In the event, he narrowly held on – and the Labour MP in the target seat increased their majority.
The target seats had been chosen by selecting those with the highest number of potential Tory voters, identified by the TVT analysis. But those chasing target voters started to become frustrated with the data, the system and the effectiveness of its predictions.
In several seats, activists were bemused that they were getting a hostile response from people whom the system told them would have a high chance of voting Conservative. Comparing the target lists with their local canvassing records, they found that the analysis rested on very old information. “I was knocking up people with data that was, in some places, 15 years old,” in the words of one. Another found herself sent to the doors of people whom the local party had last spoken to in 2005, when they were down as strong Labour and Liberal Democrat supporters. In Twickenham, which was lost to the Liberal Democrats, they found people whom the local association knew to be Lib Dem activists were identified by TVT as potential Conservative voters. CCHQ’s shortage of data had driven them to scrape the barrel so deeply that the information it was urging people to trust was simply unreliable.
This was a highly selective targeting system – including, in some places, picking out individual voters within households – based on Messina’s much-touted “1,000 pieces of data on every voter”. If it was right, then it would have been very effective. But when its underlying assumptions were wrong, it became an extremely costly waste of time. A good leafleter or canvasser can get through a street fairly swiftly. But this approach didn’t involve doing whole streets, it meant picking out a few people on each street who were supposedly sympathetic, which is very time-consuming even if the targeting is correct. “You would be sent off to knock a dozen streets, going to maybe six houses on each, which meant walking miles to barely speak to a voter,” said one canvasser.
When they did get the chance to talk, the canvassing scripts were less than ideal. They hammered home the messages – “Theresa May’s candidate”, “strong and stable”, “coalition of chaos” – to a degree that made those trying to use them cringe, and spurred negative responses from voters who felt they were talking to a robot. The scripts insisted on rating Voting Intention on a scale of one-to-ten, something useful to a data modeller but alien to how voters actually think of their intentions. “The canvass script doesn’t work. I haven’t met any sentient being that used it,” an experienced MP told me afterwards. Many activists simply ditched it entirely.
Unlike in 2015, when VoteSource, the bespoke campaign software, crashed at crucial times, the computers generally worked quite well. The problems on that front were two-fold: first, the data going into and out of it evidently was not sufficiently reliable. But second, the full suite of functions were withheld from MPs who were judged to be in non-target seats, apparently in order to push them to go and help in the targets. “The VoteSource whizzbangs were refused to most incumbents,” an adviser told me, “Some MPs pleaded until they got it.” Ultimately, though, any software can only be as good as the data put into it.
The campaign also experimented with some new tactics. One was the “Plan to Vote”, by which people were asked to specify the time of day they would be voting, and which family members they would go to vote with – an example of which is still live on the Conservative Party’s website. Apparently imported from the US, this baffled activists and voters alike. “Why do you want to know that?” was a question routinely heard on the doorstep and over the phone. The purpose of the question was never really explained to the people told to ask it, and the data gathered did not appear to be used in the operation on polling day.
Worse, in some target seats it was instructed that it should be the primary focus for canvassers in the last few days of the campaign. One experienced volunteer found herself leading a team of canvassers to TVT-identified households over the weekend before the election, equipped with Plan To Vote scripts, only to find that every single one of the people they spoke to had received a direct mail letter from the national campaign asking them the same thing a few days earlier. “We were speaking to people who’d been sent it already. Even if it was the smartest thing ever, and the greatest f*cking trick, it was a waste of campaigners.”
Election day dawns
A normal ground campaign follows a rough progression. You gather data and test messaging by canvassing for months or years in advance. You identify your voters through more canvassing during the campaign itself. And then, on the day, you go back to those supporters whom you have identified earlier, in order to ensure that they cast their ballot. The last stage is Get Out The Vote (GOTV).
The first stage was curtailed, if it had ever properly begun. The second stage was troubled by poor targeting, and messaging that became less effective over time. The third stage suffered its own problems.
Despite the inaccuracies which had by then become obvious, the decision was taken not to confine the GOTV knocking up to Conservative pledges – those who had confirmed their intention to vote Tory. Instead, troops in a variety of target seats were sent to knock on doors based on the inaccurate TVT data. A flawed hope still seemed to be lingering that the model might turn up voters who hadn’t been contacted yet.
“On the day, we were not using canvass data, but going to doors extrapolated from the demographics,” a long-serving campaigner tells me. Another found that “inexperienced centrally appointed campaign managers who don’t get elections, activists or campaigning…[were] telling activists not to call on pledges.” This appears to have been a central instruction, enforced by the Party’s local employees. The assumption that Conservative core voters would definitely turn out, which as I reported yesterday meant that CCHQ was slow to detect the damage done by the manifesto, persisted into the GOTV operation.
As the day went on, those on the ground detected that the national operation was faltering. One Association officer recalls the experience as a process of partial and conflicting instructions: “We were given bizarre instructions about removing tellers half way through the day…[which] made the whole process of update/knock up ludicrously time-consuming as we had no idea who had already voted. Moreover, CCHQ were supposed to update the app throughout the day to help us target specific areas with the GOTV operations. This didn’t happen…We had large teams being sent without prior warning from other constituencies. We didn’t need them and had no work to give them. They ended up sitting around.”
Again, some local campaigns ended up just ignoring the central strategy. In Richmond Park, “we were given data from CCHQ that expected us to knock up voters who had been recently canvassed as against or simply had no [Voting Intention] at all. In the end we used our own judgement and knocked up only Cs [Conservatives] or Ps [Probables] and won by 45.” Some spotted that they were in trouble early – Oxford West and Abingdon, where Nicola Blackwood lost her seat, was putting out calls for help to nearby associations – but elsewhere, trusting in the model, MPs were off providing mutual aid, unaware that their own seats were in danger.
By 10pm, of course, it was all over. The results began to come in, and the true scale of the disaster swiftly became obvious. Supposed targets were missed by a mile, MPs who had been assured they were safe found themselves in narrow races to hang on, and Conservative seats were lost, not gained, overall.
Then came the final insults.
Overnight, the CCHQ email system delivered thousands of emails to members and activists around the country, urging them to come and help to Get Out The Vote in battleground constituencies. “We predict a lot of seats are going to be very tight,” one declared, “…every vote really does count.” Sent during polling day, a failure to properly check that their system was delivering mass emails swiftly and promptly meant that these messages were landing at three, four and five in the morning of Friday 9th June. Those still awake, watching the results come in, felt their horror turning to anger. Those who opened their inboxes at work the following morning felt that it was final reminder from CCHQ of a chaotic and mishandled campaign.
Fuming, they waited for some kind of explanation. Instead, they got silence. Many didn’t even get a message of thanks from the Prime Minister or the Party Chairman for days, if at all.
The aftermath was little better for candidates. A blanket email went out from CCHQ saying “thank you for supporting candidates in target seats” – and was sent, inevitably, to those who had themselves been standing in target seats. “They can’t even take the time to send a personalised and accurate e-mail…those people drive me crazy,” one told me.
The hangover
Plunged into a snap election with even less notice than those at headquarters, tens of thousands of Conservative activists and hundreds of candidates battled for almost two months to make the gamble pay off – fighting the errors of their own leadership as well as the efforts of the opposition. The main problems I identify and describe in this analysis can be summarised as follows:
In the event, a proportion of the Conservative ground campaign simply disobeyed instructions that they did not have faith in. Looking at the results, the uncomfortable thought occurs that had they not done so, the result might have been even worse.
We are all now living through the consequences of that campaign. But for the Conservative Party, the effects could last longer than just this Parliament.
The targeting errors in particular threaten to have a lasting and harmful impact on future elections. Successful Conservative campaigns rest on persuading associations in either safe or hopeless seats to lend their support to target seats elsewhere – mutual aid, as it’s known. Doing so rests entirely on trust in the centre to correctly identify where troops are needed and where they are not. Those who spent their time in targets, only to see Labour win them by convincing or even increased margins, are disheartened. Many feel guilty that Tory seats which they could have helped to defend were lost, while they had followed instructions and gone to help elsewhere. Their trust in the wisdom of the centre to send them to the right places has been shattered.
“One of the impacts of this election will be the end of mutual aid,” fears an experienced candidate. Those MPs who survived the experience and held their seats agree: “We’re all going to spend more time in our own seats in future.”
That is a danger which the Conservative Party’s leadership must overcome if it hopes to rebuild and advance at the next election. The grassroots are not numerous enough, or equally spread enough, to rely on everyone just working their own patches. The central operation will have to openly admit what it got wrong, and demonstrate that it can be trusted to get targeting decisions right in future.
That process will be made more difficult by the fact that the broader relationship with the volunteers on whom the Party relies also suffered a great deal. Centralising control over selections was the first in a series of blows which left many feeling deeply alienated. One MP in a marginal seat stresses that democratic control over selections was one of the few surviving perks of Party membership: “We ask them to join, we take their money, and we give them bugger all for that except emails begging for donations.”
“The whole candidate process was dreadful,” admits a senior Parliamentarian, and draws an unflattering comparison with the enthusiasm of the Corbynite grassroots: “It’s hard to avoid the conclusion that creating an actual mass movement in which people have some control has a potential benefit.” Ruefully, he adds that the modern Conservative Party is “almost a structure set up to discourage participation. It’s not something you can love anymore – it’s something you fear because it’s a totalitarian state.”
Anger among the grassroots over the whole experience seems unlikely to fade. “We are tired of working our arses off for a party that hates its members,” one writes, a blunt summary of low morale that I have heard echoed again and again.
The question now is how the relationship between the party and its members, and the machine which is founded on that relationship, can be rebuilt. Even if the undeniable issues with policies and personalities have been corrected by the next election, the Party will still need its volunteers and its campaign machine.
The first step in setting things right is to accept that there is a problem in the first place. Indeed, that is the primary reason for the publication of this series. The Conservative Party does not have a strong internal culture of accepting and acting on negative feedback. Even one of those who held a senior role in the election campaign told me that “CCHQ reacts remarkably badly to criticism. Lots of folks have been there a very long time.” One MP cites an unwillingness to let go and localise: “They’re not willing to trust people.” To begin to rebuild, that will have to change.
> Our series continues tomorrow, with a series of recommendations to reform the campaign machine.