In May, immediately after reformers lost the AV referendum, I called on its main funders to hold an inquiry into what happened, drawing on report by one of the campaign's key activists Andy May. Now James Graham who ran its social media campaign has published an account in the Lib Dem magazine Liberator. It is cross-posted here with thanks. While the debate extends well beyond the Liberal Democrats, what lessons they draw is especially important. I understand on good authority that many if not most Lib Dem MPs and Peers (I believe they have more Lords than MPs in Westminster) and especially those in the leadership now holding government office, think that constitutional reform is an unnecessary distraction that should be sidelined. That's one view, here is another. (Anthony Barnett)

If you want to understand why the Yes to Fairer Votes campaign failed so badly, you won't get much sense out of the Nick Robinsons or even Vernon Bogdanors of this world. You need Dian Fossey. I’m quite sure the famed zoologist would have been able to explain it all.

The list of the Yes campaign’s mistakes seems to grow with every account. How could it make so many fundamental errors? The simplistic analysis was that the people at the top of the campaign were stupid or incompetent. I don’t believe this was the case, but what they certainly were guilty of was developing a management culture in which group think and the laws of the jungle were allowed to thrive and take over.

The saddest thing for me personally is that it started so well. The small group of democratic reform organisations correctly calculated that electoral reform would rapidly, albeit temporarily, rocket up the political agenda immediately after the 2010 general election, when the gross disparity between how people voted and what they got lumbered with in the House of Commons briefly entered the public consciousness.

In an attempt to capture the zeitgeist, we spent the last couple of weeks in April 2010 working together to establish Take Back Parliament. The effect of this was not just a demonstration in London that captured the media’s attention during a crucial phase in the coalition negotiations, but the spontaneous formation of dozens of local groups across the country.

Once the coalition had been formed and it became clear that an AV referendum was going to happen, the organisations again came together to start planning the Yes campaign. At first, it appeared as if we were doing all the right things: learning from the trials and tribulations of past referendum campaigns, commissioning extensive polling, and building a team with a specific focus on avoiding it being dominated by Liberal Democrats. I certainly spent the summer of 2010 feeling that, although the scale of what we needed to achieve was immense, we were at least learning from past mistakes and were determined to adopt an evidence-based, non-dogmatic, approach to campaigning.

Rigidly hierarchical

But towards the end of August, something fundamentally changed. The campaign suddenly, and at first imperceptibly, became rigidly hierarchical and obsessed with secrecy. I found myself in the odd position of being nominally in charge of the website while being excluded from talks with the contractors who were being charged with building the thing.

As the weeks went by, it became clear that the small team of senior managers was being made even smaller. The planned ‘research and rebuttal unit’ was merged into a communications unit headed by former spin doctor for Gordon Brown, Paul Sinclair. Far from a mere press office, ‘comms’ was to have control over every aspect of every statement and leaflet put out by the campaign. Yet bizarrely, this super department was to have only four members of staff for all but the last month of the campaign.

Predictably, the effect of putting so few people in charge of so much was a massive bottleneck. Slightly less predictable, but no less lamentable, was the fact that research in any meaningful sense ceased. After the initial qualitative and quantitative analysis conducted over the summer, and a huge poll in November designed to help us identify what messages appealed to each demographic, opinion poll research effectively stopped and from that point onwards we were reliant on people’s hunches to muddle us through. A frustrated research team found itself with nothing to do and was not empowered to work on its own initiative. Opposition research and proactive fact-checking simply ground to a halt.

Research was not merely not commissioned; it was ignored. Our initial focus group work clearly showed that people were contemptuous of the idea that electoral reform would prevent corruption; people only approved of notions such as AV “making MPs work harder” in the context of them having to reach out beyond their core party support during elections. Despite this advice, the campaign repeatedly sought to conflate the two. Similarly, the advice we got from veterans of the 2004 North East referendum was that celebrities were of limited value. Despite this, we ran a campaign that was obsessed not merely with celebrities but with ones who appealed only to the educated middle classes.

The campaign became increasingly reductionist in its approach. In recognition of the very real problem we faced in explaining AV to a broadly disinterested public, we adopted the guiding principle of “show don’t tell” over the summer. By mid-November, that became “don’t tell”. All proposals for explanatory videos or websites were blocked (indeed, it took a month before the comms unit was willing to sign-off any explanatory pages on the website at all).

The ground operations team was, despite strenuous objections, given explicit instructions to discourage local groups from holding mock ballots. The fear was that the people who participated in such ballots would be so outraged when they read in the following week’s local newspaper that their chosen Strictly Come Dancing contestant had not won the mock ballot, that they would instantly resolve to vote No. (At the end of March, the IPPR published research showing that support for AV massively increased amongst people who had been given the opportunity to try it out, but by then it was far too late).

Possibly the most reductionist policy of all was the decision to place so much emphasis on phonebanking. Again, it was based on the perfectly sound notion that person-to-person persuasion was far more effective than showering people with leaflets, and that we would cover far more ground on the phone than we could getting people to go door-to-door. Somehow, however, that reasonable guiding principle led the campaign to adopting an approach in which the entire ground operations campaign would be focused on getting as many people as possible to participate in one of the fifty phonebanks we were to set up across the country. At the early stages, the talk was of the largest phone operation ever seen in the UK, with 3 to 5 million contacts all but assured.

It soon became apparent, however, that not only did this strategy fail to take account of the fact that most activists did not actually enjoy phonebanking, but that considerations such as software procurement and even the legal situation had not been taken into account. In the end, just 500,000 contacts were made; and most of the data generated was not actually useable for the purposes of getting out the vote.

Notoriously Dysfunctional

So how was all of this allowed to happen? In my view, to understand that, you have to understand where most of the senior staff were coming from. Gordon Brown’s Downing Street was notoriously dysfunctional; the Joseph Rowntree Reform Trust’s POWER2010 campaign had been an expensive and chaotic disaster. And then there is the nightmare that is the Electoral Reform Society.

The trials and tribulations of ERS have been chronicled over the years in Liberator. Suffice to say that its recent history has not been a happy one. The organisation tore itself apart over the Jenkins Report in 1998 and an unhealthy ‘them versus us’ culture has existed between staff and its governing council ever since. Yet despite its problems, thanks to its commercial arm, in recent years it has been extremely wealthy (at least in voluntary sector terms).

From talking to them, and overhearing them in the open plan office, it was clear that too many of the senior staff had an outlook that was deeply cynical about political activists and campaigning in general. Idealism was very thin on the ground. In retrospect, it is extremely easy to see how such a group of people with a very similar perspective and with scars across their backs from past struggles found themselves reinforcing each other’s preconceptions rather than challenging them. And it is very easy to see how they might end up imposing a sink-or-swim, cliquey style of management.

The walls of the room in which the communications unit and most senior staff were based were covered in leafy green wallpaper; as a result, it quickly acquired the nickname ‘The Jungle Room’. Looking back on it, it is quite striking how reminiscent it was of Gorillas in the Mist. You had the silverbacks in one corner of the room, masters of all they surveyed. They, in turn, were surrounded by their trusted deputies, grooming away. Roughly speaking, the further away you were from the top table, the further down in the pecking order you were. At one point, the room was even rearranged so that there was a whole island of desks between the top table and the rest of the people in the room. The point being made could not have been more emphatic.

This is the only office I’ve ever worked in where the female staff felt it necessary to hold regular ‘ladies lunches’ in the interests of mutual support. The initial attempt to get the campaign to entrench the principles of “respect, empower, include” into the way it treated staff and volunteers was openly mocked and disparaged by members of the senior team. In the commercial sector, this would be seen as evidence of highly aberrant behaviour, yet the situation was left to fester.

For many junior staff, the situation was a living nightmare long before it became clear that the campaign itself was failing at the most basic level. I don’t think any of us realised quite what we were letting ourselves in for when we signed up. But what were we supposed to do? I came close to resigning as early as November but decided instead to try to make the best that I could in the situation.

In the end, I’m quite proud of what I achieved, winning the social media war despite having no advertising budget, and helping to raise an incredible amount of money online. I’m immensely proud of a lot of my colleagues who performed above and beyond the call of duty. And I would single out the new ERS chief executive Katie Ghose for praise; she was the only person with any actual authority in the campaign who seemed concerned about morale and improving communication. If she is given the opportunity, I am confident that she will go on to sort out many of the problems that have plagued ERS for over a decade.

But we were struggling on, having been shot in the foot and with one arm tied behind our backs. And frankly, the situation made us all complicit. I’m very aware of the number of times the stress and difficulty of the situation lead me to accept uncritically and even defend a number of things that, in retrospect, were quite wrong-headed.

A lot of Liberal Democrats have been calling loudly for John Sharkey to be held accountable in some way for the campaign’s numerous failures, and it has to be said that the buck did stop with him – at his insistence. He certainly does need to address his critics’ points. But the organisations that set up the campaign did welcome him with open arms, in retrospect with very little in the way of scrutiny. And the Liberal Democrats anointed him, having made him chair of the Liberal Democrat AV campaign and sending him as an emissary to reach out to the other relevant stakeholders. Months into the campaign, we heard numerous senior Liberal Democrats complaining about us putting him in charge, yet during the crucial planning stages of the campaign, such voices were conspicuously silent. It is crucial, after such a monumental failure, that everyone involved recognises their share of responsibility.

I will, however, end on an optimistic note. If nothing else, the referendum has clarified things. It has clarified the scale of the vested interests opposed to even the mildest political reform and the dishonest lengths to which they will go in defending the status quo. It has shown how important it is that advocates of a better, more inclusive form of politics actually practice what they preach. And it has demonstrated that the cynics can be far more naïve than the idealists that they are so quick to disparage.

Regardless of the fate of the Liberal Democrats, it is clear that multi-party politics is here to stay. With that in mind, electoral reform is liable to rear its head again far sooner than its opponents would like. If any good is to come out of this referendum at all, it is crucial that we learn from our mistakes and make sure that we are absolutely ready next time the opportunity arises.

James Graham is campaigns and communications manager for Unlock Democracy and a member of the Liberal Democrat Federal Executive. For the ‘Yes to Fairer Votes’ campaign, he worked as the web and social media manager. He writes here in a personal capacity