You would have been hard pressed to know what was happening in Alberta politics by watching a national newscast this fall.  Outside the province of Alberta, there was little to no coverage of the race to become the leader of the Progressive Conservative Association of Alberta, and with a victory, the Premier of Alberta. 

A very similar election was held in British Columbia earlier this year, as Christy Clark won the leadership race for the BC Liberal Party, and took the Premier’s chair.  Again, unless you were in BC, you probably didn’t hear anything about it. 

What is unfortunate about this lack of media coverage is that party leadership races in Canada are probably the closest things to dog-eat-dog US style political campaigns.  The political stakes can be at a very high level (in this case, becoming the Premier) and candidate’s spending limits can be unlimited, as in the case of Alberta, or have a very high cap, such as the $450,000 cap placed on BC Liberal Party candidates.  Compared to actual provincial candidate election spending, which is usually in the $30,0000-$60,000 range, these campaigns create an allowance for all types of new approaches that would be otherwise eschewed for traditional lawn sign and door-to-door campaigns.  And with high-stakes campaigns, all sorts of interesting US style tactics (including those some consider “dirty”) come into play.  

However, in a leadership race, the real-world goal is very different from a traditional majority ballot in a riding.  In both Alberta and BC, the leadership contenders were subject to a process controlled by their party, and not the public at large.  In order to have cast a vote for the party leader, you had to be a party member.  So recruiting party members to vote in your favor was the primary goal of the campaigns. 

The secondary goal was to influence members who were not voting for you as their primary candidate, to add you as their second choice on their ballot.  For example, the race in Alberta was to be the first candidate to cross the 50%+1 of ballots cast threshold.  If on the first ballot (held on September 17, 2011), there was no outright winner, the contest would head to a second ballot with the top three vote getters from the first ballot going into a run-off.  In the run-off (held on October 1, 2011), if there is no clear winner, the third place candidates’ second choice votes are split between the remaining two candidates to declare a winner.  In BC the system was a single vote process with the secondary votes from the last place candidate forward being added to the remaining contenders until a winner obtained 50%+1. 

This system created a difficult balancing act for the campaigns.  In Alberta, where six candidates were contending, the campaigns had to watch how much they are directly communicating with their current registered base.  Over-communication can create a nasty backlash, especially with five other candidates contending for the same pool of votes; generally double the number candidates for a riding in a general election.  The other issue was the traditionally slow media period in the summer, where the public and media interest in the campaign waned. 

In Alberta, the campaigns turned to social media and targeted marketing as ways of attracting new members to the party and for building influence within the party to be the second choice on the ballot. 

The Cattle Barron: Gary Mar

Courtney Luimes, who was the social media co-ordinator for the Gary Mar campaign, outlined that their strategy was to first and foremost to get Mar in front of as many Albertans as possible in order to sign new members and influence party members to add Mar as their second choice.  It was an old school campaign, but social media was invaluable in that process of organizing and getting people to attend events for Mar.

As a secondary benefit, social media helped the campaign rapidly respond to an attack made by candidate Rick Orman.  Orman had attacked Mar on comments he had made about privatizing healthcare in Alberta.  Mar’s response, entitled “Encouraging Healthy Dialog” was carried on his website and both Twitter and Facebook in response. 

Targeted marketing was also an effective tool, and the Mar campaign leveraged the widely used customer relationship management portal Salesforce (www.salesforce.com) to store member information and send out targeted emails.  This is one of the first campaigns in Canada to move away from proprietary software generally used by political parties to a more open and modern software-as-a-service (SaaS) cloud based offering. 

Using targeted campaigns managed through Salesforce, the Mar campaign was able to emphasize specific platform points through emails timed and sent just as the campaign heads into a specific town or region.  The result has been open rates between 40% and 50% in targeted regions. 

The Poker Player: Doug Griffiths

“We had to go ‘all in’ on social media,” explained Troy Wason, the campaign manager for the Doug Griffiths campaign.  The Griffiths campaign did not have the financial means of Horner’s or Mar’s campaigns.  It also did not have the MLA or cabinet minister endorsements that help draw the core member ballots, so they have had to push all their chips in on social media.  Griffiths as a result has turned his attention to Twitter. 

Wason used tools like Twitter Adder (http://tweetadder.com) to target engaged Albertans on various Twitter hashtags such as #abed and #ableg.  The tool configures your Twitter account to follow anybody using the hashtag, and invites him or her to follow you back.  The result of this approach is that the number of followers ends up equaling the numbers of people you follow.   This approach is also derided in some social media circles as “astroturfing” where you are building your social media base without knowing for sure if the people you are adding will truly support you. 

Wason’s counter to this claim is simple; he not only wanted to, but had to engage all types of Albertans to win.  Griffiths, who personally handles all his own tweets, had out-tweeted all his opponents approximately 10:1 by early September. 

The Good, the Bad and the Ugly

What has been hard to measure is the direct impact these tactics have had on the ultimate outcome of the leadership race.  Namely, what are the real-world effects of social media clout?  Looking at Christy Clark and some of the other notable campaigns and politicians in the US and Canada, helps to derive a sense of what that successful clout benchmark is:

 

Candidate Facebook Likes Tweets Twitter Following Twitter Followers Global Website Ranking (www.alexa.com)
Christy Clark (@christyclarkbc) 3836 549 4402 9398 http://www.christyclark.ca/: 2,987,219
Naheed Nenshi (@nenshi) 20,787 6194 844 26,618 n/a
Danielle Smith (@ElectDanielle) 3236 3233 1629 5035 http://www.daniellesmith.ca/:  14,354,538
Stephen Harper (@pmharper) 65,120 471 12,256 166,202 http://www.conservative.ca/leader/: 665,622
Jack Layton (@jacklayton) 193,831 926 11,267 126,286 http://www.ndp.ca/jacklayton: 149,931
Cory Booker (@CoryBooker) 51,494 11,237 57,367 1,097,557 http://www.corybooker.com/: 2,665,645
Rick Perry (@GovernorPerry) 147,958 1,192 31,214 79,563 http://www.rickperry.org/: 95,747
Barack Obama (@BarackObama) 22,950,522 1,777 690,138 9,957,162 http://www.barackobama.com/: 7247

 

Data: September 6, 2011

A pattern that emerges among these successful social media politicians is that your Twitter followers and Facebook followers should relatively align with each other (say within a scale factor of 2-3x), and a person with clout generally follows approximately an order of magnitude (10x) fewer people than are following him.   There are also two types of politicians: those that tweet for themselves, and those that let others tweet for them.  The tell tale is the number of tweets.  The larger the number of tweets, the more obvious it is that the person is tweeting for him or herself and engaging in conversations through Twitter.  It’s also very clear that you can be big on Twitter without tweeting for yourself (ex: Clark, Harper, Layton, Perry & Obama), but it’s also clear that those who do tweet for themselves  (ex: Nenshi, Smith, Booker) can build a support base that can provide the necessary votes for leadership.   Using these models as the archetypical successful clout profiles, it is then interesting to judge the current clout of the candidates in the Alberta race:

 

Candidate

Facebook Likes

Tweets

Twitter Following

Twitter Followers

Global Website Ranking (www.alexa.com)

Doug Griffiths (@griffmla)

789

3112

3291

3292

http://www.betteralberta.ca/: 2,519,105

Alison Redford (@alison4premier)

2570

471

695

1778

http://www.alisonredford.ca/: 3,101,711

Gary Mar (@garymarpc)

942

307

373

1391

http://garymar.ca/:7,236,310

Doug Horner (@hornerforab)

3808

250

85

1018

http://hornerforalberta.ca/: 6,464,518

Ted Morton (@morton4premier)

5886

163

157

880

http://tedmorton.ca: 12,313,705

Rick Orman (@Rick_Orman)

210

265

285

712

http://www.voterickorman.com/: 2,946,315

Data: September 6, 2011

As of September 6, 2011, none of the PC Association of Alberta candidates were close in terms of absolute numbers to the successful candidates.  This was likely due to the lack of interest and media coverage in the campaign to this point.   It is also apparent at this stage that there is no candidate so numerically far behind that they could not otherwise catch up.   

But there were some interesting patterns that stood out in the various campaigns. It’s clear at this point that both Ted Morton and Doug Horner had focused on collecting Facebook followers.  Morton’s campaign team’s approach was to Google advertise his Facebook site instead of his website, which had seemed to hurt his website ranking, which was the lowest of all six candidates.  This would also seem like an odd tactic as your website is where you would want to influence people to signup for memberships. 

Doug Horner used a slightly different approach, running Facebook ads posing questions about people’s love or hate for various aspects of the province and its current policies, that when clicked automatically “Likes” his Facebook page.  While it was a sneaky tactic, it clearly helped build his Facebook base.  However, like Morton, it didn’t drive much traffic to their campaign website. 

Rick Orman was clearly running an early-2000s style campaign, as his social media numbers were barely registering, but his website ranking was good.  This would likely indicate that his team has chosen to run an email campaign over employing a more modern social media strategy. 

Alison Redford was the most balanced of the campaigns in terms of across the board social media numbers, but it was still too early to tell at this stage with the bulk of the campaigning about to happen.  Gary Mar and Doug Horner were both better funded and endorsed, leaving them in the best position to spring forward in a sprint to the first ballot. 

As of September 6th, Doug Griffiths had a reasonable social media clout lead, and it had translated into at least one important real world activity for his campaign: donors buying fuel for Griffiths’ truck. Griffiths’ campaign has listed on hits website 6 things people can do to support him, and one of those activities was buying a tank of gas for Griffiths’ Ford F150 truck.  The result has been donations coming in frequently in the $40-$100 range to cover the fuel for his campaign.  With this support, Griffith has logged over 100,000km driving to debates and campaign stops, all helping him get in front of as many voters as possible.

But would it translate to votes on September 17th

Shootout at the PCAA Corral

Between September 6th and the vote on September 17th, some interesting shifts occurred as the race took off.  The first was that interest in Mar’s, Horner’s and Morton’s campaigns all increased significantly as demonstrated by the website ranking increase.  This was likely due to the increased media attention and interest.  Redford and Griffith’s website rankings experienced smaller percentage gains, and both maintained their lead over all other candidates. 

YouTube views were heavily in Mar’s favor (12,603), with Griffith’s a close second. (7,902) and Morton (5,577) in third.    In the new media realm, Mar was in control. 

However, the real-world votes was all that mattered on the first ballot.  Mar took double the number of first place votes of Redford, with Doug Horner sliding into third place.  Mar then received the endorsements of Orman, Morton and Griffiths going into the run-off.  Had those votes all tallied up on that evening, Gary Mar might likely have been the Premier of Alberta today. 

Griffiths’ social media lead did not turn into many votes, however he likely captured every vote he received through his social media channels.  The all-in social media strategy fell before the voting block of the insiders.  In poker terms, he was drawing dead. 

 

Candidate

Facebook Likes

Twitter Followers

Global Website Ranking (www.alexa.com)

Votes

Votes: Facebook Ratio

Doug Griffiths (@griffmla)

838

3435

http://www.betteralberta.ca/:    2,193,040

2,445

2.952898551

Alison Redford (@alison4premier)

2724

1942

http://www.alisonredford.ca/:2,096,358

11,147

4.092143906

Gary Mar (@garymarpc)

1093

1567

http://garymar.ca/: 3,378,258

24,445

22.36505032

Doug Horner (@hornerforab)

4387

1089

http://hornerforalberta.ca/: 4,101,373

8,648

1.971278778

Ted Morton (@morton4premier)

6222

973

http://tedmorton.ca: 5,483,667

6969

1.120057859

 

Rick Orman (@Rick_Orman)

227

778

http://www.voterickorman.com/:2,735,872

6010

26.47577093

Data: September 18, 2011

What happened in the two weeks between Mar’s massive first ballot victory and the final outcome could only be summarized as stunning. 

The system used to select PCAA leaders had never resulted in the first ballot winner going on to win the party leadership, however nobody had achieved the first ballot margin Mar’s campaign had, and nobody had the number of endorsements he had collected.  Going into the final two weeks, he looked poised to overcome the system handily. 

The Marshall: Alison Redford

While Mar was the clear leader, he was also a polarizing figure, and it was that polarization that allowed Redford to position herself effectively against him. 

Redford had only one MLA endorsement, which made her look like a party outsider.  This was also played up in the media, but within the party everyone knew she had worked for Joe Clark for years, giving her plenty of unspoken credibility.  Redford was also able to coral support from more progressive elements that were critical Premier Stelmach’s plan to cutback on education spending.   Redford’s sold a large number of PCAA memberships to teachers, Liberals and Alberta party members looking to reverse this plan. 

Redford then wisely endorsed Doug Horner as her second choice on the ballot.  Even without having Horner openly endorse Redford as his second choice, it created an atmosphere that there was a quid pro quo voting arrangement in Redford’s favor in the Horner camp. 

The last event that took place had nothing to do with political tactics or planning.    Redford’s mother passed away.  Unfortunate in any circumstances, it brought an extra amount of pressure to bear on Redford personally during the final stretch of her campaign.  Right after the death of her mother, Redford attended the final leadership debate, and with all eyes on her, she performed with composure and strength. 

 

Candidate

Facebook Likes

Twitter Followers

Global Website Ranking (www.alexa.com)

First Choice Votes

Total Votes with Second Choice Votes

First Votes:
Facebook Ratio

Alison Redford (@alison4premier)

2724

1942

http://www.alisonredford.ca/: 1,694,808

28,993

37,104

 

8.132678822

Gary Mar (@garymarpc)

1093

1567

http://garymar.ca/: 1,913,764

33,233

35491

26.39634631

 

Doug Horner (@hornerforab)

4387

1089

http://hornerforalberta.ca/:3,409,438

15,950

N/A

3.37995338

Data: October 2, 2011

Going into the final balloting, this setup Redford as the sentimental favorite, while Mar was still the numbers favorite and had momentum.  Redford had pulled close enough with the last surge in memberships collected, so when Mar didn’t receive the majority of votes on the second ballot, it was clear the system was going to decide the premiership, and that system favored Redford and the broad support she had collected through her maneuvering during the final two weeks of the campaign.   

For the Mar campaign the final results painfully proved that the system doesn’t pick the candidate with the most first choice leadership support within the party, but the one with the broadest amount of support from the membership.  While the goal was to influence both first and second choice voting, the campaign was positioned to win on the first vote.  Clearly here, the balanced campaign run by Redford proved to be the most effective in capturing both first and second choice ballot results, even if it catered to more progressive elements. 

Looking at the real-world vote to social media scoring, Mar clearly had a huge vote capture to Facebook ratio through the first round and second, but Redford doubled her vote capture to Facebook ratio from 4 to 8 between the first and second round.  This is a measure of the campaigns digital clout; their ability to draw out supporters and get them to vote in the real-world.  This doubling was a good retroactive measure of the momentum in the Redford campaign, but is not the real-time social media index we are all looking for as campaign managers and pollsters.  The use of better ‘digital asks’ to test the engagement of supporters to vote in the run up to elections is needed to get a more accurate picture of the real clout of a candidate.  

Sidebar article: The Loyalty of Twitter Followers

Dr. Anatoliy Gruzd is an assistant professor at the University of Dalhousie in Halifax and performed an interesting analysis of the 2011 Canadian Federal election this spring. Below is an excerpt from a blog post he did during the 2011 Federal Election call “Imposters among the believers… fakers and spammers “following” party leaders” (http://socialmedialab.ca/?p=1952). Dr. Gruzd found two metrics especially useful in measuring the strength and the level of engagement with a candidate’s followers on Twitter: their total number of “loyal” followers and the total number of “fake/spam” followers. By considering these two metrics in conjunction with each other (i.e., “fake/spam” accounts are identified and removed from among the “loyal” supporters of each candidate in a race).  An interesting picture of which leader had the most “loyal” supporters was created. 

Loyal Followers/Supporters

For the purpose of this test we define “loyal” or “dedicated” followers to be those who follow only one of the party leaders. To estimate the number of actual party supporters (“loyal” or “dedicated” accounts) for each leader, we counted how many users only follow a single candidate (See the chart below). As of April 15, Harper had the highest number of loyal followers among the five party leaders; out of about 126k users who follow Harper’s Twitter account, 51% do not follow any other party leaders. Duceppe came in second with 39% of loyal followers. While Ignatieff, Layton, and May have about the same number of “loyal” followers, estimated in the 27-29% range.

Fake/spam followers (Imposters among the Believers)

Due to the open nature of Twitter, there are a lot of different bots (automated programs) that can easily create fake user accounts and start following real accounts in an attempt to disseminate various types of information or simply increase one’s number of followers. One of the main characteristics of bots is that they tend to “follow” many accounts, but very few would follow them back. In order to assess how many bots follow each party leader, we calculated the Follower-Friend Ratio (FFR), for every “loyal” follower, as defined earlier, for each of the 5 party leaders. FFR is a ratio of the number of followers to the number of friends (accounts that are being followed back). For example, if a user follows 100 accounts, and only 50 accounts follow that user back, then their FFR = 50/100 = 0.5 which is considered low. Most of the “human” accounts have a ratio close to 1 (following as many accounts as they follow back). Celebrity-type and news accounts would have FFR higher than 2. For example, Harper’s FFR = 10 and Ignatieff’s FFR = 8. For the purpose of our analysis, we flagged all accounts with FFR less than 0.1 as potential bots. Below is the chart with the results.

Of all the leaders, Duceppe has the highest number of followers, which exhibit bot-like behaviour; 41% of all of his “loyal” followers have FFR<0.1. Elizabeth May has the least number of bot-like followers (only 9.6%). Harper, Ignatieff, and Layton have between 17% to 22% of such accounts. This suggests that of the five party leaders, May pays the most attention to who follows her on Twitter and Duceppe the least. It also suggests that whoever is maintaining May’s twitter account is actively culling and weeding the Green Twitter follower base to prevent spammers from taking root and that Duceppe’s Twitter account is open to all followers whether they be “human” or “bots”. Further analysis is necessary to confirm whether or not this is a case of a computer-savvy supporter or detractor creating fake accounts to follow Duceppe’s account and driving his numbers up, or is it simply the case that Duceppe’s social media team decided not to clean up their Twitter account from bots and spammers who can easily and quickly take over one’s followers’ list.

 

 

John Craig is the VP of Sales and Marketing at Purple Forge (www.purpleforge.com) that provides iPhone applications to politicians and political parties across North America, in the UK and elsewhere.  John was named to the 2011 Aristotle Dream Team for Mobile Campaigning around the Globe. 

  

Tags: Alberta PC leadership race, Alison Redford, BC Liberal Leadership, Christy Clark, Doug Griffiths, Gary Mar, Rick Ormandy