“Harper’s Team” reviewed: Lessons for the Green Party

Magikal Tome?

Magikal Tome?

Two weeks ago I picked up a copy of “Harper’s Team” by Tom Flanagan. I had seen a recommendation in Report on Greens, and being naturally interested in a first hand view of how the conditions were created for the rise of the Conservative Party from has been to the currently dominant political force in Canada.

I have spent, (according to my wife, wasted), thousands of hours canvassing, fund raising, devising and designing GOTV programs, implementing databases, and data management programs on behalf of various GPC campaigns.  The sundry challenging tasks of Green Party cat herding, that some call campaign management have taken their toll. I yearned to find the secrets to how to really make it work. ‘Surely’, I asked myself, ‘there is some magic potion that these Conservative rascals imbibed to bring them invincible strength? Perhaps the formula lies between the dusty covers of this rare tome?’

Now I can reveal the truth to my readers. Nope, no magic potions here. Once again, the truth’s within those dusty (actually shiny new) covers is prosaic. The most important thing about it is that it doesn’t come from an obscure Green activist, and businessman crying ‘but it’s OBVIOUS what we have to do.’ It is coming from the Campaign Manager that deployed prosaic tools and organizational principles to seize power within our Parliamentary Democracy.

My first comment is a negative. This book isn’t a work by a master of communication and disinformation like Warren Kinsella. Kinsella’s book, “The War Room” rambles a bit, and is kind of heavy in the name dropping, and self congratulation departments, but you have no doubt that you are imbibing the real brew.

What this book is, is a well written eye witness narrative of Harpers career from 2002 on. It shines a light on the early history of the Alliance, and the subsequent PC merger that gave birth to the CPC. I won’t go through the nitty gritty of who did what, with whom, and for whoever. I will once again filter the book for lessons for the Green Party to apply.

Start every campaign with analysis, and a plan. SWOT analysis, familiar to business managers stands for Strength Weakness Opportunity Threat. Look objectively at all sides of the equation. Develop a strategy to maximise, minimise, exploit, and counteract respectively. Sounds easy? It isn’t. It needs to be concise, and readily grasped and expanded on by a whole team of individuals.

From that starting point, develop the campaign theme, and message. This shouldn’t be an ad hoc exercise. You might start with a gut feel, but you pursue professionally conducted political issues research. That means opinion polling, with a focus on what messages, and presentation will achieve your strategic ends. For example, if you are the strong front-runner, and your primary objective is to protect your lead, and resist encroachment, then you will research amongst supporters to determine why they support you, and what message is most likely to re-enforce this support.

The structure of the campaign is important. To a certain extent, the strategy will be formulated based upon the skills available. The managers task is to ensure that the right skills, and people are in place to implement the plan. A strong chair is essential, because there will always be stress and tension within the team. The Chair must ensure that the team ends up pulling together. That means managing, and yes, even manipulating these tensions to spur on better results.

Flanagan stresses the importance of advance planning of the campaign based upon a campaign calendar. My personal opinion is that the emphasis on planning every single day of the Campaign which he believed was essential proved to be the CPC’s undoing in the 2008 campaign. Their plan was so rigid, that they continued to try to control and direct the message, even while stock markets were collapsing around them. They failed to adapt, and it made them look totally out of touch for the tail end of the Campaign. Kinsella understood the need for adaptability, and stresses that again and again in his book. Had Kinsella run the Conservative War Room, we would probably have a CPC majority government today.

For the actual mechanics of campaigning, this is where Flanagans book gets really interesting. Arguably the single most important tool that the CPC and Stephen Harper have employed is direct mail fund raising. In different circumstances, his various campaigns were able to direct hard hitting, very specific appeals for cash to their various mailing lists. There would always be some specific objective and appeal. Not just a generic, ‘give us some cash’ but more like, ‘together, we need to stop X,Y, and Z happening. We need this much to do it. Give generously using this postage paid reply envelope.’ Every campaign started with such direct appeals, and they really worked. The CPC continues to go back to the well again, and again, and this tool forms the basis of their financial muscle.

Another interesting outreach tool discussed in depth by Flanagan is various types of phone bank. One of the earliest tools employed was predictive dialler’s. Flanagan assumes his reader knows what this is, but for the uninitiated, I’ll offer a brief description. A computer can be adapted with a special phone board to manage multiple phone connections at once. A single board can handle up to 32 separate lines at once. Predictive dialling software/hardware setups will automatically dial 32 people at once, and there will be a number of separate phone operators networked with this dialler. When someone receives such a call, they can either be patched through immediately to a live operator, or they could have pre-recorded option presented to them like ‘dial two to speak to an operator’. Using fairly small teams of volunteers, the campaign can plow through large numbers of phone calls fairly quickly. More sophisticated types of dialling technology can be employed, depending upon the task being performed, and the numbers involved.

The lessons ultimately learned by Flanagan were as follows: For ID’ing the vote, on a really large scale, the best way to make a large number of these relatively simple phone calls is to hire a professional telemarketing firm. There is little, or no persuasion going on, other than a few simple stock pitches, so paid telemarketers can do this very efficiently indeed. The calls can be targetted by sorting a large database to separate the polls which had good Conservative (or Green), turnouts in past elections. That way, the best response rates can be guaranteed. The CPC does this ongoing, even outside of election periods. It enables them to build up supporter databases so they can go to these people to Get Out The Vote in future campaigns, and it gives them hot pre-qualified prospect lists for their direct mail fund raising letters.

For persuasion type phone banking, volunteers are better. They will generally have more policy knowledge, and more interest in spending some time persuading people to change their vote. The Conservatives actually own their own predictive dialler’s, and set up phone banks in their War Room, or wherever the volunteers can most easily be gathered. They use these teams to target undecided voters, to convert them to Conservatives. They will also target soft supporters of the other Party’s. As I have mentioned elsewhere, they could have the twin objectives of either converting these votes, or suppressing turnout for the opposition Party’s. Volunteers are much better at raising money as well. Because they cost nothing in wages, and can be more adaptable, a volunteer phone bank can generally be relied on to break even, or even turn a small ‘profit’ by asking for donations as part of their standard pitch. Any type of phone banking can be targeted anywhere in the country. If the campaign calls for an extra effort in one particular riding, then all the phone banks can concentrate on it, until it has been thoroughly canvassed, and persuaded. Obviously, this calls for good data management, and co-ordination.

The Data management discussed was, in my opinion, the single most important innovation of the Harper team. They invested some pretty serious money in creating a single database hosted by the National Party headquarters. The reason this was so important is because, just like the Green Party, the majority of the contacts made at the local level during a campaign would not be recorded, or preserved between elections. These supporters would therefore not be accesable to co-ordinated fund-raising efforts. They would also be lost before the next campaign, and the voter ID effort would have to begin from scratch. The Green Party of Canada has made some feeble efforts along these lines with their CIVICRM, but frankly the software is crap, and no-one in their right mind would base their campaign data management on such a time consuming clunky piece of junk. What is needed is a fluent, easy to use database. It needs to have easy user interfaces, and should be blindingly fast, no matter how many users there are.

Flanagan wrote a lot about message, and the air war. The air war consists of advertising, plus earned media. Frankly, while it makes for OK reading for a campaign buff like myself, if you want to learn about communications, read Warren Kinsella’s “The War Room”.

In conclusion, the Conservative Party has many lessons to teach the Green Party of Canada. It’s not that the Green Party doesn’t know these lessons. It’s more that they need to have their noses rubbed in the fact that Yes, these things work, and are capable of propelling an extremist party into government when used correctly. (hint hint)

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to Ma.gnoliaAdd to TechnoratiAdd to FurlAdd to Newsvine


9 Responses

  1. Hey great post, I’m going to need to buy this book! Just to clarify one thing we use CiviCRM to track doners and members but not for voter ID or GOTV. For voter ID and GOTV we use GRIMES which was developed by Craig Cantin.

  2. I’m glad to hear you extracted the same message from Harper’s Team. Knowing your background, the “must-read” status was more for fledgling Greens who are still pretty politically naive.

  3. Hi Mark,
    I still think this was a must read. It’s like Kinsella, the more you know about your foes, the more effective youy will be against them. I think that the command and control model of politics has some inherent weaknesses that could be exploited.
    Hi Dave,
    The last I used CiviCRM was when it was being touted as an ID-GOTV database by the hub during the London North Centre by-election. It was crap, and we quickly ramped up a local database app. for these purposes. Does Grimes work?

  4. Hey BGB,

    We did use an add on for CiviCRM called CiviCanvass for the LNC by-election but it was total crap.

    When Craig replaced Pierre as the manager of technical services he developed GRIMES which is an acronym for something but I can’t remember exactly what.

    I’m pretty sure we first used GRIMES for the Toronto Centre by-election (again I might be wrong but I’m pretty sure it was the Toronto Centre by-election.) It’s good and getting better but for its effects to really be felt all the campaigns need to use it because it is worth more the second campaign you use it because you get to build off of the last campaigns data.

    I really think Craig did a good job, it does most of what we need in terms of data management, and performed well enough in the past elections. The problem was that very few ridings used it which was not a failure on Craig’s part. He developed it and supported it and encouraged people to use it BUT there wasn’t enough leadership from the organizers (not all organizers.)

    • I wouldn’t mind seeing it.
      Don’t be surprised if not a lot of campaigns used it. The vast majority of campaigns don’t do any ID-GOTV. I remember in the runup to the 2006 campaign, there were campaign schools, and generally an effort to identify target polls to canvass and ID the vote in. As far as I can tell, Toronto-Danforth, Toronto Centre, Etobicoke Lakeshore, Beaches East York, Trinity Spadina, Parkdale High Park, and a couple more Toronto area EDA’s actually put resources into ID-GOTV. It was tough though. A lot of canvassing to scrape together a few hundred supporters names. I suspect that a great many of those people who learned the basics then have dropped out of the GPC since.
      How do you populate GRIMES with data? Does the GPC load it up with the national electoral database? What about phone numbers? If there is national data, plus phone numbers, provided the the voter list revisions, and advance polls get updated, then you’ve eliminated the need for a campaign to have a competent data jockey in order to conduct a proper canvas. Has to be searchable by poll, polling station, street names and address ranges. Also the big campaigns would benefit from barcoded paper printouts. Saves a lot of data entry after a long days canvassing. The ideal will be massaged to target polls by demographics, and census supported Green voter profiles. That would trump the conservative system, and really squeeze the last drop out of our limited volunteer base. (See my posts on supporter profiling.)
      Keyboarding and data entry is always an issue though. If the GPC ever manages to roll out a decent national database, then they’ll grow with every subsequent election. Especially if systematic mailings go out to ID’d supporters.
      I know that’s pretty ambitious stuff, but if enough council members were to push for truly campaign centric planning and preparation by the hub, then this is easily within reach. If you’ve been on council though, you’ll know just how unlikely it is that they could ever focus on something so mundane as election readiness.

  5. GRIMES is a web based system that allows people to do various campaign tasks based on their level of access (different users have different permission sets.)

    This past election was our first general election to use the software and we started by loading the Elections Canada voters list into the database.

    For canvassing purposes users are assigned to a riding and to a poll in that riding. They can login and make canvassing calls (each riding can make their own call script which is displayed on screen) or print off a voters list for their poll so that they can foot canvass.

    When you make a canvassing call you enter information into a HTML form and click the next call button to make your next call so data entry doesn’t take much time.

    We do have many phone numbers but not everyones phone numbers. One thing that we did quite successfully in Guelph was in the office there were a few phone books and people who for various reasons wanted to volunteer but didn’t want to go door to door would look up phone numbers for lists of voters (by poll) that the office manager printed off from GRIMES.

    It is searchable by poll. I don’t think it is searchable by polling station. It is search able by address, phone number, first name, last name. Searches are limited to a particular riding so if you search for “John Smith” you only get the John Smitth’s you need to worry about. Lots of printouts and reports are available but not barcoded. There is currently no demographic system in place but Craig is open to suggestion but to be fair the vast majority of his time developing this software was done unpaid during evenings and weekends.

    It really is a great tool. It’s not at the level of the conservatives or the liberals but I have heard that we might have a leg up on the NDP who I have been told still rely on Microsoft Access.

    You’re exactly right about the database getting better and bettter with each elections which is why it is unfortunate that more people didn’t use it since the riddings that did will have such a great head start.

    It also keeps track of signs so in the next election the campaigns that ID’d supporters will be have a ready to go sign list so that they can get a lot of their signs up on day one.

  6. Hey, GRIMES sounds like a real leap forward! I’ve got an access canvassing database that worked pretty well in the past few elections. It drives autodiallers, providing you have a conventional Bell line, and an old dialup modem to generate the dial tones for you. We basically doubled the throughput on phone canvas. It’s the presentation, and search functions plus the extra data I scarfed that made it good. Ethnicity, and historical political data, like Party supported last time, sign taker for which Party, renter/owner/spouse. School board, (Catholic or Public) I should throw up some screen shots. One of the tricks was to be able to automatically display all the different electors at a single unique address. That way you can ask for everyone in the same household by name. This was a really tricky wrinkle that took a lot of effiort to iron out.
    For phone numbers, getting them can be laborious if you don’t purchase the big list from Bell. I have heard of it being done by writing a script that repetetively queries Canada411, or one of the other online equivalents. There is a certain amount of labour, and I’m not sure it is permitted use of the webservice, but About 25 volunteer hours located, sorted, and matched about 65,000 numbers in 2005-6-7 for three campaigns I worked on.

  7. […] as I explained in this previous post. Read ‘Harper’s Team’, or at least read my synopsis of lessons for the GPC.  Start sourcing subcontractors for direct voter contact. Even better, hire a Call Centre […]

  8. […] posted from `Not an Official Greem Party Site` dated February […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: