Part 2: Greater Democracy – History of DemComm

Following the previous post, a little bit on the history of DemComm.

If not for Mark Gorenberg…
One interesting anecdote from my time with the Kerry Campaign – when I originally negotiated my role, my primary tasks was to build the network of online Kerry supporters – through some form of structure which we had dubbed the Kerry Leadership Network. But as things quickly escalated, the KLN was quickly deprioritized over the course of the first two weeks of the camapign. When I arrived, the team was focusing on the Alexa stats that one of our earliest Internet supporters, Marc Gorenberg, was focusing on. At the time, Gephardt’s Alexa ranking was much higher than ours – and we needed to determine if the stats were truly reflective of our traffic.

After investigating our own traffic stats (which were confusing and wrong), it became obvious that issues concerning technical infrastructure were a higher priority (and we had Erin Hofteig and Dave Patten working with community members), I switched to my technical side and became the CTO. But, I always kept my involvement with community effort – working handedly with Dick Bell, Erin Hofteig and Dave Patten on our community efforts.

Aside from implementing an enterprise-worthy email CRM solution, one of our skunk-works projects were the creation of John Kerry personas on the social networking sites of the time (Ryze, Friendster, tribe.net to name a few). Erin’s work on these sites were so successful that we got a press hit regarding the number of “friends” JK was getting at Friendster – though one part of the story not reported was that, during the course of the campaign, JK’s profile was so popular and there were so many pseudo-JK profiles being created that Friendster had to create new features like an “official” profile and to overcome their limit of 500 friends at the time.

Returning to March 2003
As Kerry began to win primaries and other campaigns began to fold, we began to bring on more staff to handle the increasingly complex management of a national campaign. In the Internet, bnuilding an organization that three months prior, was the size of a small startup and now had to service the needs of a customer base the size of Amazon.com with similar needs in terms of customer service – the tasks was incredibly daunting.

As experts became available, we were hiring on members to become part of the Kerry team. It was in late March/early April that we hired Amanda Michel and Zack Exley to fill out our community efforts. Amanda came to us from the Dean Campaign, highly recommended by Jim Moore for her work through Generation Dean. Zack was brought on to focus on community – and, at the time, focused on the execution of effective email campaigns.

I had met Cam at eTech and lobbied for him at Kerry because of his exceptional work with the Clark Campaign and the Clark Community Network. He came onboard to spec out the next generation of a community platform that would engage our supporters beyond the typical forum (which had been staffed and maintained by our supporters) or blog (run by Dick, Peter Daou and Ari Rabin-Hayt).

Creating DemComm
The challenged we faced at this time was dealing with various issues that startups see frequently when they have explosive growth – moving from a simple infrastructure and small staff to an enterprise-worthy infrastructure, serious customer service and integrating an infusion of new staff. Couple this challenge with the fact that we were not be allowed to “officially” work with the DNC until we were “officially” the nominee (though being presumptive allowed us to finally begin our integration efforts), we had a extremely large task in terms of technical infrastructure. And, with Amanda, Zack and Cam joining us in April, we were trying to ramp up quickly.

From discussions with online futurists (like David Weinberger from ClueTrain Manefesto and Howard Rheingold), it became obvious that creating a team of community experts and leveraging some of the collaborative technologies could get us further along in our underestanding and development of a online community strategy that was far beyond our personal bandwidth and budget.

While we were attempting that model with demtech (and having difficulty with specific requirements to help us build a solution), the concept was still sound and lead to the creation of DemComm.

Creating another skunk-works project
On April 2nd, seven people were added to the DemComm group which, as Jock mentioned, included Amanda, Cam, Howard, Nanci, Jon, Jerry Michalski, John Coates and myself. One of the problems I faced was, while I knew this was an important effort, I had a number of other projects I had to address (including completing the infrastructure buildout for the website, a new Online Action Center, and a total rebuild of the contribution, marketing and reporting engines by our tech team), we agreed that Amanda take charge of this group – since she had the strongest reporting connection through Zack.

As Amanda wrote:

Howard is absolutely right – the success of our efforts will rely on wide acceptance throughout the campaign and the grassroots (via bi-directional communication). For example, promotion of the online community and its efforts and aims needs to become part of the campaign’s message to the grassroots. Of course, self-organizing won’t work if our architecture won’t
support it.

A plan is the best way to leverage support and involvement throughout the campaign. 100% support won’t make the difference, but substantial support will. Pushing for the plan early on will also ensure that our efforts get worked into the campaign’s general election strategy.

Pushing for direct action throughout our community is essential. And not just because of its immediate results and the benefits of immediate engagement – asks for direct action are good reminders of our goal.

I propose that during our meeting on Thursday that we discuss the process for putting together the plan. Most everyone has raised questions that need to be answered, or at least addressed before we make any big decisions. Let’s figure out how to pursue possible solutions in a timely way – and how
to delegate work among us all. We don’t have much time to put this in place – six to eight weeks is the max for planning time.

The team began to work on the plan – and, through the hard work of the people on the team, we had the initial draft that Jock shows on the Greater Democracy post by the self-imposed deadline. The challenge we had was, at that time – the campaign was focusing on fundraising, staffing up and the insanity behind building up for the coming Convention.

But it should be clear – that DemComm was another skunk-works project: no one in the senior staff (with the possible exception of the Dir of Internet) knew about DemComm. We all knew that the goal was to prepare a proposal for the campaign that would be guidance for development – and help in supporting Cam’s community solution. Cam’s proposal (which I think is still one of the better ideas the campaign generated at the time – combining the best parts of threaded discussions, forums AND blogging) – was not accepted due to cost concerns and potential political liabilities (“What if someone said something on a Kerry Community blog that was racist or anti-American? Even though it came from an outside user, it still is a Kerry-branded site…”) So, while we had some of the best people working on DemComm, the challenge was – as a priority, the community effort had a very different focus. Which leads to Kock’s point of “sheeple” and how the campaign operated in 2004. And that is the topic of the next post.

Tags: DemComm, Kerry Campaign, online organizing, eCampaigning, Greater Democracy, Howard Rheingold, Jerry Michalski, Amanda Michel

This entry was posted in Campaign 2004. Bookmark the permalink.

1 Response to Part 2: Greater Democracy – History of DemComm

  1. Pingback: Weblogsky » Blog Archive » DemComm

Leave a Reply

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

This site uses Akismet to reduce spam. Learn how your comment data is processed.