Measuring the MySpace Primary

Chris Bowers at MyDD recently wrote about the “MySpace Primary” - the race to acquire visibility and support among social networking users. The blog is worth a read, but some of the real action was in the comments, two of which struck me as great observations:

From Shaun Appleby:

Just a couple of comments. The Obama Facebook groups One Million Strong for Barack and Students for Barack Obama (which itself has over 52k members) have discussion boards (equivalent to MyDD diaries) numbering, as of now, 326 and 269 respectively. These attract between a handful to several hundred posts each, averaging out to about a dozen or less, just on a guess. The content of the posts just on a skim are roughly as serious and informed as those at DailyKos, both pro and con, if less self-conciously ‘Democrat’ and seem to cover a range of relevant and frivoulous topics. Seems to me there is some actual content in these groups which qualify them as another ‘branch’ of the netroots whether they are aware of it or not.

I think this is probably right - especially the bit at the end about this being a new branch of the netroots. It requires that we broaden our definition of the netroots to include more than blogs, but there’s nothing wrong with that. I don’t see this developing into a permanent piece of infrastructure, like the blogs. Rather, MySpace and FaceBook - and any other social network - offers the social space and the technical resources for political activity to coalesce for brief periods (a campaign cycle) and then disperse when the goals are met or interests wane. Much like SMS and MySpace allowed students to organize for last spring’s immigration protests.

The next big question is channeling discussion into activism and measurable activity. Metrics in particular will be important to campaigns as they seek to convert “friendship” and popularity on social networks into tangible support. I can think of a couple metrics of the top of my head, but I’m sure there are plenty more (leave ideas in the comments).

  • Fundraising: directly to the campaign via ActBlue and a strategy similar to this effort by an Obama FaceBook group; or indirectly, perhaps using ChipIn Widgets and their own social networks to form the Millennial/Internet equivalent of Bush “Pioneers” and “Rangers.”
  • In-site advertising: yeah, I mean campaign badges, but I also mean encouraging supporters to change their Pic/Icon to something representative of the campaign. If Barack has 100,000 “friends” in a group that’s great at generating some buzz, and could be indicative of a lot of support, but really means very little in terms of tangibles. However, if 20,000 of those friends all changed their pic/icon to one of Barack Obama - that is worth something because that change would reverberate through MySpace as a visible sign of support appearing on friend lists probably an order of magnitude greater in size. That’s the electronic equivalent of a maildrop or yard sign, and it wouldn’t surprise me if it actually carried some of the weight of a peer-to-peer contact.

More after the jump.

Micah Sifry wrote (among other things):

2. That gets to my second point. It could also be that we’re just looking where the light is brightest, and the real measure of online action and energy is going to be on the candidates’ sites and in the peripheral support sites that their backers create. I haven’t had the time (yet) to look at all the unofficial groups online for the candidates; nor have I looked at such basic building blocks like Yahoo groups, which were critical to the Dean campaign’s base. The Edwards campaign has built a very robust online platform for its supporters, and I think they could make a very good case that if they manage to attract and engage lots of people directly on their site, their lower MySpace total may matter much less.

I think this is right too. Student sites, OneCorps, MySpace and FaceBook Groups - support among young internet users going to coalesce in a lot of different spaces. It’s all niche now. Lack of support (or greater support) on any one of these sites isn’t a good indicator in and of itself.

Let me throw it open - what metrics do you see, and what tactics do you think are going to be rolled out on these platforms in 2008?

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

advertising metrics

Social Networkin sites are basically venues for promotion. That means advertising metrics are appropriate. In advertising, you measure the # of impressions made.

number of profile views + number of friends x bullitins posted

A tactic like the one you talk about w/ Obama’s pitcure would count as well. Measuring that would mean: number of people switching their photo to obama x # total cumulative # of friends (wow that’d be a lot!)

In my mind, social networking has always been about communicating to people in a more comfortable venue. Where as many people get angry at the number of email newsletters they receive, no one gets sore over the number of bullitins posted on their myspace.

Then there’s the other side: people like me who have a myspace, but never really check it out. I prefer email communications.

So the purpose is to hook your communications campaign into as many people as possible, using the communications venue they prefer. That means taking some time to integrate different online venues together- your myspace connects to your campaign profile, which connects to your you tube, which connects to your facebook. The more seamless the integration, the more natural it will feel, the more successful the tactic.

my .02 at least.

Conversions vs. Metrics

I guess I’m thinking of a different kind of metric. I’m not sure that advertising impressions is a good measurement, because at the end of the day, what is that really worth? Is it worth a vote? A donation? X Volunteer Hours?

Maybe metrics is the wrong term. I’m talking about conversions as my standard of measurement. We have X friends on FaceBook and X% of them switched their pic, or donated through ActBlue. I think that’s a much harder (concrete, not difficult) metric and more indicative of the health of a campaign.

So the purpose is to hook your communications campaign into as many people as possible, using the communications venue they prefer. That means taking some time to integrate different online venues together- your myspace connects to your campaign profile, which connects to your you tube, which connects to your facebook. The more seamless the integration, the more natural it will feel, the more successful the tactic.

Absolutely.

is this comparable to anything?

Obviously there isn’t a 1/1 correspondence between myspace adds and votes converted… but is this comparable to anything that existed before social networking? Not trying to be a dick, mike, but i think trying calculate conversions from myspace and facebook metrics may be an excercise in futility.

Although there was something I saw a while ago… one of the leaguers made a “I pledge to vote no on 75” form on his myspace. Maybe some kinda thing like that… “if you pledge your vote to obama, put his picture in your top 10 until election day”

You could even make that the avatar… a picture of obama photoshopped to say “obama’s got my vote.” As long as you made it clear that people were “pledging” their votes, then you could start to draw some conclusions. It still wouldn’t be a 1/1 correspondence, but it might be nice to give it a try. Like putting the little yellow ribbon on your bumper.

Votes aren't the only way to measure

Votes aren’t the only way to measure conversions. In fact, I don’t think that vote conversions from a social network are accurately measurable at all without exit polling. Those statistics won’t be available until after the election when CIRCLE and YVS do their exit polling analysis. I merely mentioned votes as a gesture towards arguments I’ve heard questioning what is a MySpace or FaceBook friends list is really worth.

This is weird because I’m trying to create/adopt an unfamiliar language to describe what I’m talking about.

If I’m an Obama staffer, and I wanted to get a better sense of what my friends list can do, and how many are actually engaged, I need a metric besides the size of the list. What I’m talking about is a way to measure that engagement.

If I photo shopped a pic (or asked my list to create a photo shopped pic in some sort of contents), and offered it for download to my list for use in a campaign as I described, then I could measure downloads. Not everyone who downloads the pic will post it, but its a much more accurate measurement of how we are doing among social networking users than list size alone.

The ActBlue idea is measurable on a one-to-one basis because you can see how many people donate.

If you use a widget like ChipIn on your and your members MySpace page, ChipIn tracks some metrics for you.

Add a bunch of these up, and throw in some traditional metrics like referrals to your main web page and email signups, and you’ll get a much fuller sense of what a FaceBook group/friends list is worth.

So I don’t think measuring conversions is at all an exercise in futility as long as you don’t define conversions solely as votes.

now i get you

that was the disconnect. When you talk about conversions, you’re talking about how many people you’ve actually turned on to greater involvement in your campaign through a social networking venue.

MFA has a few thousand myspace friends. Maybe we could try some tactics on them.

ohh… that sounds like fun.

Metrics

One of the interesting things that should be possible as politics moves into public social networking spaces is the creation of metrics based on network analysis. For instance, if you have a means for identifying your strongest supporters, and you can place them within a network map which is data-mined, you can do some weighting for supporter value, analyze clusters, spot holes, etc.

Another possibility with a campaign-operated activism site (OneCorps sort of qualifies, but i’m really thinking of GOPTeamLeader here) is very close tracking of volunteer activity. A sophisticated campaign could create a social environment for its highest-value activists, and through that interface provide tools to do outreach, rapid-response, online/offline canvassing, remote phone banking, etc. You could do a lot with tracking, and it would give you a lot of data at the top. On the other hand, you’d still be missing out on all the stuff people do outside of your official campaign toolbox, and providing the level of quality that would really get mass participation would be a pretty big challenge.

Recommended Reading - Network Analysis

Got any recommended reading on that subject?

Good Questions

For the pop intro, I like Linked: The New Science of Social Networks. For practical stuff, I’ve been bugging my friend Luke who’s getting a PhD in sociology and doing his thesis research by building a dataset of business mergers around the world (and the people involved) over the past 30 years, then feeding that into his network analysis programs (I think mainly packs for SPSS).

There’s a number of decent software products out there that can do useful analysis. I saw some stuff back in January 2006 at BarCampNYC where kids had spidered a few hundred thousand myspace profiles and built that into a dataset. I think it was built on a FOAF visualizer. We had one of those for DealLINK back in the day too. It’s sort of dried up as a technlogy, but there’s still stuff out there.

Back on the academic tip, you might try The International Network for Social Network Analysis.

But basically you want something that’s going to map connections, look for clusters, overlay data about geography, interest, activities taken, etc.

Linked

Thanks for the reminder. I’ve actually read that book (three years ago) and still have it. I remember that there was a bunch of math I kinda glossed over as I was just getting into the theory at that point.

Maybe that’s a good book to revisit for a quick primer.

Meh

It’s not totally practical, but the bits on power-law expansion are good. He doesn’t really talk much about network analysis, more about the math that seems to govern network growth and behavior. This stuff is good to know too as it can help you target expansion, etc.