I remember it clearly, even though it was 22 years ago. It was Hannukah, 1985. There was one album I wanted more than anything: Starship's Knee Deep In The Hoopla. Lucky for me, my mom trekked down to Cal-dor and bought it for me.
Less than a year later, by late '86, I considered this album to be among the worst rock albums of all time (at the tender age of 9, no less!). How quickly my tastes changed, huh? I asked for Bon Jovi's Slippery When Wet for Hannukah '86, and the rest, as they say, is history.
It just goes to show you that what may be accepted as totally rad one minute can often be seen as totally bullshit and useless about six months later. I think we're at that point in analytics and assessment metrics, when it comes to social media.
There's been so much talk lately in the blogs about "What are reliable assessment metrics?", especially when your brand is combining web strategy with social media strategy. I saw a pretty solid web analystics talk on Owyang's blog last week and it really got me thinking. People are trying to measure social media in this crazy 2002 mindset. Well, that'll work all right, assuming the last new album you bought was All That You Can't Leave Behind.
If your brand strategy has changed since '02, then we have to do a little breakdown right now. I wrote a post on Return On Participation the other day; that's one social media assessment metric. That post was a jumping-off point, but not even the be-all-end-all. The idea of this post is to separate out the metrics that aren't really meaningful in social media, versus the ones that are.
So, here's the 2002 toolset - these were the five types of assessment questions that were rockin' the house alongside QOTSA's Songs For The Deaf and Beck's Sea Change. What I'm saying is, these were classics in their time, but things have evolved a bit since then.
- What are the most common paths to the most important pages on our brand's website?
- What keywords bring the most traffic to our brand's website from search engines?
- How many new users visit our site every month?
- How long are people spending on our website?
- How many visitors do we get from New York every month?
So, these old web metric questions (taken from a data mining book that actually came out this year) strive to confine the web experience by a few really narrow parameters:
- Time on site - How long do people stay?
- Common paths - How do they usually get here?
- Total traffic - How many people are here?
- New users per month - Are we still getting a lot of new people?
- Geographic distribution of Visitors - Where are people from?
Nowhere in these parameters is it even alluded to that the content on this site will be (1) delivering value to users (2) allowing them to react with the content on-site or (3) allowing them to engage with the content and do things with it - remix it, mash it, whatever they want to do with it. Oh, and here's the rub: most of this engagement isn't taking place on companies' websites - it's taking place on content websites, blogs, shopping websites, social networks and places that brands have little direct control and a high signal-to-noise ratio to bear.
An aside: you know, one of the reasons that I became such a huge fan of the Clash (about three and a half years after I got that crappy Starship record) is because they went from being an amazing punk band to being an amazing alt rock/funk band. And the reason that they did stuff like this is because they were brave enough to take things they loved, like hip-hop records, and turn them into things that no one would have imagined otherwise. Where do you think "The Magnificent Seven" came from? So here's the deal:
If your brand is making content that users can't engage with, then assess that interaction with the 2002 metrics.
If your brand is even trying to engage in social media, then you've got to change your assessment metrics, now.
Before we dive into the discrete metrics, let's talk about a new assessment philosophy. Every social media initiative involves working in a few different tools. Whether your brand is using wikis, podcasts, vodcasts, blogs, widgets or social networks, you've got to think of social media assessment metrics as a mesh or a grid. You're evaluating based on these questions, which are intertwined with one another:
- How engaged are people with each of these tools? Do they come back a second time to use any of these tools?
- Do people agree that the content that our brand is delivering is content that delivers utility to them?
- Are conversations started as a result of these tools? Do we have proof that these conversations are started? Can we follow them?
If you're expecting a 2002-style "conversion" based on a user using one of these tools (e.g. joining the WalMart Facebook Group), you're dead wrong. No one is going to run out and buy a six-pack of Busch Lite because they saw that your brand started a social network. Even most Busch Lite drinkers aren't that stupid.
But, the next time they're at the liquor store, deciding between Busch Lite and the beer that doesn't let them engage with their product, they might just pick Busch Lite. Maybe. It depends on how well you engaged them.
Tomorrow, we get into the discrete metrics on a per-tool basis, that will allow you to answer these three critical questions, so you can connect social media assessment to goals that create business value for your brand.
-----
Part two of this post will go up on Tuesday or Wednesday and then we'll be having some guest authors while I'm down in Mexico for the week. Any videos mentioned on MetzMash, from here on out, will be archived on the MetzMash YouTube widget, down there on the right-hand side of the blog. Also, big thanks to the Oakland Public Library's Terrence Egan and the Netlibrary in the creation of this post. While it's cool to get the first chapter of a book from Amazon, it's a helluva lot cooler to get the whole damn thing with your tax dollars from the Netlibrary.
Link to original post