I've been blogging a lot about Social Media for the past few days, so I figured I'd better move back a little closer to my original premise with an analytics post. I've spent the past week participating in the Inbound Marketing University for the past week and it had been very helpful. I fully embrace some of the core principles of encouraging others to find your business through less invasive tactics like Search, Blogging, and Social Media. People are less and less anxious to be "sold to" these days and inbound marketing both recognizes that and systemizes the efforts and measurement of interest generating activities.
So what about the measurement of Social Media? That was the unfortunate topic Marshall Sponder tried to tackle today. I know Marshall better than he knows me, but I also know that he's been mired in the minutiae for several years. He uses lots of tools- most are free. I certainly understand the appeal of data. I too am a data junkie, but I heard all sorts of advice about learning things from the numbers and not too much about what to DO with any of it.
So that leads me to some ideas for you.
1. Find your Key Performance Indicators. This applies to "on site" analytics as well as "source" analytics. Time on site is an old standby. It's a good bench mark. It's not perfect, but it's worth tracking over time.
2. Watch your bounce rate. I think I heard someone say today the sometimes bounce rate doesn't matter. I still can't think of any reason why a high bounce rate is desirable. I will say that there really is no such thing as an industry benchmark. You'll need to track your own bounce rate and measure it over time.
3. Measure Social separately. Regardless what tools you have access to, there is likely to be some method of segmenting groups of inbound links, called referers. (BTW, referer is purposely misspelled. It's kind of a geek thing based on the fact that in log files, it's misspelled and true industry nerds know that. Admittedly, that may only be 12 people on the planet, but I digress... ) By segmenting into groups, you can build a social segment.
Now check all three metrics Social VERSUS the non social traffic. There you have it. The basics of social analytics. It doesn't make any sense to just start look at numbers if you have nothing to compare it to.
OK so now how to get actionable data. Here's the deal- we know the numbers will be different. But if your KPIs are "lower" in your social segment, it doesn't mean you should abandon your strategy. After all, time on site from paid search is often shorter than organic search- it doesn't mean you should abandon paid search. It means you need to learn more. Oh no! are we gonna get complicated? Not really.
Now it's time to consider some new KPI's. Be careful not to violate the cardinal rule of analytics. It can be very dangerous to "add new KPI's" or to try to get the data to tell the story you want told. Don't do that. Just think a little more about what you would expect a social visitor to do on your site. Consider where they are entering your site, what they are doing and how long they stay. If your site has many goals, which many sites do, it's possible that your social visitors are providing more value towards some of your goals and less for others.
WARNING- This may get complicated. Consider this bonus content.
Let's say that your time on site is longer for social visitors, but they don't seem to be converting very well. Perhaps you need to consider that you're customer life cycle is different for Social Media. If you don't already have a latent conversion tracking method, you might need one now. Check out the concept of closed loop analytics- the idea is to follow the whole customer cycle start to finish- including off site metrics in some cases. The other related idea is attribution management. Their exactly the same, just different. Clear as mud right?
If you're confused, I'll offer you the same advice David Spade gave (the late) Chris Farley in Tommy Boy. "Take tylenol for any headaches, Midol for any cramps..."
No comments:
Post a Comment