Blackbeaks Blog….All things Analytics

Questions I’ve had about the measurement of the effects of social media on your website like blogs, frappr, flickr, myspace, facebook, LinkedIn, 2nd Life and a bunch of other Web 2.0 stuff has got me thinking recently.

My first question is “Why does it worry you if you have good KPIs?”

When you use KPI’s to drive actions on your website social media becomes irrelevant. Let me illustrate with a scenario.

Scenario

You run a website selling products and are measuring the reach factors, the various referrers that visit your site shows that some social media is more effective than others at driving sales. Cool. This same KPI also compares the other reach factors you operate therefore doing anything special is not necassary. Let’s say for instance we find that 10 sales were drven by something going on in 2nd Life, 10 by blogs and 10 sales were driven by your Google campaign. Your KPI should tell you which one was more cost effective and where to focus your spending (social media stuff always has a related cost too, even if it is only in time you spend on it).

So what’s the problem?

Ahh I hear you say, but what about where to focus my social media efforts? How can I find out if my prospects trust me? How do I know how to improve my offers based on what people are saying about our products? Or can I take action on my website because people in social media are saying “x” about us? Now we’re talking about something web analytics can’t help you with. We’re talking about qualitative data that should be combined with the quantitative stuff that web analytics gives you.

So my second question is “Have you effectively built a process and have tools which help you combine the data?”

Scenario 2

You run a website selling products and are measuring the reach factors, the various referrers that visit your site shows that some social media is more effective than others at driving sales. Cool. This same KPI also compares the other reach factors you operate therefore doing anything special is not necassary. Sound familar? Ok this is where it gets different. 2nd Life sent you 10 sales, Google sent you 10 sales and a referrer from some blogs sent you 10 sales. The qualitative data in these 3 reach factors are, 2nd life, none, basically you have no idea why they bought unless you set-up some kind of 2nd life survey (maybe from the 10 customers for instance). 2 Google - You have the keywords they used tto find you and again you can ask the same survey questions of those customers. 3 Blogs - You have the blog posts they made talking about your product and the survey data asking them why.

The trick is to look for qualitative data that allows you to make decisions. You might have 10,000 visitors (quantitative data) that come from blogs and 500 blog posts about your product. Remember that theoretically from these blogs you had 10 sales. If from the 500 posts people were derogatory about your product then take the information they give you and make changes based on those comments. 10, from 10000 is poor any way you look at it.

If however the 500 posts were good news, then you would expect that sample of happy posters to represent a high level of satisfaction meaning that the 10,000 visitors have not had all of their questions answered on your website about your product. This is a clear indication you need to go back and fix your website so that you are answering the needs of your visitors.

Social media measurment tools you could use;

BlogScope Is an analysis and visualization tool for blogosphere which is being developed as a research prototype at the University of Toronto. It is currently tracking over 12 million blogs with over 100 million posts. For instance I did a search about web analytics and then wanted to see where people were discussing about it the most, this was the result. There is of course much more to it than this.

Xtract - Xtract Social Links¢€ž¢ turns raw data into a marketing tool. By targeting the well connected influencial people in their communities this tool gives you the qualitative data you need to make decisions on who can help you with your branding.

Buzzcapture - Buzzcapture¢‚¬„¢s proprietary technology retrieves data that are relevant to you from Consumer Generated Media and filters the data through its linguistic analysis technology. Basically that means it trawls the Internet for conversations about your brand and gives you the data in digestible form.

Onalytica - By analysing what has been published online about a particular topic, as well as who is paying attention to it, they extract valuable qualitative knowledge that delivers insights.

TNS Cymfony - Our friends at TNS Gallup recently bought Cymfony. In their words; Cymfony, a division of TNS Media Intelligence, is a market influence analytics company that sifts and interprets the millions of voices at the intersection of traditional and social media such as blogs and social networks to gain consumer insight and develop stronger bonds with influencers.

NeilsenBuzzmetrics - I can’t mention TNS without mentioning Neilsen! In their words; Nielsen BuzzMetrics is the global measurement standard in Consumer-Generated Media. They are apparently the technology behind BlogPulse another free tool which shows below the percentage of all blog posts they have measured when you give them a keyword;

Interesting peak there around the beginning of May. Could this be when the debate around Web analytics blogs started when every man and his dog was discussing who was top blog? Opinions welcome.