Showing posts with label webtrends. Show all posts
Showing posts with label webtrends. Show all posts

Thursday, August 1, 2013

Tracking individual users in Google Analytics

Several people have recently asked me if it is possible to use Google Analytics to track and store information on specific individual visitors to their site. Usually the popular analytics package only reports trends and grouped user behaviour, you never get to see the granular detail of each person (e.g. their specific browsing path around the site, etc.). However this can be quite annoying for some marketers who want this information and who have used competitive packages from companies such as WebTrends or Adobe (Omniture) in the past.

If you read the Google Analytics support documentation on Google’s website, you will see in several places there is a reference to NOT using PII (Personally Identifiable Information). This is data that can be used by Google to identify an individual and includes info such as: Name, Address & Email address. However PII can also be a mobile phone’s unique identifier or some other way to recognise a specific device.

There are work-arounds to this restriction (such as using Custom Variables that hold randomly-generated reference for each specific user), but these come very close to violating Google’s End User License Agreement and are definitely not in the spirit of the platform.

So can you upgrade to Google Analytics Premium (the paid-for version of GA) and then start to store important user data? No. The collection of personally identifiable information (PII) is in violation of Google Analytics entire EULA and therefore paying $150k still doesn't let you use the platform as you might have hoped.
Google has therefore been pretty specific in its user agreement (with its new Universal Analytics product also currently having the same restrictions) and even gives the warning that:
Your Google Analytics account could be terminated and your data destroyed if you use any of this information.
However, I have one important point to raise that has been bugging me…

In GA there is the feature to understand your Multi-Channel Funnels. This is lets a site manager understand the interactions between different online media and see how the channels work together to trigger sales. Since this report gives a breakdown of all the multiple digital customer touch-points over the last 30 days… if Google doesn't uniquely identify individuals, how does it know when specific people use each channel and then join them up to create a complete picture of the steps customers take before actually converting?

Thursday, November 8, 2012

How much value is your website data giving you?

I'm on a mission to make sure companies get more value from the online data they collect. Why?
Well most companies now know that they can plug analytics into their website, ranging from the free (but still incredibly useful) Google Analytics, to paid for products such as: Coremetrics, Adobe [yes, we do still call it Omniture] and WebTrends.

For me the diagram below highlights the increasing value of website analytics and their importance to the business.
Reading the diagram from the bottom up, you firstly move on from a position of no data (or badly collected data) by setting up correct data capture processes. This can be anything as simple as collecting log files from your web server of visitor traffic. However by taking the next step (e.g. implementing website analytics) an organisation makes the move to having information it can use, now typically presented in the form of tables, diagrams, etc.

But, that is typically where most companies stop. However there's an additional layer above that, delivered by carrying out data analysis on the available information and creating valuable online insight.

Monday, October 22, 2012

Where have all the website analysts gone?

All to often I speak with clients who have great websites and have therefore implemented analytics, such as those by Google, Webtrends, Omniture / Adobe, etc. However a lot of them also admit that they are not getting as much from these analytics packages as they should be.

They often claim that they either don't have the skills in-house or an aligned digital agency who can give them timely and relevant insight. This is a real shame, as these organisations usually understand that their business could be that much better if they could just gain better explanation of what their customers are doing (or not doing).

To back this up, a recent posting from eConsultancy highlighted an important and worrying trend..... website analysts are disappearing.

If you look at the graph above, you will see that in the recent survey, for a question asking "How many dedicated employees does your organisation have doing analysis of web data?" the number of organisations with zero staff responsible for web analytics has actually increased.

Increased? Surely it doesn't make sense that, in an increasingly digitally-savvy business environment, companies are downsizing their online analytics capability? It would seem so, and this is further backed up by the other data in this survey.... that clearly shows that most companies (i.e. those with up to 3 people responsible for online insight) have been reducing their focus on this important function in the last year. Perhaps unsurprisingly, only those large companies with five of more dedicated website analysts have increased their numbers.

It seems that the either role of website analyst is disappearing, or the smart staff required to fill these roles just aren't around any more. Where have all these people gone?

Monday, September 27, 2010

Measuring Social Media activity across your business

If you work for a larger company with multiple divisions (and marketing initiatives) then there is a good chance that you have more than one piece of social media activity going on at once.

There's also a pretty good chance that these different initiatives are not connected and may even be duplicating effort (or worse, working against each other).

So what do you do?

1. Audit all activity
Firstly, do an audit of all activity involving social media across your company.  Create a list of everything from those business Twitter accounts in the USA office, through to those Facebook Pages created by the German Marketing team.

2. Measure reach, impact and engagement
How? Well, for reach, its a numbers game. With services such as Twitter its easy to see the number of followers an account has and its just as easy to record the number of Facebook 'likes'. Counting up the number of people who have 'like'd each clip* or are following your channels on YouTube is obvious. Also most blogging platform now either provide statistics for the number of visitors to your blogs or allow you to integrate Google Analytics or other stats package into each page.
*YouTube also allows users to give each clip a 'thumbs down' plus has basic stats built into it now

However measuring impact and engagement is a different matter and I would therefore suggest you look at these methods as a starter:
  • Twitter: 
    • Look at the number of re-tweets each account gets, this will give you some idea of how valuable the followers regard the postings.
    • Measure clicks on hyperlinks. You can do this either by looking at your referrer information in your site analytics package (assuming the links point back to your own site), or if you use a service like Bit.ly which integrates with Twitter, this will show you the clicks on each Tweet.
      Hayden's tip: Use Google link search to see how many people hyperlink to your Twitter account e.g. http://goo.gl/6hkt

  •  Facebook

    • Look for the amount of comments and 'likes' on postings. Obviously the more followers you have, the greater the likelihood of having comments and 'likes'...
      Hayden's tip: To get a real measure, work out your average number of responses per posting and divide this by your total number of followers.. and then measure this over time.
  • YouTube, Flickr, etc.
    • Both these sites allow commenting on each respective file uploaded and YouTube even allows video responses to be posted. Reviewing this feedback over time will give you some idea of the content that people find interesting (in a positive or negative way - so remember to )
      Hayden's tip:
  • Blogs
    • Comments are the obvious way of measuring engagement with your blog content and because they are easily spider-able by search engines and social tools, you can use a variety of methods to understand the quality of the people contributing to them (e.g. http://socialmention.com/)
      You can also use your analytics package to measure more factors about engagement such as the amount of time spent on each page (paid for analytics applications may even have a specific way of working this out, such as this method from WebTrends)
      Hayden's tip: Again use Google to measure the number of inbound links to each blog: e.g. http://goo.gl/tjTn
There's obviously lots of other social networks to assess and the ones you audit and measure will depend upon the countries your business is active in. But without first taking stock of your entire social media presence, there's no way you decide which efforts are working and which are not.