Analytics Market Chart

Google Analytics and Social Media Tracking

Many business are tapping into the value of social media. It operates unlike any other type of online marketing, and the rules are constantly shifting. With this shift in focus, web analytics vendors have started to announce partnerships or tools to integrate social media tracking into their product.

Companies using web analytics have to decide how they will track and measure their social media efforts alongside their online marketing and website traffic. How can social media be tracked in Google Analytics? How can a company compare their social media against their banner ads and cpc?

The short answer is, they shouldn't.

There are (at least) three reasons social media can't be tracked accurately in web analytics tools. Some kind of integration isn't intrinsically bad, but making web analytics software a one-stop source for measuring everything online is a mistake.

1. What is "Success" in Social Media?

There is no consensus yet on what a "successful" social media campaign looks like or what it achieves. In the loosest sense, companies engage in social media to increase their influence and reach. That is an ethereal objective, and it can only be meaningful by measuring its "symptoms": its tendency to increase sales, decrease complaints, increase customer satisfaction, increase referral rates, etc. How do you accurately measure the social media "symptoms"? Sales influenced by social media often can't be tied directly back to a specific post (or source).

There is also no consensus on the best approach to social media. Should you work to increase the number of followers? Should you try to increase your retweet/tweet ratio? Should you entice more fans to say they like your Facebook post? Should you aim for a higher click-through percentage for your links?

Do you do this by being a person, assuming a fictional character or presenting a formal corporate face? Some organizations feel that they ought to follow influential people. Conan O'Brien has almost 1 million followers, but he only follows one person.

Social media is still so new that there are no firm rules for how to do it. Therefore, there are no predefined metrics for a tool to calculate. When a vendor proposes specific metrics to evaluate "success" for social media, it is almost an arbitrary decision.

2. Measuring Social Media Gives False Comfort

Integrating social media into more conventional web reports implies a false correlation between them. Tweets can't be compared to email campaigns, for example. If an email campaign brings more visits than a Twitter campaign, it doesn't mean the email campaign "did better". Social media is an entirely different animal from email, cpc and banner ads. There is no direct comparison to be made between them. (For that matter, there is often no comparison to be made between different types of social media.) They each need to be evaluated intelligently for what they are, not automatically by predefined filters.

Putting everything into one place for analysis implies that it is a comprehensive measurement of everything relevant. However, web analytics software is only one piece of the puzzle. To really evaluate a company's online presence, an analyst needs to consult click-path tools, social media, competitive intelligence, voice of customer, test results, etc. Although each of these categories has some influence on the other, it is nonsense to suggest that they can be meaningfully integrated into one place. Unless, of course, "integration" means putting them all into separate reports.

3. Integrating Social Media is an Inflexible Model

Social media morphs quickly (and it should!). Evolutions in what products are used and how are user-driven. Companies will always be a step behind users in determining how to use a medium effectively. This is the essence of social media. With this in mind, it seems ludicrous that a web analytics vendor would attempt to hardwire integrations for every current social media platform into their product. Where would you be today if you used a tool with a built-in MySpace integration? How long will Facebook last? What will replace Twitter? Who will come up with the next bookmarking service? What will advancing mobile devices do to the industry?


None of this is to suggest that integrations wouldn't be convenient. A good integration may save some time for the analyst. But it won't obviate the need to use other tools and methods for evaluation.

In the end, choosing a web analytics product based on its social media integrations may be a red herring. The foreseeable future will require skilled analysts to work with multiple products to get a holistic picture of what's happening and what needs to be changed. And as the terrain becomes more complex, the required skill level of the analysts will likewise rise.