Understanding Facebook Link Shim Referrals

Facebook Link Shim Referralsview at full size

The Referrals report displays multiple values for Facebook.

If you filter your Referrals report for facebook, the “l” (that’s l as in lemon) referrals were directed through Link Shim, which Facebook uses to protect clickers who may be unknowingly accessing malicious sites. If your site is legitimate, Link Shim is no problem, but the referrer will still appear as the l version.

This apparently represents an improvement for analytics, because the current Link Shim versions previously registered in Google Analytics without accurate source attribution, so you could not even tell that the visits originated from Facebook.

To sum up the four combinations of desktop/smartphone and Link Shim:

  • facebook.com – desktop or tablet, link not directed through Link Shim
  • l.facebook.com – desktop or tablet, link directed through Link Shim
  • m.facebook.com – smartphone, link not directed through Link Shim
  • lm.facebook.com – smartphone, link directed through Link Shim

For more information on Link Shim referrals in Google Analytics, check out this great post by Jules White.

Please share this post:
Facebook Twitter Plusone Linkedin Email

Google Analytics “Hit”: Definition and Limits

Although pageviews are central to Google Analytics, the calculation of interaction metrics and limits in Google Analytics is based on a broader set of interactions.

A Google Analytics hit is defined as a pageview, a virtual pageview, an event, a social action, or an Ecommerce transaction. Unlike regular pageviews, the other types of interactions are not generated by the default Google Analytics tracking code but instead require special coding.

For calculating time metrics and bounce, Google Analytics treats all hits equally. For this reason, if a virtual pageview (from a PDF download, for example) or an event (from a video play) occurs during a single-page visit, that visit is not counted as a bounce (unless, in the case of the event, you have coded the event as non-interaction).

Google Analytics bases interaction limits on hits and not exclusively on pageviews. Google Analytics can capture up to 500 hits per visit and 10 million hits per month. (The monthly hit limit for Google Analytics Premium increases to 1 billion.)

Note that hit in Google Analytics differs from the general hit of Web server parlance, which instead refers to a request for any file from the Web server, such as HTML, CSS, JavaScript, or an image. (It also differs from the layperson’s usage of hit as a visit to your website.)

Although hit does not appear anywhere in the Google Analytics interface, it’s an important concept for Google Analytics practitioners.

Please share this post:
Facebook Twitter Plusone Linkedin Email

Add Campaign Parameters Before Shortening URLs

Add campaign parameters to a URL before shorteningview at full size

You can add campaign parameters to URLs before shortening for social posts.

As with any inbound campaign link, you can track the performance of your social campaigns by tagging any associated inbound link with Google Analytics campaign tracking parameters.

Even if you already track clickthroughs directly through the analytics within shortening services such as goo.gl and bit.ly, you’ll gain insight into metrics such a goal completions within Google Analytics if you append the campaign parameters to your URLs, as in the example below, before shortening.

http://www.mysite.com/?utm_source=twitter&utm_medium=social&utm_campaign=springpromo

As with other instances of campaign parameters, it’s recommended to use the Google URL builder just to be certain of correct configuration.

Please share this post:
Facebook Twitter Plusone Linkedin Email

12 Google Analytics Resolutions for 2013

 

1) Set up goals for all form submissions.

Goals aren’t retroactive, so make sure to set them up for all form submissions as early as possible in your Google Analytics implementation. Without conversions, there are no conversion rates.

 

2) Learn the ins and outs of funnels.

Once you have a goal, it’s very useful in many instances to build a funnel on top of it so Google Analytics can calculate metrics such as Abandonment Rate, but to use funnels most effectively, you should understand a few of their idiosyncrasies and limitations.

 

3) Don’t ignore Page Value.

The unassuming and underused Page Value metric (formerly $index) is indispensable in determining which pages are pulling their weight towards goal conversions and e-commerce transactions. (For this reason, make sure to assign values even to non-monetized goals, in a separate profile if you don’t want to skew actual e-commerce and goal monetization.)

 

4) Tag all inbound links with campaign parameters.

How effective is your monthly email newsletter at driving visits and conversions? How about your press releases? There’s no excuse to be in the dark on these basic questions – tag those links!

 

5) Track PDFs views and external links with Virtual Pageviews or Events.

It’s easy to forget that views of PDFs (and most other document formats) and clickthroughs to external sites are not captured with the default, pageview-based Google Analytics Tracking Code. Virtual Pageviews and Events provide an easy solution.

 

6) Avoid the built-in “Non-paid Search Traffic” Advanced Segment.

Lumping branded organic traffic and non-branded organic traffic is, for most purposes, a very bad thing. Define your own non-branded organic Advanced Segment. (And define other Advanced Segments of your own as well – better segmentation generates better insights.)

 

7) Track different social actions differently.

A Facebook Like that originates from your website can mean two very different things. Understand how social actions differ and track them appropriately.

 

8) Keep a chronology with Annotations.

When did the new home page go live? When was our company mentioned in the Wall Street Journal? You can’t make correlations between your Web data and relevant events if you don’t remember when those events happened. Maintain a comprehensive timeline with the convenient Google Analytics Annotations feature.

 

9) Incorporate qualitative evaluations into your conversion optimization plan.

An in-person or online user test can quickly reveal a conversion pitfall that could otherwise lurk in your Google Analytics reports indefinitely.

 

10) Run Content Experiments regularly.

Nothing like a head-to-head contest to tell you quite definitively what works better for conversions. If you haven’t yet run a Content Experiment or used a third-party A/B testing tool, start now.

 

11) Establish core KPIs (key performance indicators), and check them at least monthly.

As Brian Clifton says, Web analytics is like going to the gym: if you don’t do it regularly enough, you’re not going to be satisfied with the results. Consistent scrutiny of meaningful KPIs will lead to useful and actionable insights across the board.

 

12) Use Those Same KPIs for Daily or Weekly Custom Email Alerts

Don’t wait unnecessarily to find out that visits or conversions have dropped by 20%. Take advantage of Custom Intelligence Events to actively monitor your essential metrics and receive email alerts for substantial deviations.

 

Best wishes for a customer-focused, data-driven, conversion-optimizing new year!

Please share this post:
Facebook Twitter Plusone Linkedin Email

A Like Is Not a Like

view at full size

Using more specific action values when you call _trackSocial makes the Social Plugins report more meaningful.

You’ve undoubtedly heard jubilant exclamations along the lines of “We got 10 likes on Facebook today!”, but what specifically does that mean?

While any type of social media validation is a good thing, it’s important to recognize the two fundamentally different types of social actions that can originate on your website:

  • liking, tweeting about, sharing, or +1ing the content on your website
  • liking, following, +1ing/circling your organization’s pages that reside directly on the social networks

To distinguish between these two different types of actions when calling _trackSocial in Google Analytics, you can use separate values for the “action” parameter, as in the examples below for Facebook:

  • _gaq.push(['_trackSocial', 'Facebook', 'Like']
    (for directly liking your website content)
  • _gaq.push(['_trackSocial', 'Facebook', 'LikeFacebookPage']
    (for clickthroughs to your Facebook page, which users can then like once they’re there)

There is a less confusion on this point for Twitter, since the “tweet” and “follow” concepts are quite distinct. The same is true for “share” and “follow” on LinkedIn.

The equation is, on the other hand, even a little more complex for Google+. In addition to the +1 for your website content and the +1 that a user can perform after clicking through to your organization’s Google+ page, there is the Add to Circle action that a user can also perform after clicking through to your Google+ page. For this reason, it’s helpful to use a broader social action value such as the following to record clickthroughs to your Google+ page:

  • _gaq.push(['_trackSocial', 'Google', 'ActionOnGooglePlusPage']
    (for directly liking your website content)

If you use sufficiently specific values for your social actions when calling _trackSocial as above, your Social Plugins report will provide a much more accurate representation of the social activities that originate from your website so you know specifically which social media victories you can celebrate.

Please share this post:
Facebook Twitter Plusone Linkedin Email