How can I track traffic from social networks?

If I post it will they click it? And if they do, will I be able to count them? What post or topic did they like the most? Shall we focus on one network or another? Raise your hand if you have not been asking yourselves these very questions while trying to figure out how to efficiently keep on promoting your content on social media.

This tutorial is based on the assumption that you want to collect some useful data on how much traffic your site or blog receives from social networks but you do not have access to resources and tools for sophisticated analysis.

There are many possible ways of measuring performance on social media. Metrics should be selected on the basis of what is really important for your communication objectives. Therefore, it is impossible to be exhaustive in one tutorial. The three options proposed here can be useful starting points in collecting and analysing Web site traffic data generated from those networks that are not consistently captured as referrers in traffic analysis software. The goal is to be able to disaggregate the visits generated by clicks on links promoted on social networks such as Twitter and Yammer, so as to optimise online promotion.

Why disaggregate traffic from social networks?

People can access social networks such as Twitter, Facebook and Yammer via a variety of tools, not just a browser: mobile applications, desktop applications and email notifications are alternative ways of reading and participating in a network. Clicks on links from one of these applications are not captured as referrers in traffic analysis programs: the visits they generate get bundled together as “direct traffic”. For example, if I click on a link from a person I follow in Twitter from the browser, the twitter.com domain will be captured as a traffic source for that site. But if I click on the same link in my TweetDeck desktop application, the visit I generate will be captured as direct traffic, as if I have typed in the URL manually in the browser.

If we do not capture these visits in traffic statistics, we lose track of interesting data on the benefit we expect from promoting our content on social networks.

How to track traffic from social networks

To see some ‘clean’ numbers for people who followed a link to our latest post on a specific social network, we want to focus on the URLs that we post to the networks, so that the visits they generate are captured by our traffic analysis software. We have three options: each one implies varying levels of manual work and data accuracy.

Option 1. Use a URL shortener to count clicks

Shorten the link with a URL shortener with click-counting capability, e.g. bit.ly, and post them to your favourite social networks. In your account, you get a click count, and aggregated information from other bit.ly URLs for the same link, mentions in Twitter, shares and comments in Facebook.

Alternatively, you can post your links using Hootsuite, a popular social media dashboard that lets you post to several networks, shorten URLs and analyse different metrics straight from the dashboard. Note that not all the latest analytical reports come free.

Pros:

• Low level of manual work required.
• You can get an idea of how much traffic comes to your site via specific links, and how much interest a certain topic has raised.

Cons:

• Counts may not be reliable and include so-called ‘decodes’ i.e. hits by bots and search engines, not necessarily intentional clicks by humans.

Option 2. Add tracking code to URLs, shorten, post and analyse in Google Analytics

This option has the advantage of being more accurate than Option 1: if you use Google Analytics (GA) to analyse your Web site traffic, the link you will post will include parameters that GA can track as sources and that will return a disaggregated number of visits generated by clicks on a specific URL. This method is also called URL tagging: there are hundreds of tutorials online. For introductory information, you could check this tutorial which I found pretty good (you can ignore the title,and jump straight to section Tagging 101).

The procedure consists of three steps:

1. adding Google Analytics tracking code to the URL,
2. shortening it with a URL shortener and
3. posting it to your favourite social networks.

For example, I want to post the URL of a recent blog post to my network on Yammer. I want to be able to see the number of visits to our site that have been generated by that specific post.

The first step is to go to the Google Analytics URL Builder, enter the URL of the post, and the following tracking parameters:

• Source: yammer
• Medium: post (if I was coding URLs for an email newsletter, I could put ‘email’ here)
• Name: tutorial (a naming convention for the type of post)

Parameters in Google Analytics URL Builder

I click on Generate URL and get this:

http://ictkm.cgiar.org/tutorials/how-can-i-write-a-proposal-document-collaboratively/?utm_source=yammer&utm_medium=post&utm_campaign=tutorial

Then, I shorten it and post the shortened link to Yammer.

GA will show the clicks on that URL as visits from source: ‘yammer’ and  campaign: ‘tutorial’:

Campaign data in Google Analytics

Pros:

• More accurate traffic measurement for individual content items promoted on specific networks.

Cons:

• Labour-intensive: you need to tag each and every URL for each and every network.
• Requires planning and consistent use of the naming conventions for sources, campaigns and mediums.
• It’s never going to be comprehensive of all traffic generated on a specific network, simply because people may have found your content through other channels and used their own shorteners of choice before sharing the URL.

Option 3 – Use an automatic routing service with tracking capabilities

One way of having the tracking code added automatically to your URLs is to use a ‘routing’ service like dlvr.it. You open an account and the service will take the titles and URLs of new posts from the RSS feed, shorten them and post them to your favourite social network account.

Dlvr.it can attach GA tracking code to the URL it sends out. If you maintain the default settings, traffic from dlvr.it URLs appears as source: ‘dlvr.it’ in GA. All parameters can be changed based on your naming conventions.

Tracking code parameters in dlvr.it

Pros:

• Less manual work is required, since the key content for the message is sourced from the RSS feed, and the tracking code is added automatically to the shortened URL.

Cons:

• A downside of relying only on a routing service is the limited freedom of wording the message: posting a blog post title as it appears in the feed, with no tag and no personalised message, may result in some sort of “dry” way of communicating. In practice, if you get the URLs posted by such a service, it would be advisable to retweet or repost after a while.

Conclusions

Adding tracking code to URLs is a tried and tested method that provides accurate measures of traffic generated from specific social networks. However, developing naming conventions, preparing URLs and posting them manually to different networks entails a workload that may be hardly sustainable for small teams without resources dedicated to promotion.

In such resource-poor scenarios, URL tagging could be reserved to priority campaigns to promote an event or a featured publication. The information that tracking code can collect can be very useful for refining promotional efforts.

The difficult balance between the cost of data collection and the benefit of accurate analysis affects the Web community as well: Web site traffic is one metric for promotion, but there are more ways to monitor and evaluate social media activities. What metrics do you track to show usage and relevance of your information?

I would love to hear your take on this.

Photo credit: pie chart color by svilen001