Subscribe to Dr. Granville's Weekly Digest

Bit.ly is an URL shortener (like tinyurl.com), but it offers much more than redirect services (everybody can create a redirect service), thanks to its interesting reporting capabilities. They recently hired a famous data scientist, Hilary Mason, see picture below.

hilary mason

And this service it is an example of great big data - provided you are aware of their incredibly simple - yet incredibly powerful dashboard (I love simple dashboards). It is also a classical example of big data that is useless unless you are aware of the features coming with it, and how to leverage it. So here's what I found to be especially useful. Below is a screenshot of the dashboard in question.

You can find the amount of traffic (page views) from your own bit.ly redirects in the "CLICKS VIA YOUR SHORTLINK" column. Now the same link might have been shortened by other people. That's why there is an extra column labeled "TOTAL CLICKS" that tells you total traffic coming to the target URL, both from you and from other people who also used bit.ly to shorten the same URL.

Assess traffic from competition

This is very useful for publishers or media companies. For the goPivotal link we created at 5:15 pm, 155 clicks are from us, and another 114 from competitors. So we delivered better than competitors on this one. Of course not everyone uses bit.ly in their marketing campaigns, but many do because it offers traffic statistics (by referrer, country and day) to the client. Simply add a + sign at the shortened URL: https://bit.ly/11xTDUD+. Et voila!

Note that the goPivotal link created at 7:22 pm contains special proprietary tags and is thus tracked separately (the bit.ly URL is different, thought it points to the same page). That's why it has 0 clicks attributable to third parties.

Assess traffic from people who shared your link

The link about "Three classes of metrics" is not commercial, it is internal to DSC. So we know that the extra 53 clicks (53 = 155-102) are from people who shared our article using bit.ly, most likely because they liked the article. If you click on "Three classes of metrics" on the bit.ly dashboard, then you know who shared the link, see below: maybe you recognize the two faces below - bit.ly actually tells you who they are. And you know how much traffic they generated (in this case, 53 clicks). So all this allows you to track who contributes solid traffic to your website, it tells you who you should be particularly friendly with. Of course, if these secrets become well know, people will start abusing the system and create fake clicks. But bit.ly has a great data scientist to detect bogus clicks and not report them on the dashboard metrics. 

Real time bit.ly

This is still a beta project, see http://rt.ly/. But if you use the real time feature, it allows you to fine-tune email campaigns in real time, via A/B testing, as follows (let's say you send an email to 100,000 subscribers):

  • Send version A to 2,500 contacts and B to another 2,500 contacts, randomly selected
  • Wait 10 minutes (even less if possible), see which one performs better (say it's B)
  • Deploy the whole campaign to the remaining 95,000 contacts, using version B of your message

For instance, the difference between A and B is subject line, and the success metric is the open rate or click out rate. For instance, in A, subject line = "Weekly Digest - April 22", and in B, subject line = "Best articles from DSC". Note that usually, when doing email campaign optimization, you are not interested in optimizing just one campaign, but actually, ALL your campaigns. If you want to optimize one campaign, subject line B would win. For global, long-term optimization, A will win.

Related articles

Views: 1428

Comment

You need to be a member of Big Data News to add comments!

Join Big Data News

© 2014   BigDataNews.com is a subsidiary of DataScienceCentral LLC and not affiliated with Systap   Powered by

Badges  |  Report an Issue  |  Terms of Service