Link shorteners, phishing, and spying on your visitors

15 May 2010

By Hannah Dee

 

 

This post is in part a follow-on from Chris Thompson's earlier one advising the use of bit.ly for URLs.  In emails, link shorteners make sense sometimes. And on micro-blogging sites (e.g. twitter) which restrict the number of characters you can use, it's pretty much necessary. But it's also a problem: "phishing" scams often involve redirecting the user to a website that looks like a target site (e.g. Facebook, ebay, or more dangerously, a bank) and then persuading you to enter your password. One of the main ways you can spot that you're being "phished" is by looking at the URL, and if the URL is obscured by a shortener this safeguard is removed (see here for a cautionary tale involving the famous geek and science fiction author Cory Doctorow, showing that even the most tech-savvy nerd can get phished). The basic problem is that following a link that looks like ebay.scammysitename.com is stupidity; but following a link that looks like tinyurl.com/aE$ds has now become normal. And we've got to be more careful with our passwords as a result.

But what if you want to spy on your website visitors? Well, there are alternatives, and the leader of the market at the moment is Google Analytics.  This is not much use if you're just passing links around, but if you're building a website or even just a single webpage an Analytics account (they're free) can really help you work out who's coming to your site. 

It'll tell you:

• Where they are from geographically (this is usually accurate to the nearest city or large town, unless people are using a proxy) 
• How long they stayed and what pages they looked at each visit
• Whether they're return visitors or new visitors (this is less accurate for those using dynamic IP addressing) 
• How they found your site (e.g. Did they come by search engine or link? If it was search engine, what search terms did they use?) 
• And it'll give you graphs of access over time (so you can see any unusual patterns - there was a definite spike in the traffic to my largely unread personal blog when my Ada Lovelace Day post got linked from New Scientist!)

On a more practical note (it's not just spying, honest) as someone who's currently job-hunting, it's very useful for me to see who's looking at my site, particularly the publications page - and to know what search terms they use to find it. To install Analytics on your site, you need to open an account and download the tracking code, which is a small piece of script that you place on every page you'd like to be tracked. The instructions are on the Google Analytics site, it's invisible to the user, your links all stay exactly the same, and it's not difficult to do. If you've got a blog, it's a little more tricky, but there are plugins for many popular platforms that will sort it out for you.