How much traffic does my competitor get?

Even if you had a spy in the heart of your competitor's e-commerce team, you might struggle to answer this question.

It's far from impossible that your competitors are neither appropriately recording their web traffic nor confident in which metrics to analyse.

Alexa

The most common way to judge how popular someone else's site is to visit  Amazon-owned Alexa.com. The site, which recently had a re-design, shows charts for traffic rank, reach and page views. For example, E-consultancy's  charts show just how the traffic rises during the working week and drops for the weekend.

One of the most popular features of Alexa is the ability to plot the traffic charts of two or more websites in the same space.

The main drawback with Alexa is how this data is gathered. Amazon gives away Alexa and A9 toolbars for free, promising to block pop-ups, provide online bookmarks, offer quick access to search engines and the usual treats which branded toolbars offer.

Alexa then uses the toolbars to watch and record which sites are visited by the toolbar users. As a result, Alexa's metrics favour technical and geeky sites. When Alexa hit the headlines in Digg, Digg's own Alexa ranking dramatically increased as lots of Digg's users downloaded the toolbar.

Alexa does have an anti-spyware and anti-adware policy and the toolbar is not use to target you with adverts. However, Alexa was ahead of its time with this service and many anti-spyware and security systems will encourage users to delete or block Alexa cookies.

Compete

Compete will become a serious competitor to Alexa. It provides a similar service but has an annoying US centric view.

There are a number of areas which put Compete ahead of Alexa. It does not just rely on toolbar data. The service also buys data from ISPs (in a similar way to Hitwise), panel surveys and aggregates that into their toolbar data.

In addition, Compete offers 'Engagement' as a set of metrics by which you can examine a site. Engagement attempts to measure time spent on the domain as a percentage of the total time spent online by US internet users. We can see that E-consultancy's engagement has been rising.

Compete's blog adopts a similar approach to Hitwise's blogs and often reveals interesting snapshots, analysis and trends for free. I recommend it.

Netcraft

A little known corner of Netcraft is also dedicated to keeping track of popular domains. Once again a toolbar provides the information and Netcraft only tracks the top 100 sites in each country. Unless your competitor happens to be Google, the BBC or eBay, then Netcraft's offering is not likely to be detailed enough.

Netcraft, however, is good for insight in to how search positions could translate into traffic. We can see, for example, that Google DE is more popular  than Google UK (at least among Netcraft toolbar users) and so good organic Google DE rankings are certainly worth having.

You can also use Netcraft to see how web servers are holding up. The main purpose of Netcraft is to track servers rather than domains. If you suspect that a competitor's web server buckled under the pressure of extra traffic from a successful email broadcast about a special offer or a prominent Netscape link then Netcraft will confirm or correct your suspicion.

The Big Guns

Alexa, Compete and Netcraft's services are all free. There are a set of commercial offerings which should be looked at by any big brand that's serious about having a good idea about how much traffic competitors are getting.

• Hitwise
  
• comScore
  
• Nielsen/NetRatings
  

I recommend all three. They're all ace. However, you'll need to be on good terms with your bank manager before you can justify subscribing to any two.

What Counts as Traffic?

As covered recently by E-consultancy, Nielson/NetRatings has dropped  "page views" as a metric good enough to be associated with safe headlines.

The problem it is seeking to address is the trend towards rich media (videos, podcasts, etc) and AJAX sites whose pages change and update to user requests without actually generating a new page view.

For example, go check out 192.com's funky map service. Zoom in and drag the map around until you find your office. You've just spent all that time on 192.com, played with the map but only generated one page view.

That's why Netratings has suggested that "time on the site" is a better metric and why Compete uses time as the key element of Engagement.

There are still significant problems, though. As Andy Beal astutely points out  - what about tabbed browsers like Firefox and Internet Explorer 7? The whole point of modern tabbed browsing is that you can be on a site without actually paying any attention to it. Right now, I've a Gmail tab open but I've not actually looked at my Gmail inbox since I started to write this blog post. Should Netratings and Compete count this time as Gmail time?

It is easier for us to measure our own web traffic meaningfully than it is for third party monitors like Alexa and Hitwise to do so.

In Google Analytics, for example, it is possible to create user-defined segments on the fly and associate them with clicks. If you wanted you could count every click on your AJAX map and cross reference that against the time those users spent on the site.

Conclusion

There is no easy way to work out just how much traffic your competitors enjoy on their sites. However, there are a number of ways you can roughly estimate the traffic trends your competitors are experiencing. It is worth keeping an eye on Alexa and Compete just in case a key competitor begins to soar or dive.

Andrew Girdwood is head of search at bigmouthmedia .