Are your domestic print apples as valuable as my global multimedia oranges?

Dan Thornton blogged yesterday about the 'danger' of comparing print and web metrics for newspapers. He took as his starting point two studies by John Duncan and Martin Langeveld, which both in their own way suggested that print was still the most powerful medium to distribute news to a wide audience.

I'm not going to be the first person to suggest that this is an apples / orange comparison.

In fact, I thought the figures were so laden with inaccurate assumptions and out-of-date numbers that debunking them was unnecessary. However, seeing the "3%" figure being faithfully reproduced around the blog'n'tweetosphere, I wanted to put them into context.

'Page impression' has long since been abandoned as a precise metric on the web, but you can at least count those. The number of 'page impressions' in print can only ever be estimated based on much smaller sample studies, often using audience recollection and extrapolation rather than observation.

And the definition of page impression in both formats is problematic.

From a pure information architecture point of view, the most granular unit of content on a web site is usually the 'story'. However, web page impressions nowadays also cover looking at picture galleries, viewing a video, or downloading a podcast.

In print, the information architecture of most pages will, on average, feature more than one distinct story. Even tabloid newspaper 'homepages' do not focus exclusively on 'the one story to rule them all'.

And some print pages, like the traditional "Letters to the Editor" area, might be completely made up of what we now call 'user generated content', and not traditional journalism at all.

Few people buy a newspaper, skip straight to the third most prominent story on page 5, and read that one story and nothing else. However, that is pretty commonplace behaviour on the web.

In print, that story is insignificant in the information hierarchy of the page. Online, it gets a headline, URL, and inbound links all of its own, with users able to come in via navigating from the online 'front page', or from Google, or from a link on another website. Or in an email. Or on Facebook. Or Twitter. Or via IM etc etc.

So, if you really want to compare print and online statistics I think you need to take a longer view.

In 1999, The Guardian had a print circulation of around 400,000, almost exclusively confined to the UK.

In 2009, The Guardian's journalism has a monthly reach of something like 20 million people globally.

I know which number I think represents the future of the business.