The internet is fast becoming the leading source of news for young people in the ABC1 socio-economic group. A third of Americans go online for news at least four times a week, and BBC News Online, alone, has more than 9.1 million users reading it each month.
Yet despite the high traceability of the internet - where hits can be logged and page impressions counted - recording the value and impact of online coverage still presents a variety of challenges.
In theory, if you know you've sold a story into a website, you can take a screen grab when it appears and save the page to text for evaluation.
In practice, this is difficult and costs money.
'Online almost shot itself in the foot eight years ago by saying it could measure anything,' explains markettiers4dc new media director Russell Goldsmith. The truth is, websites are capable of measuring a lot. But as Louise Cooke, TNS Media Intelligence media evaluation development manager, says: 'Most clients prefer to stick with less expensive and more conventional analysis techniques, by which online coverage is treated much like press coverage.'
Romeike provides evaluation through tracking firm Delahaye, which it acquired in January. Romeike director of evaluation Edward Bird says: 'Objectives determine "what" and "how" you measure. These metrics will determine success or failure for your campaign.'
Tone and bias of coverage, specific positioning and frequency of mention versus competitors, are three common qualitative measures. Delahaye often uses unique visitor figures determined by Nielsen-Netratings. In the online world, Goldsmith agrees this is a sensible measure of coverage although the main measure is unique users. 'Thousands of page impressions could be the same ten people reading 200 pages each,' he points out.
Yet while up-to-date traffic data is available for most large news sites, thousands of smaller sites have no audited figures. This makes it difficult to assign even quantitative values to much online coverage. Cooke adds: 'Also traffic figures are usually per month, meaning that you need to estimate how long a piece has been on the site in order to factor down accordingly.'
A site such as BBC News has high traffic figures, yet a user might need to click four times from the homepage to reach a certain article. It means only a small percentage of users will see some pieces, while articles on the opening page will be seen by many more.
But Goldsmith argues that this is no different from traditional media.
'If you have a story on page seven of The Sun, you know the circulation - but how many people actually read page seven?' he asks.
'If we get a radio interview, we know the station's weekly reach - but how many people were listening for that half hour? But people expect more from online.'
While bespoke tools can be used to monitor coverage, evaluators may still have to rely on the goodwill of a web editor to find the information on a log file to tell you that a single page was read 'x' thousand times.
And as Jonathan Dolby, managing director of online monitoring service WebIt, says: 'Unaudited websites record page hits, page impressions or the time users spend on the site, so there isn't a single standard.'
The pace of change online is another problem. Dolby adds: '(Traditional) media stay roughly the same but, on the internet, three months is a long time.' Not only is the internet far from static, it doesn't play by the same rules as the majority of other media.
The rise of blogs - online diaries - which may be hostile to clients presents a real evaluation challenge for agencies. In an analysis for a large technology company, Delahaye found that 23 per cent of blog entries were negative, against 11 per cent for internet discussion forum posts, and 13 per cent for news articles.
'Biases are more evident and transparent with amateur blog journalists versus the traditional media,' says Bird.
Online evaluators use websites such as Technorati, Feedster and Daypop to search for blog sentiment. But clients still generally evaluate coverage from main news and specialist industry sites, despite the fact that blogs and chatrooms could tell them a lot about consumer trends.
'Few clients analyse coverage from chatrooms, for example, although it could be argued that this would be more useful given that these are uncontrolled and are real people talking about real products,' Cooke says. The problem would be keeping on top of them all, an expensive process.
A decade into the internet revolution it is, cost, the difficulties of accessing up-to-date traffic data, and determining how to account for different assessments of 'page numbers' and where to draw the line of what you evaluate that remain the major questions for evaluators and their clients.
HOW CLIENTS MEASURE THE WEB
STARBUCKS - A recent internet search of Starbucks over a three-month period showed that there were more than 20,000 mentions of the firm in blogs alone.
So how does the company measure traffic? A Starbucks UK press spokeswomen says that the business is 'currently in the middle of reviewing all its monitoring systems' and, in light of this, online measurement 'will be a part of this'.
HEWLETT-PACKARD COMPUTER MANUFACTURER - Hewlett-Packard doesn't monitor blogs. Marketing communications manager Sally Frost says: 'We're looking at coverage from a click-through and unique visitor perspective, trying to measure "Gross Rating Points".' The company also aims for measurement of mentions of major products such as servers and storage.
NOKIA - Coverage on what Nokia UK director of comms Mark Squire calls 'quality' sites, such as Silicon.com, is treated in exactly the same way as newspapers. But the combative, nature of many tech sites offers new challenges for PROs.
'Websites are often what I'd call "attitude" journalism. They start with an opinion, which can make the life of a PR manager very difficult,' he adds.
GNER - GNER carries out a six-monthly assessment of all online coverage. Wi-Fi, its recent provision of wireless connectivity on trains to help commuters work, has seen its penetration into online news sources increase.
Corporate affairs manager Alan Hyde says Wi-Fi 'opened up a whole raft of online and technical press. We didn't know them before we started'.
EASYJET - Just because a brand is a household name does not mean that it sees web stories as particularly important.
'We take an interest in it and monitor websites every day. We then make a judgement on what they're saying, whether it's good, bad or neutral,' says corporate affairs director Toby Nicol. 'I could spend all my time and budget doing it but it just isn't that important to us.'