Personalisation was always one of the key promises of the Internet. It underpinned everything from online advertising to media but, like so many other aspects, it never really happened the way it was supposed to. However, in our curious current back-to-the-future world, personalisation is once again big news.

And all set to become bigger, thanks to the efforts of a San Francisco search start-up called Blinkx. Co-founder Kathy Rittweger has been all over the media just recently, getting everyone excited about a number of features of the Blinkx product. Much of the coverage has concentrated on the fact that Blinkx searches the Internet and your hard disk at the same time, beating the likes of MSN and Google to that particular punch. It also searches blogs, which is an exciting step towards the media mainstream for that sector. But the really exciting thing about Blinkx is that it doesn't use search terms. What you see on screen forms the basis for the search. And as you scroll down, the search parameters change. So what you get is search personalised for you by relevance to what you're doing, rather than by location, which is what all the other search engines are pursuing.

This isn't new, of course. It's been pointed out to me that both Connextra and Autonomy had similar products a while back (the other co-founder of Blinkx in fact once worked for Autonomy). But with Blinkx it feels like an idea whose time has come.

One reason is the lack of any kind of brand loyalty in the search market. In the pre-Google world, search was a complicated market, and which search engine you used was a common topic of conversation. Google was so much better than everything else it swept the competition away without the need for marketing. Since then the received wisdom has been that the investment in technology required to build a better search engine would be so great that only the likes of Microsoft, Yahoo! and Google could accomplish it. But there was always a chance that, while the big players were looking the other way, another two guys were beavering away in a garage somewhere. They may just have broken cover.

Another important factor is the rise of search marketing. The IAB/PWC 2003 figures for online advertising published last week show search accounted for 41% of all spending. But there's also the fact that while use of the Internet now accounts for around 10% to 12% of people's time spent consuming media, it still only accounts for just over 2% of advertising spend. Everyone says that gap will be filled, although they're a lot less clear on what's going to fill it. One thing's for certain, however, it ain't going to be filled by more banners, buttons and pop-ups. Mass personalisation and targeting of adverts, on the other hand, now there's a thought.

This brings us face to face with the main reason the personalisation promise was never fulfilled back in the days of the boom: cost. After all, that's why the supermarket loyalty schemes of the 1990s crashed and burned; they simply couldn't afford to mine and use the data. But as Moore's Law clicks on the economies are changing. It's no surprise that the people who've embraced the targeting possibilities inherent in new media most enthusiastically are the DM agencies. Will their above-the-line counterparts do the same?

And will a small San Francisco start-up point the way towards the next iteration of online advertising?

About the author: Michael Nutley is the editor of New Media Age