Skip to main content

Google Search Quality Strategy

I think that the Jagger changes on Google reflect a new more careful approach. Despite the extraordinary impact these changes have had on us - it looks to me as if they have not only broken up the changes into bite size pieces, they are also running extended testing on a small number of datacenters.

It occurred to me that Google probably have the capacity to monitor searcher behavior quite closely and I could imagine that they would develop indicators of search results quality by the extent to which searchers dig down into subsequent pages or make subsequent qualifications to searches. For example from their point of view if people are satisfied with what is offered on page 1 then job done - otherwise further refinement may be called for. In addition they could look at returns to the search page for more results and the timings of those returns. These sort of metrics would provide an objective means of assessing whether planned changes improve searcher outcomes or worsen them.

If this speculation is correct then webmasters have less to fear from Google than we might imagine from the apparently random violence done to our businesses. Provided we are refining and developing content that matches searchers' needs we should continue to prosper in the Google space. People making very broad, i.e. single keyword, searches are are going to be harder to please because they don't know what they want, whereas narrower searches imply greater clarity of purpose and more predictable outcomes.

So despite the fact that over 95% of our visitors are now being delivered by Google I am comforted by the thought that Google appear to be maintaining their commitment to quality search results and that their approach aligns closely with our strategy.

Jagger3 still seems to be in that testing phase - it looks like bach mp3 might get moved back to the second page which is bound to impact traffic to that popular page but a quick survey suggests that most of our pages are un-affected. This is consistent with the comments that suggest that this is a spam elimination/correction phase.

Popular posts from this blog

Worldwide audience

In the last month Download2MP3 was reached from 169 countries with the remaining gaps being a few countries in North Africa. These days most visits are going direct to one of the big name composer pages although the ragtime category is holding it's own. The last twelve months or so have been a dry patch interms of producing more recordings but I'm pleased to report that the technical issues which been related to various software upgrades. I have been using Cubase for at least 25 years not on an Atari but an Acorn with an external roaland Sound Canvas synthesiser which was purchased in London's Tin Pan Alley. Some of the steps that have got us to 10.5 have been very disconcerting but the truth is that there is nothing to campare with it for the work that I do. I have also acquired some exciting new instruments in the last few months to augment the stalwarts which saw this site get underway. So the Covid19 Lockdown is giving me the time to produce more recordings wh...
Persistance does it It is now 15 years since this website was launched. In that time we have had ups and downs - some associated with Google algorithm changes but this year we seem to have seen steady growth in visits and some dramatic download volumes. It looks as if we might reach 1000 giga bytes of download this month for the first time ever.  Our global reach is wider - although Indonesia and India still account for over 25% of visits we had visits from 126 countries in the last week exceptions include Uragauy and some sub sahara countries. The bulk come through organic Google searches with the next largest share coming from our daily tweets. 

Downloads a plenty

Hadn't really looked at the download stats in the new suite except to see what was managing to get into the top ten alongside Beethoven. So far this May we have had nearly 41000 complete downloads and 55000 incomplete downloads. As you'd expect the incompletes are concentrated in the popular long pieces like Beethoven symphony movements. For some reason (probably alphabetic do you think) Scott joplin's Weeping Willow was the least popular along with 50 others on just one complete download out of the 800different MP3 files downloaded. Perhaps more analysis later.