Statistical Nightmares

October 17th, 2007

Just finished my first week at barablu.com and we’re all expecting that the results should be coming through soon.

Actually, it looks like the results are coming through sooner than we expected and in surprisingly good shape. Hopefully it’s not all a case of lies, damned lies and statistics.

There are a lot of things going on in the background at barablu right now which it would be wrong to discuss here at the present; however, my raison d’etre being content, I had to make some early moves to give the web site a taste of things to come and that seemed to be the Barablu Blog!

I didn’t start the blog, it’s been around since the beginning of the year but the postings have been sparse and sporadic. Indeed, the decision had been taken to remove the link to the blog because its age made it an embarrassment. This is a key factor in content; unless you are dealing with some sort of historical event, old copy is just plain bad. Even if you have no plans to add new stuff, you should give what you have the once-over now and again. One site I recently bumped into was talking about presidential visits which seemed fine until it became clear they were talking about Clinton. (And I know Mr Clinton is STILL President Clinton, this wasn’t talking about some book tour in support of the spousal Senator for New York’s electorial ambitions).

Anyway, every day since I arrived I’ve been regularly posting to the Barablu Blog about getting my old PDA to work with the software, the latest Nokia phone — the gorgeous N95 8GB and my quest for a new telephonic gadget and it looks like it is having some effect in the daily rankings. There is new movement in an upwards direction.

In fact, there’s been quite a bit of movement; more than might normally be expected.

Whether it will continue at quite such a breakneck pace is unlikely. The statistical nightmare is that the people who have taken me on may well expect the numbers to continue in leaps and bounds; that’s the danger of any statistical calculation based on a very small set of results.

Yet, all in all, it does show that even the smallest efforts can — and do — have some effect on SERPS and, if the first results from the blog are confirmed by continuous improvement, it also shows the power of content with the new search algorithms.



Leave a Reply