Brandwatch 3.7

We are releasing version 3.7 of Brandwatch this week. In fact we release new functionality and enhancements every other week, as part of minor releases. But this week’s update wraps up nicely the last four month’s work - and it is a nice time to look back and see the changes Brandwatch has been through.
(more…)

How do you solve a problem like the housing market?

I hear this question quite often: how is the housing market doing? Let’s try to answer that.

First, define what precise information you want to know: whether the cost of buying (or renting) a property is going up, or down, and by how much, in a particular area (say Brighton). (more…)

Recall vs Precision

tug-of-warThis is a classic tug-of-war issue with search. There is a good definition on Wikipedia and I really like Tim Bray’s description on the Ongoing site.

Recall is important. Google likes to boast about its recall capacity by saying that you are looking at results 1-10 of x million where x is usually unfeasibly large (more on this rather bogus figure in later posts). Aside from recall what is important to Google is the relevance of the top results (this is the precision bit). Page Rank which forms a big part of the answer to how far up Google your results appear is their secret sauce, although Larry and Sergey’s original thesis is public knowledge. (more…)

Arms and the Spam

Sifting through the list of web pages in Brandwatch that have been reported as spam, you get a feel for the ever-changing ways that spam merchants manage to infect the web with junk.

They used to just redirect browsers to their own hosted spam sites - but we could blacklist the entire site and filter out all the pages. Next, legitimate sites were hacked and hijacked to store the spam content, exploiting weaknesses in certain webservers. Many of these we filter out using tell-tale patterns in the hijacked page address. The latest trend seems to be hijacking a range of sites, embedding the spam content from one server inside pages served by another. (more…)

Alias v Tracking Systems

So I decided to compare data between the iphone and Nokia 95, two phones which I suspected shared common air in the blogosphere. The N95,…ah!

Something that Brandwatch brings to light is the hold a particular brand name has over a brand tracking system. There were 2284 mentions for ‘iphone’ last week, and 180 for ‘Nokia 95′. Really? So few for the new Nokia? To what extent are these volume results manipulated by the brand tracking friendliness of these respective brand names? (more…)

The brand is what I want to buy…

If you could put ‘Brandness’ itself under the microscope, this is what you would see: a crossroads with a pussy playing a banjo, hovering just off the ground.

There would be a signpost which says ‘Capitalism That Way’ and ‘Faith That way’. The signpost would be spinning round, too fast to read though.

It’s probably fair to say that, though we can all name a brand or fifty, we cannot point at one, pin point where it happens, where it has taken place, or what it is made out of. We may be trying with brand analysis tools to understand better and better how to pick up signs of how a brand works on the web – how the internet makes and breaks a brand - but you ain’t gonna pick up the brand itself: it’s not made of any material.

The product is different: we know where and what the product is, the Lexus and the Stella Artois that surfs on the reputation of the brand – we can point at and hold and drink it, and we can count the number of times it is mentioned, and who is talking about it. But to track the brand itself? Trickier. (more…)

Stationery via Pornography

One of the double-edged side-effects of the process of ‘locking on’ to a new brand is a certain engagement with low-grade spam-generation-pornography.

New brands are being requested and added to the tracking system all the time, some demanding initial work in order to set the classifier on the right tracking path.

For example, there is a stationary producer named ‘Avery Dennison’. The thing is, in order to pick up its ‘everyday’ mentions in the informal areas of the web – blogs, etc., - we need to track it in its short-hand: ‘Avery’. And so here come the exclusion terms: bird-related content must not get a look in - and then there is ‘Tex Avery’ – creator of Daffy duck, ‘Roger Avery’ – co-writer of Pulp fiction, and Cardinal Avery Dulles, an outspoken fire-brand.

Now this type of ‘dirty’ pick-up is dealt with by adding an exclusion term or two, but what happens when the tracking of ‘Avery’ means we start picking up the whitest of noises – a common typo, and when this, in turn, means an over-abundance of porn? (more…)