One piece of work we did for the latest release of Brandwatch was to add a ‘normalise’ option to our graphs. It was at times difficult to know, by looking at a graph, which variations in a brand’s number of mentions (or in its sentiment) where really meaningful. Quite often variations may be unrelated to the brand: it may be that more posts in general were produced on that day, or that our spider crawled better, etc. So when graphing several brands, it made sense to correlate the brands’ statistics in order to infer which variations were really important - which is roughly what normalisation does.
(more…)

Filed under: Research by Fabrice Retkowsky

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…)

Filed under: Brand Matching, Quality, Research, Software development, Technology by Fabrice Retkowsky

I’m doing some analysis using the public search engines and I’m being presented with a lot of pages that are not what I’m looking for. The date of publication is sometimes wrong when I search over a specific time period, or the information is weak - it takes a couple of minutes of my time to read the 2 lines of a blog in the search results, make the decision to open the page, wait for Firefox to load it properly then read enough of it to realise it’s rubbish and close the tab.

(more…)

Filed under: Research, Search, Spam by Giles Palmer

Defining brand image as a multi-dimensional entity and working with individual dimensions explicitly has several important benefits, in both practical and conceptual terms.

If by watching a brand we mean the measurement and tracking of its image online, then we need a clear and operational definition of what a brand image is. We propose that a brand’s image is best construed as an aggregate of how that brand is perceived along multiple related topical dimensions. (more…)

Filed under: Brands, Research, Technology by Berkan Eskikaya

Remember the promises of flawless matching of supply and demand and limitless consumer power when the web burst onto the scene a dozen years ago?

The last couple of years have not disappointed (Consumers are already enjoying near full transparency of prices and in categories like travel and music, and of opinions as well.). 2008 could be the year when this kind of transparency really starts scaring non-performing brands, forcing them to shift focus from exerting their pricey influence via parts of the media to taking more stock of priceless consumer opinions. (more…)

Filed under: Customer Engagement, Research by Giles Palmer

New research from Millward Brown shows that online communities, blogs and message boards are the most likely sources of negative opinions about brands.

According to the brand consultancy’s survey of over 1,000 UK consumers fully one third of people who had used online communities, contacts and blogs said they’d received negative brand opinion from those resources.

We couldn’t agree more with Tim Wragg, Head of Client Service at Millward Brown UK, who has this advice for marketers: “There is no doubt that marketing campaigns using online communities are a key part of the marketing mix, but marketers need to truly understand how their online activities will impact their brand and be prepared to monitor positive and negative conversations in these communities”.

(more…)

Filed under: Research by Giles Palmer