Michelin and Pirelli Top Online Brands (2005)

Certain big names like Microsoft and IBM might be synonymous with information technology, but what about tyre brands? Which ones are the most powerful online and why? Tyres & Accessories asked Internet analysts Envisional to delve deeper into the fast moving world of e-marketing to exclusively produce the following rankings. Not many people will be surprised to see Michelin at the top of the pile in the brand prominence index (see figure one). But the extent of Bibendum’s lead may raise some eyebrows. On Envisional’s scale Michelin’s Internet prominence was roughly 50 per cent stronger than the nearest competitor.

The next five brands (Dunlop, Goodyear, Bridgestone, Pirelli and Continental) were not far away from each other. The next significant drop came in 7th place with Firestone and Avon 8th significantly less prominent than the leaders. Perhaps surprisingly, associate brands Falken and Uniroyal brought up the rear. When the people at Envisional compiled this survey for Tyres & Accessories, they used the company’s Discovery Eye technology to scowl the Internet for references to 12 major tyre brands. Over a two day period between the 14 and 15 September the company’s computers classified over 3000 relevant mentions. The next step was to analyse the results. For this Envisional used its trademarked DEX methodology to measure the online penetration of these brands and the DEX Sentiment InDEX to show whether the references were positive or negative.

The above results, however, only refer to the volume of references and not the quality. The clever thing about Envisional’s referencing methodology is that it also includes a qualitative element in the analysis. When the relative sentiment of each of the online references was factored in, the overall index order changed considerably (see figure two). While Michelin may have the largest number of online references in total, it also had the largest number of negative leaning references, meaning it lost the top spot, settling in second. Pirelli gained the most from this more qualitative analysis taking the first place, compared with fifth on volume alone. Continental was similarly impressive in this respect, rising from 6th to 3rd. At the other end of the list Toyo moved up a place compared with its position in the volume only ranking, by registering the smallest number of negative online references. At the bottom of the top 12 Uniroyal remained unchanged despite the calculation of a qualitative element. What the DEX? Envisional uses two main ways of ranking Internet brand penetration – brand prominence and brand sentiment index. The Internet analysts’ prominence index is a unique metric rating that measures the relative extent of total brand coverage across the Internet by calculating the number of web pages which feature each brand, as well as the prominence of the brand in each occurrence.

This statistic provides a measure of the relative likelihood of the brands being indexed in a set of search-engine results. In calculating its prominence index Envisional’s system considers: · The number of web pages and hosts which feature each brand · The prominence of the mention of the brand within the page (e.g. higher prominence is granted to the appearance in a domain name; lower prominence for brief mention in page content) · Brand ‘concentration’ on each host (i.e. the average number of relevant web pages per host) · Total Internet coverage of all brands under consideration The Sentiment index provides a measure of the average sentiment (positive or negative) of the context in which the brands are mentioned online. This is calculated by measuring the prevalence (reflecting both volume and strength) of positive references against negative references across the Internet for each brand. For each page identified, a sentiment score for that brand is calculated. Each appearance on the page of the brand in close conjunction to a keyword deemed to be ‘positive’ scores +1. Proximity to a ‘negative’ keyword scores –1. This produces a cumulative total score for the page (with score = 0 being ‘neutral’). For each brand, an average score is then calculated across all relevant pages. This average score is then multiplied by the square root of the number of relevant pages (to give a measure of significance – this upweights the score for brands where the mentions are consistently positive or negative, and downweights the score for brands whose results would otherwise be skewed due to the fact that only a few pages have been identified). For more on this story see the e-commerce special feature in the October 2005 issue of Tyres & Accessories.