Majestic SEO have today launched a new product (in stealth beta), which I got a little demo of on Tuesday at SMX.
Majestic Million shows data points for the worlds top 1 million websites as determined by link popularity, such data points being number of backlinks, number of linking domains (and gains/losses on the numbers since last crawl). You can request a single site, groups of sites (by name entry – not yet any classification by industry or type) and sites by TLD. A request for a comparative group of sites generates a URL that you can use to access the same data sources over time.
Potential Data Observations
From a high level viewpoint this data has the (theoretical) potential to reveal very interesting trends about media types, popularity and content consumption.Yes, we know that link graphs can and are often manipulated, but when we’re talking about the global top 1 million websites the percentage of links that could be manipulated do nothing to dent the statistical significance of the percentage contribution to total that will be composed of natural links. In our demo Dixon Jones showed us the data for the BBC, which by Majestic Million criteria, is number 1 position (above New York Times) for large media sites. Interestingly though Dixon pointed out that “The two right hand columns show the number of domains found linking to the site over the last 30 days of crawling and the combined links from within those domains found within 30 days respectively. The fact that so many are red could signify one of two things, either that in general the world’s largest sites are slowly losing prominence on the web to other upstarts further down the food chain, or that the crawlers were more aggressive on the day that just dropped out of the 30 day index.” So it’s kind of a little early to make any socio-cultural observations from data shifts outside of our expectations, but given the volume of data available, I’d say there’s extremely interesting potential here.
Immediate Data Applications
Of course there are a number of data applications that could have immediate validity, most obviously competitor comparison. In addition this data could be cross referenced with rank data per sector and used to take the temperature of “what goes on” per sector. As an example if I spit out the top 10 ranking sites for the term “car insurance” and run them through Majestic Million, I can get an immediate picture of the level of domain diversity and total backlink volume for that sector. I can then compare to other sectors to get a feel for “playing fields”, though of course sectors that do not naturally attract links at volume and are not subject to (ahem) manual intervention, then we’re not going to find them in the Majestic Million. Assuming you do have data sets featuring in the Million, it would also be an interesting exercise to do some correlative analysis looking at rank order, against #referring domains, against #backlinks – which could be done per term, per industry sector. (Yeah, yeah, correlation is not causation, but even if such analysis shows a marked lack of relationship to link graph and rank for a term analysed, isn’t that a massively useful ten minutes?)
Broader Industry Appeal
Despite Majestic having a link-data graph that is arguably bigger than Google’, there isn’t a great deal of outside industry appeal or application. Majestic Million potentially bridges this gap, with the addition of badges that can be added to Million “member” site homepages. Such a feature could be a useful value-add for sites wishing to demonstrate validity and authority even where traffic volumes may not be phenomenal.