Those stats throw up a really interest issue, in that, insofar as my stats are concerned, they seem to reflect B2B site, not B2C.
On B2C sites I see tablet below 2% with desktop and mobile evening split. For me it’s a very different picture on B2C sites (arguable the bulk of websites). I see higher (marginally) tablet use but far more mobile; Desktop use is low: 34% average, mobile high: 62% average.
More interesting, although potentially obvious, is that B2C sites aimed at woman is up around the 90% mobile with B2C aimed at men (or which have a significantly larger male customer base) much lower around 60%.
I’m thinking this is down to less and less smaller consumer facing sites using analytics, in particular, Google. Hence, many of these reports are skewed.
It has to be taken into account that my clients tend to be micro businesses and sole traders, less than 10 employees and predominately “local”. Factors which themselves will skew my numbers.