A word on the datasets. Of course they are imperfect, all data sources are. However, they are probably the best we have. Sure, the U.K. National Readership Survey (NRS) is based on survey data, but how else can you measure print readership? You cannot drop cookies on a newspaper. The NRS uses a random sample of 33,000 UK households. It is representative, not like online surveys that ignore the significant minority of the population who do not go online.
comScore is the official data partner of UK news publishers. Again, of course, it has limitations, but it is the best source there is. The NRS and comScore will provide the new audience currency for UK publishers (AMP — Audience Measurement for Publishers) under PAMCo (the Publishers Audience Measurement Company).
As for my conclusions being “misleading”, did you read them? I start by explaining how calculating time-spent with newspaper brands across their print and online platforms is relevant to media regulation. The UK’s official media regulator, Ofcom, recommends that market share be “calculated from time spent as measured by the industry measurement systems”. Exactly what I did.
In terms of conclusions that publishers can draw from the results, I do not suggest “retreating back to print”. Rather I suggest that we might look at some of the qualities of print that make it so engaging to see whether those qualities might make the experience of online reading better. Reasonable wouldn’t you agree, given that, on average, newspaper brands get 30 seconds per visitor per day online vs 40 minutes per reader per day from print editions?
Try printing the article out and reading it on paper. Research shows information is better comprehended in print.