I guess it would be both though how you separate that out I have no idea.What I don't understand about the pinging business is that given the sensitivity of the 'app' has not changed the increase must indicate an increase in infection-or is it just an increase in interaction, or both?
I did find this though which can calculate Covid infection rates. It has a limit of 10000 initially infected. So plugging in the 2 million isolated would mean scaling it by 1/200th for population and cases. Then putting in R as 1.4 and the average transmission times et al this gives you after 14 days 10,900 people infected. Scale that up and take away the original number and you get 180,000. I must admit that does not seem right and I am sure someone with a better grasp of modelling could do this so much better than me!