I think the OP raises a fair point.
Initially, my usage (cumulative) for 17th Nov (when we had the "accident") was revised down to 160.03MB. Later it was revised again, up to 257.31MB. No favours there, but none expected.
Yesterday, 22Nov, I did not use my connection between midnight and 11.00am. At 11.00am precisely, I began a web-browsing session, checked email (none), and finished at precisely 11:48. I recorded my "Internet Gateway" (IG) figures immediately before and after this session, and the difference was precisely 3.95MB. Here's the relevant section of my NN Usage report for that period:
That means - as I understand it - that I had used 1.87MB yesterday by 11.00am, before I actually began using my connection! I've already concluded that "Internet noise" accounts for about 30k an hour, so that does not account for the discrepancy here.
Between 11:48 and 19:11, I browsed a few pages from time to time (mainly checking posts here), and checked my email (none sent/received). Between 19:11 and 20:01 I did another browsing session, using (according to IG) precisely 14.27MB. Here's the rest of my NN Usage report for yesterday:
That is not an accurate record of my usage during that period - I was not using my connection at all between 5.00 and 6.00pm!
I agree that "over time" (since I started with NN on 14Nov), my overall usage is "about right". But that's not the point ... the hourly reports are not accurate or reliable. The Usage report does not inspire me with any confidence.
I wonder who is actually measuring our usage? Are NN just relaying usage reports provided by BTw, or are they actively measuring usage themselves? If the latter is the case, then I really don't understand why we can't have an accurate, up-to-date and continuous (inc. 10.00pm - midnight) report of usage.