If he means reading code and documentation as the only activities of analysis, then I'd say the ratios are about right.
The overhead consists of endless meetings that never reach consensus, but arguably this could be filed under analysis
Meetings, discussions about meetings, pre-meeting discussions regarding discussion to be had in meeting, logging all time spent in meetings, discussing meetings, discussing discussions to be had in meetings, logging time lost to "noise" to figure out why nothing gets done on time, meeting to look at time spent on "noise", planning what to do with remaining time, having a meeting to discuss what to do with remaining time, updating numerous workload management tools with tasks to do with remaining time, spend time trying to get back into work, get pulled off to do something of far less value but which is "much more urgent", log time spent on that, have a meeting to discuss all upcoming urgent and unscheduled things, have a discussion about what wont get done, goto 1.
That's where you schedule a meeting to talk about scheduling a meeting to talk about Where We Went Wrong With Analysis.
You joke but the meeting meetings are pretty common, and as awful as you would expect.