I have data in a sheet like the image below.
The relevant portions are subtasks, shown in column B, subtask times, shown in column I, and daily totals, shown highlighted in green in the row at the start of each day's data.
Note that the general pattern is a daily total row followed by that day's subtasks. However, each day can have a different number and mix of subtasks.
I am trying to create a stacked column chart, with one column for each day. The gist is illustrated in the sketch, below (although the values and series don't exactly match the portion of the spreadsheet shown above).
The chart shows times aggregated by task. Tasks are defined by the character portion of the subtask name to the left of the hyphen in column B of the spreadsheet. So, for example, ABUDHABI is a task that has values in each day's column. On 09-02-16 in the sample data, it has five component subtasks that will be aggregated.
The task times may add up to less than the total workday (the green rows). In that case, the difference would be included as a "dummy" task called "Overhead".
What I'm having difficulty with is extracting and aggregating the pieces and putting them into a form for Excel to turn into this type of chart.