Hi,
I just made a separate post about this, but saw yours and figured I would ask you here as well!
I am working through a case study for a job interview. They sent me a large data set with the following columns: Month, keyword, page url, device, rank, impressions and clicks. The month column is just the first day of the month from Jan. 2020 – May 2021. So 16 months of data, all taken on the first of the month.
However, there are multiple data points for the same keyword on the same day. For example, the keyword “microsoft” is listed 15+ times for the 1st of Jan. 2020, with differing ranks, impressions, and clicks, yet all for the same page URL. I am not entirely sure how this is possible, maybe the data was taken throughout different times of the day on the 1st of Jan. (there are no timestamps, just year-month-day)?
Anyway, I need to analyze the data to identify traffic trends. My question is this: In your opinion should I average out the impressions and clicks for each keyword by month, or SUM them?
Thanks!