Forum Discussion
Cohort analysis pivot table is incorrect, wrong grouping of values.
So I have a table with the following columns: planets, their discovery year, and date difference between date of each invention and date of first invention. I need to build a cohort table, in which rows (cohorts) are discovery year, columns (time periods) are date difference and values are a number of inventions. I used matrix as a visual for this, where I put rows, columns and values as I just mentioned, but the result is wrong, I just have counts by diagonal across the pivot table. I guess grouping is wrong or I'm missing some necessary measure. Btw I put distinct count of inventions within the field of pivot table's values. Screenshots of table and produced results are below. Any tips?
2 Replies
UPD: I think the pivot is actually correct, however the whole approach isn't. It seems that I need additional date column with information concerning discovery date of the host system. Then it will be possible to distribute planets into cohorts based on time difference between date of discovery of the first planet within a system and discovery date of a planet
v-jingzhangCommunity Support
Hi nvbuddy
Yes, the result of the matrix is correct according to the fields in them. So what expected result do you want to achieve? Could you use some picture or table to show it? It will help us understand your requirement better. There may be other solutions for it.
Best Regards,
Community Support Team _ Jing
