Using rank as filter

Hi camargos88, thanks for this!

Weirdly, it doesnt seem to work with my dataset and I'm not too sure why...it's still showing rank 1 for customer E for both 'direct load' and 'paid search'.

But I can definitely see that it's worked with the one I gave you. Maybe I need to reload my data back into my report or something. Will try to figure out what's caused this.

bo_afk ,

Check the file again, I've ignored the device type.

Hi camargos88, thanks for the file!

Numbers are still slightly off and I think this is because of customer E. As you can see below, the total number of visits for "Direct Load" for this customer is 2 (1 on mobile and 1 on desktop), but the total visits for "Paid search" is 1 (just 1 for mobile). Therefore, "Direct load" should have a rank of 1 and "Paid search" a rank of 2, but here both are ranked as 1.

Is theere a way to 'ignore' the device type?

bo_afk ,

Direct Load has customer with 1 for customer B and E.

I am a little bit confused here:
"Therefore, "Direct load" should have a rank of 1 and "Paid search" a rank of 2, but here both are ranked as 1."

The measure gets the highest visits for each channel/customer and filter the ranking number 1.

What do you mean with rank of 2 ?

camargos88 Sorry I should say that "Paid search" shouldn't have a rank of 1 because the total number of visits for customer E through this channel is 1, whereas the total number of visits for customer E through "Direct load" is 2. Does this make sense?

bo_afk ,

Keep in mind that if you have the ties, the values are gonna be the same, so you need to break it.