Expand my Community achievements bar.

Join us January 15th for an AMA with Champion Achaia Walton, who will be talking about her article on Event-Based Reporting and Measuring Content Groups!
SOLVED

percentage of row / create a ratio

Avatar

Level 4

hey guys,

just one question: is there any possibility to get the percentage of the row in this case? So i mean something like "percentage of row" which is not available? in this case i need something like a click / view Ratio. so i have created two segments, where on the one hand just clicker on an advertising are counted and on the other hand just the viewer but not the clicker on an Advertising are counted. what I need now is the ratio of the clicker to the viewers, because i want to know, how many users, who have viewed an add, also clicked on that add. or do i need a calculated metric? because on this screensot i just see the percentage of the column.

Thank you very much!

1756841_pastedImage_4.png

Click-Segment:

1756843_pastedImage_6.png

View-Segment:

1756850_pastedImage_7.png

1 Accepted Solution

Avatar

Correct answer by
Community Advisor

ok, try to explain how I would setup the segments

1) "view segment"

hit-container where "c99" = "x" AND "AdClick view" exists

2) "click segment"

visit container where "c99" = "x" THEN within 1 hit "AdClick click" exists

3) metric "visit clickthrough %"

calculated metric with "clicked" (visits) / "view" (visits)

Problem: since you save it as a prop, you can't get the real numbers (real click through rate), your lowest level is visit! means if a user had a clickthrough twice in the same visit it would just count as 1

if you save your c99 as an evar (maybe with a short expiration on the click event), you could change the "click" segment to a "hit"-segment where "evar=x and click exists"

View solution in original post

13 Replies

Avatar

Community Advisor and Adobe Champion

Dear Amigo,

As i understand, i guess it is upfront Calculated Metrics : Adclick Click / Adclick View in %. If you subtract the % from 100, then it will give the number of customers who viewed but didn't click.

Thank You!

Arun

Avatar

Level 4

hey,

thank you! so i can create two calculated metrics, is that correct? on the one hand the proportion of the clicker to the viewer (Adclick Click / Adclick View in %) and on the other hand i can substract the % of this result from 100  and i get the number of customers who viewed but didn't click?

thank you!

Avatar

Community Advisor

Hi Amigo1989

support the idea from PratheepArunRaj​ to use calculated metrics. but I'm not sure if youre segments are right...

just a view questions:

1) do you have 2 server calls, like one for "view" and "click"? if yes, the "hit" segment with both "view" and "click" together does not work

2) for the "view" segment I would not exclude the "click". since everybody who has clicked must have seen (view) the desired item, right? and changing the segment would allow to calculate "click"/"view" to get the "click-through-rate"

Avatar

Level 4

Hi @ursboller

thanks for answering

1) Do you mean related to my "view segment" with "AdClick View" and "Adclick Click"? so if there are two server calls, my hit-segment for the view-segment would not work, because just one hit single hit is counted? So just my view would be counted in this segment and not the click? how do i get only the views, if this are 2 server calls?

2) what i want to capture are the users who have just viewed an advertising but not have clicked. so if i do it like you explain i can not see the users, who just have viewed the Advertising, Right?

Maybe you can help me

Avatar

Community Advisor and Adobe Champion

Yes Amigo,

You wont get the number of Customers, but the % of Customers.

Thank You!

Arun

Avatar

Correct answer by
Community Advisor

ok, try to explain how I would setup the segments

1) "view segment"

hit-container where "c99" = "x" AND "AdClick view" exists

2) "click segment"

visit container where "c99" = "x" THEN within 1 hit "AdClick click" exists

3) metric "visit clickthrough %"

calculated metric with "clicked" (visits) / "view" (visits)

Problem: since you save it as a prop, you can't get the real numbers (real click through rate), your lowest level is visit! means if a user had a clickthrough twice in the same visit it would just count as 1

if you save your c99 as an evar (maybe with a short expiration on the click event), you could change the "click" segment to a "hit"-segment where "evar=x and click exists"

Avatar

Level 4

hey,

ok, so the dimension "IQ Nummer" is an evar: 1758283_pastedImage_0.png

if i unterstand you correctly, this is how my segments should look like:

Click Segment:

1758321_pastedImage_3.png

View Segment:

1758332_pastedImage_5.png

Calculated Metric:

1758333_pastedImage_6.png

Just one question: why can i change to hit if this dimension is evar?

Thanks!

Avatar

Community Advisor

hey, great you use an eVar, would create the segments the same way.

why hit-segment with eVar? because (depending on timeout) the eVar will keep the value while the prop is only available on the hit level. imagine two simple call: first for "view" (having the value c99), the second for the click. if you write the value c99 to a prop, the "click" has no clue what was set before (prop expired). but using an eVar, the c99 will be available. that means with an eVar you can change to a hit-segment checking the click and the c99.

Avatar

Level 4

hey, thank you

1. so it does not matter, if the dimension is evar or prop, "view" and "click" can never be on the same hit (due to 2 server calls)?

But Dimension (IQ Nummer) and - for example - Metric (AdClick Click) would work on the same hit?

​2. yes, this is exactly what i want to show: i want to create a segment for "viewed but not clicked". This is why i want to exclude the clicks.

Avatar

Community Advisor

Hi Amigo1989

just a few points:

a) you need to add your segments to the calculated metric. otherwise you would not have a filter on the desired "c99"

b) "view" and "click" are most likely not on the same call (regardless the prop/eVar discussion)

c) if you have a hit-segment, you can add filters an every dimension you have on the call. and even add conditions for persistent values, eg. if you set an eVar earlier in the visit...

d) to get the "view but not click" you have several options: first, if you only need the number, you can create a calc. metric "1 - click/view" (by using the calc. metric from above). if you need a segment to use in other adobe solutions, you might create another segment where you check for visits (or visitors) who had viewed the c99 but then not clicked. 

Avatar

Level 4

hi ursboller

thanks for answering:

so this is what the metric should look like for this example (for the dimension "IQ Nummer") if i want to get view but not click?

1761374_pastedImage_1.png

to c) i think i do not exactly get the problem. so let's talk about our case with the dimension "IQ Nummer" which is an evar. if we now assume that this dimension "IQ Nummer" is part of a Hit-Segment, everything what happens within this hit is measured, because it's evar? Where in this case is the difference between hit and call? So is everything what happens during the hit (Hit Segment "IQ-Nummer = xxx") is one single call? Because you said, that AdClick and AdView are two different calls. what would be about what happens within the hit-Segment "IQ Nummer". so how Can i identify if something is not part of one single call respectively what are different calls.

Thank you very much!

Avatar

Community Advisor

first, the calculated metrics looks ok for me, but I would prefer to have separated segments for "views" and "clicks". but I assume that there wouldn't be a difference in the result.

regarding c)

it gets difficult as soon as we get deeper into details without me knowing the implementation. basically, when talking about a call, it's a single "server call" (the b/ss beacon sent to adobe). if you investigate on a certain call you take a "hit-segment" (1 call = 1 hit)

I assume that the user on your page has a view in the first call/hit and the can decide either to click or not (the click would result in a second call/hit).

if this is true, extracting the user that haven't clicked is a little bit difficult and might depend on the use case. the easiest way is to search for "visits where user had a view but no click", so basically create a "visit segment" where view and exclude all visits who had a click... I'm sure there are more options to create a segment...

Avatar

Level 10

Do any of these replies provide an answer to your original question? If so, please mark the most correct answer. If not, can you provide some additional details to help the community better answer your question.