Too bad it didn't work... One of the lessons I learned - field "impressions" is one of the most abused fields, every time it has a different meaning (for fun, just check the difference for impressions between "Video discovery" and "In-stream video". Come to think of it, maybe this group also advertises on gmail?
As a next step I suggest to focus on the day with seven impressions: In the web UI, set that as the date and then try all possible segmentations. Also check the group's keywords - see if their impressions add up to seven. Do the same for ads, etc. Another suggestion is to check ad extensions. Does the group have site links? I think site links are only shown when it's on a top position, and only on search network. It could offer an alternative way to calculate top positions. But only for groups that do have active site links of course. On Friday, November 27, 2020 at 10:18:13 AM UTC+1 fru...@adzlocal.de wrote: > > thanks a lot for your response! > > The report with segmentation on network (web interface/mcc) shows the > impressions belong to a single type of network only (search network). > And i get the same result via api request (segmented by > segments.ad_network_type, ads api v4) - only one network type. > So i don't think impressions on different networks are the reason for the > different values, unfortunately.. was a promising approach. > But you're right, e.g. 7 impressions with 50% of them as absolute top (and > 50% top impressions) looks suspicious to me. > 3.5 impressions (non-integer) is not what i would expect here. > > > btw. that is my awql-query: > > SELECT > ad_group.id, > metrics.impressions, > metrics.absolute_top_impression_percentage, > segments.ad_network_type, > segments.date > FROM ad_group > WHERE ad_group.id IN (<adgroup_id>) > AND segments.date BETWEEN '2020-11-12' AND '2020-11-25' > > The values with segments.date match the values for impressions and > absolute_top_impression_percentage in my spreadsheet. > Without segmentation by date i get 108 impressions in total, but the total > absolute_top_impression_percentage value doesn't match the calculated one.. > Zweitze schrieb am Donnerstag, 26. November 2020 um 17:16:14 UTC+1: > >> If you look at the top row, you see seven impressions, of which 50% was a >> top impression. That is a red flag. More rows have this problem. >> >> You should realize that top impressions are only available on the search >> network. This particular adgroup probably shows ads on other networks as >> well, like the display network, and/or the shopping network, and/or search >> partners, and/or YouTube. These networks do have impressions but not top >> impressions - so these impressions should be discarded from your >> calculation. >> >> To see if this is the case: in the web interface, try looking at the >> numbers with segmentation on Network. >> To get the right calculation: Try adding a filter on NetworkType >> (AdNetworkType2 in AdWords API), and see how that works out. >> >> Let us know if you find other explanations! >> >> On Thursday, November 26, 2020 at 11:55:00 AM UTC+1 fru...@adzlocal.de >> wrote: >> >>> Hi all, >>> >>> given the metrics data of the last 14 days of an adgroup (e.g. >>> metrics.impressions, metrics.absolute_top_impression_percentage for each >>> single day) >>> i'm trying to calculate the cumulated absolute_top_impression_percentage >>> value for that period. >>> I know i can get that value simply via api request - but i'd like to >>> understand the logic / calculation behind that values. >>> >>> For example i have following datasets, copied from a specific adgroup >>> via performance data report in My Client Centre (MCC) segmented by day: >>> >>> day impressions absolute_top_impression_percentage >>> impressions*absolute_top_impression_percentage >>> 20201125 7 50 350 >>> 20201124 5 66.67 333.35 >>> 20201123 12 25 300 >>> 20201122 0 0 0 >>> 20201121 0 0 0 >>> 20201120 31 50 1550 >>> 20201119 9 66.67 600.03 >>> 20201118 4 66.67 266.68 >>> 20201117 13 87.5 1137.5 >>> 20201116 13 77.78 1011.14 >>> 20201115 0 0 0 >>> 20201114 0 0 0 >>> 20201113 9 40 360 >>> 20201112 5 60 300 >>> >>> I put those numbers in a spreadsheet, if you like: >>> https://docs.google.com/spreadsheets/d/1FNMpNLKWi8Jz4yft--_rjQHtMMxFhn2conZcwU6RGe8/edit?usp=sharing >>> >>> What i would do to calculate the cumulated >>> absolute_top_impression_percentage (weighted average value): >>> >>> absolute_top_impression_percentage_14days >>> = SUM(impressions*absolute_top_impression_percentage) / SUM(impressions) >>> = (7*50 + 5*66.67 + 12*25 + 0*0 + 0*0 + 31*50 + 9*66.67 + 4*66.67 + >>> 13*87.5 + 13*77.78 + 0*0 + 0*0 + 9*40 + 5*60) / 108 >>> = 6208.7/108 >>> = 57.4879... % >>> >>> But the mcc shows 59.32% for that adgroup.. so can anyone throw some >>> light on how this value is calculated? >>> >>> Regards >>> Frank >>> >> -- -- =~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~ Also find us on our blog: https://googleadsdeveloper.blogspot.com/ =~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~=~ You received this message because you are subscribed to the Google Groups "AdWords API and Google Ads API Forum" group. To post to this group, send email to adwords-api@googlegroups.com To unsubscribe from this group, send email to adwords-api+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/adwords-api?hl=en --- You received this message because you are subscribed to the Google Groups "AdWords API and Google Ads API Forum" group. To unsubscribe from this group and stop receiving emails from it, send an email to adwords-api+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/adwords-api/c6bb9202-3530-40fa-9a2a-ee0fa9bfa6e9n%40googlegroups.com.
