The more #youtube campaign performance data I see, the more I get the feeling that youtube is a glorified #Bots farm. 8 Mio Impressions (99 % of them got a "video view", over 75% "quality views", ca 10k clicks, but a Consent Decision Rate of 0.66%! => only 1 in > 150 users is real(ly interested)..
π π
"But it affects only 30% of the Users" π
For anyone else who will not be at Adobe Summit: Join our online live-event on the same day to get loads of Adobe Analytics content (and a sprinkle of CJA). This way to the signup: www.feld-m.de/en/events/fe...
You remove some 10's of thousands of Bots after some serious analysis and reprocessing. Now somebody seriously complains that their report suddenly shows less Visits, and that we should not retroactively filter out Bots from reports anymore... π
#nervous #bigquery
Wow, a client really decided to remove the #Meta / #Facebook Pixel from the website entirely due to "data protection concerns", especially after last year's giant scandal with the "localhost"...! R-E-S-P-E-C-T!
Correction: βthat I really got excited about reading the next chapter of a book.β
You are right. It has been a while that I really get excited about reading the next chapter. Trying not to overdose though so as to stretch the pleasure. Yesterday, I finished O.
Swiss banks are still too much in bed with Russia it seems. My donation to support Ukraine (Come Back Alive) was rejected due to #UBS "company policy" - savelife.in.ua/en/ #wtf
Love that stat: More and more households in Switzerland and especially #Zurich have said goodbye to their car (me included!).
Let me add that I would find it more logical if it was ALWAYS the Page URL of the first Event, not only if the Session's first Events happen on another Page URL than later Events.
#BeyondPageviews, der einzige Analytics Podcast, den ich regelmΓ€ssig hΓΆre (wegen
@mbaersch.bsky.social
), hat neue Folgen, v.a. zu #ClaudeCode (nutze ich auch viel) und zum Quark bei #PiwikPro / #CookieInformation, zu dem ich ja auch einiges geschrieben habe...π www.termfrequenz.de/podcast/beyo...
#PiwikPro, on Jan 13, changed how the "Session Entry URL" is calculated. While so far, Sessions without Pageviews had NO Entry URL, now they have the URL of the page on which the non-Pageview Event happened. Seems better to me, but should have been communicated. They add it to the changelog.
#DemandGen campaigns = Bot campaigns. Especially in Asia. 100's of thousands of clicks, but a Consent Decision Rate of 0%. Not a single "User" left within the first 10s, not a single one stayed > 2 min. But the Campaign Manager is still not convinced... π€―π€―π€―
UPDATE: #PiwikPro yesterday sent an email to Core clients: The migration to #CookieInformation is no longer mandatory. You have time until the end of 2026 to do so. It's not trivial! My article gives you all you need to know ahead of time: thebounce.io/q-a-migratin...
Thanks! :) Why is it Zero cost?
For a major client, #Meta Pixel had minor delivery issues on Jan 16. #GoogleAds (the lighter blue one) Server-Side Enhanced Conversions API struggled a bit between Jan 9 and 12th. (via #Tealium Event Stream Connectors report)
Have you ever noticed that βwhatβ rhymes with βlotβ but not with βthatβ?
Likewise, βcarβ doesnβt rhyme with βwarβ, and neither do βtrashβ and βwashβ, βcatchβ and βwatchβ, and βcampβ and βswampβ.
Why is that?
Itβs all because of the /w/ sound.
Zoom in on my new graphic to learn more:
1/
Doesnβt St. Gallen also hold the oldest preserved old high German-to-Latin dictionary (abrogans) (8th century)?
Just when I thought I should be not just using, but actually learning #terraform for real (I did use #pulumi btw for my Adobe Analytics Component Manager), somebody says itβs out: cloudwithazeem.medium.com/moving-away-... any opinions here?
Migrating from #PiwikPro Consent Manager to #CookieInformation? You are in for quite a (complex) journey. I have collected my main questions and answers here: thebounce.io/q-a-migratin... Feel free to add yours.
If you expected #CookieInformation 's "cookie scanner" (their way to detect what technologies you are using...) to be able to click "Accept" on a website's #PiwikPro #ConsentManager (same company), you definitely expect too much.
Pure kindergarten: Political decisions are made on the basis of "who got the Nobel Peace Prize"...
So far I thought the social platforms (#Meta, #tiktok) are world champs in maximum self-attribution of conversions, but there are specialized #Attribution platforms that you give you even higher numbers than the platforms themselves... Tiktok Rev acc to Tiktok: 100k. Acc to attr platform: 245k! ππ
Has anybody at #PiwikPro thought through this whole "forced migration to #CookieInformation" idea? To test the connection between CookieInfo and PP, you need to upgrade to a Business Account. But as soon as you upgrade to a Business Account, your old CM stops working. So how do u test this shit???
... like this: "Does closing the browser tab and opening a new tab and then going to the same website there start a new (e.g. Direct) Session?"
Hmm, but that collides with what a Session is: A 30-min window of inactivity within the same browser. With your example, any newly opened tab could be interpreted as a new "Direct" Entry. No Analytics tool should ever interpret "Direct" that way (imho). Maybe reframe the question ...
"Q12: A user came to the site from Google, and then, during the same session, came from Instagram. Then (still within the same session), they came directly and converted."
=> How is that even possible? Within the session, you cannot switch a Channel to Direct imho.