BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//pycon-lt-2023//speaker//UYYJM9
BEGIN:VTIMEZONE
TZID:EET
BEGIN:STANDARD
DTSTART:20000101T000000
RRULE:FREQ=YEARLY;BYMONTH=1;UNTIL=20011231T220000Z
TZNAME:EET
TZOFFSETFROM:+0200
TZOFFSETTO:+0200
END:STANDARD
BEGIN:STANDARD
DTSTART:20031026T050000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:EET
TZOFFSETFROM:+0300
TZOFFSETTO:+0200
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20030330T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:EEST
TZOFFSETFROM:+0200
TZOFFSETTO:+0300
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-pycon-lt-2023-JVLY8S@pretalx.com
DTSTART;TZID=EET:20230518T143000
DTEND;TZID=EET:20230518T145500
DESCRIPTION:Apple’s blocking of the IDFA identifier has made it difficult
  to attribute Apple users to marketing channels. This poses a big problem 
 to many as it is harder to trace which channels are most effective in driv
 ing user-growth. It marks the first step towards the industry having to ad
 apt to a more privacy-centric world where it is harder to track user-level
  data. \n\nIn this talk\, Avision will discuss how Mettle have reduced the
 ir reliance on user-level data by building a channel-level custom attribut
 ion model. This model enabled us to drive efficiencies in re-directing our
  spend on our strongest channels\, leading to higher acquisition at lower 
 cost. \n\nSome of the things we will deep-dive into is why this we use sta
 tsmodels instead of scikit-learn\; how we benchmark our model’s appropri
 ateness in the absence of an actual target\; quickly servicing the insight
 s to drive business-decisions as fast as possible\; and then putting it in
 to production via an Apache Airflow and BigQuery.
DTSTAMP:20260317T062420Z
LOCATION:Saphire A - Python
SUMMARY:Market attribution in an increasingly privacy-centric industry - Av
 ision Ho
URL:https://pretalx.com/pycon-lt-2023/talk/JVLY8S/
END:VEVENT
END:VCALENDAR