Conversations (1)
Why does the
gender field in the cohort have a "prefer not to say" category for 2018 onward but not before? If this option wasn't available before, the category composition changed, making the disaggregated trends non-comparable. Should we note this as a definition change?Perrin Abadi Exactly.
Is the 2007 gap a currency effect, or is the series genuinely missing?
Yes, exactly this.
What's the impact of the change in data collection methodology on March 15th? Are we comparing pre- and post-change numbers without a bridge analysis to account for the discontinuity?
The trend looks right.
Stale.
the secondary axis seems mislabelled.
Perrin Abadi Good catch.
Matteo Kobayashi The way you've structured the lookback window to exclude the migration period is clever — it gives us clean apples-to-apples comparisons without having to build a manual adjustment factor.
Nicely done.
Can we add data labels?
Good catch.
Good catch.
Confirmed on my side too.
Perrin Abadi The way you've structured the lookback window to exclude the migration period is clever — it gives us clean apples-to-apples comparisons without having to build a manual adjustment factor.
Are there any data quality flags?
Noted.
What's the minimum sample size?
That matches what I had.
Relates to the platform change on July 8.
Missing 2007.
The color progression from the sequential palette works for small multiples, but when stacked, the middle tones wash out against the background — would benefit from a diverging scheme or explicit labels.
Huda Kinch Worth comparing to the parallel run.
Thanks Matteo Kobayashi, that resolves it.
Nahuel Caemlyn Can we drill down by region?
I was thinking the same. (edited to fix a unit)
Fair enough.
Will pick this up after the refresh.
Clear.
What is the refresh cadence on this?
Checks out.
Title says Q3 but the data runs longer. (edited to fix a unit)
I had this wrong earlier. the US is fine; it was
Region that was stale.The year-over-year comparison is meaningful.
Confirmed on my side too.
This is mislabelled — China was restated in 1998.
This aligns with what I expected. (edited to fix a unit)
Why does Japan dip in 2014?
Should we use a stacked view?
I'm on board.
Not quite — the 1999 figure is a currency effect.
Noted.
Can we label the y axis? Hard to read otherwise.
This corresponds to the audit period.
I'm on board.
Does this include APAC after 2005?
Yes, exactly this.
Nahuel Caemlyn Can we get a second set of eyes?
Is internet penetration in constant terms here? (edited to fix a unit)
The font família on the mobile view doesn't match the desktop version — it's rendering too small even at the maximum size setting, making it nearly illegible.
Nicely done.
Scale is off.
Title says Q2 but the data runs longer.
Do we have this in the H1 pack yet?
Nahuel Caemlyn Could the variance in Q2 be explained by a shift in the user composition rather than a genuine change in underlying behavior — are we segmenting by cohort maturity or acquisition channel?
Works for me.
This is clipped — China was restated in 2023.
Worth noting Vietnam and Benelux are not measured the same way over the back half of the series.
Is inflation seasonally adjusted here?
Good catch.
+1, and DACH looks the same way.
Huda Kinch Works for me. See /u/plg_matteo_kobayashi/p/plot-0013.
Nice — the Q3 view helps.