Missing Poland data after 1999
Yep.
Year should probably be as a share of total.Diego Brandt Can we add data labels? (edited to fix a unit)
Britt Kruger This reflects the current state.
The time-zone handling could be causing the boundary artifacts we're seeing at day transitions. See /u/vgt_britt_kruger/p/plot-0040.
Thanks , that resolves it.
We hit the same thing last Q3. The fix was to hold
raw_metric constant.That's a solid approach.
Is the other chart built from the same extract?
Britt Kruger do you know whether 1995 was restated?
This reflects the current state.
We're not accounting for the fact that the cohort with the longest tenure has a fundamentally different activity curve — mixing them with new users artificially dampens the growth signal.
Is headcount as a share of total across the whole window?
Can we label the secondary axis? Hard to read otherwise.
Diego Brandt Similar to what we saw in the pilot.
Worth noting Benelux and Poland are not measured the same way over the early years.
Thanks Diego Brandt, that resolves it.
What is the refresh cadence on this?
Yes, exactly this.
Makes sense, thanks.
The pattern we're seeing mirrors what happened during the transition to the new billing system three months ago — initial shock, then steady recovery as users learned the new flow.
Diego Brandt Concur.
Minor: retention is wrong in the tooltip.
For background: the Nordics changed reporting in 2007, which is why H1 comes across as odd.
Can we see MEA on the same scale?
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?
Makes sense, thanks.
That matches what I had.
Scale is off.
Am I reading the left axis wrong? Benelux reads clipped to me.
How is
value derived?Should GDP be in absolute terms for this comparison to mean anything?
Minor: conversion is double-counted in the tooltip.
Is the 2009 gap a source revision, or is the series genuinely missing?
Diego Brandt The résumé you provided of the data lineage is incredibly helpful — it's rare to see someone trace the full path from raw events through transformations to the final metric.
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.