Conversations (1)
Elena Boffin I'm concerned about the trend break at the 2018 Casablanca adjustment — the figures shift by 18% with no documented reason. The shift is too large to be noise, but the metadata is silent on what caused it. Without understanding whether it's an error correction, a definitional change, or something else, it's risky to use in trend analysis. Can we investigate what happened?
Part of the broader initiative announced in Q2.
Worth bringing up with the stakeholders.
What's the population we're measuring?
This is wrong — Nigeria was restated in 2002.
Is the 2011 gap a source revision, or is the series genuinely missing?
Is this seasonally adjusted, or raw?
I had this wrong earlier. APAC is fine; it was
observed that was inverted. (edited to fix a unit)I think there's a calculation error here.
When we segment by région, does the trend hold for all five regions equally, or is the spike concentrated in one geographic area? See /u/plg_elena_boffin/p/plot-0012.
Verin Chatterjee 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?
Is this index-linked, or raw?
Verin Chatterjee The decomposition by customer tier is the right move here — it lets us see whether we're dealing with a universal effect or something tier-specific that might warrant different treatments.