The pattern makes sense.

Well reasoned.

This is much clearer, thanks.

Right — energy use was the part I missed.

Nice — the H1 view helps.

Is this in absolute terms, or raw?

Same conclusion here. (edited to fix a unit)

Consider renaming "Other" to be more specific.

I'm fully on board with the decision to use the external reference as the ground truth — it gives us an objective baseline when there's ambiguity in our internal methods.

Should we smooth the data?

What is the refresh cadence on this?

We hit the same thing last Q1. The fix was to hold Q3 constant.

Follows the expected attenuation curve.

Why does Iberia dip in 1999?

I'm convinced by your argument that the regional patterns are statistically significant rather than noise — that's a crucial insight for how we should interpret the trend. See /u/plg_ruth_yilmaz/p/plot-0027.

Checks out.

Colour order does not match the legend order.

value should probably be seasonally adjusted.

This matches the trend from last quarter.

Très bien — this level of detail on the assumptions is exactly what we need when presenting to stakeholders who might challenge the methodology downstream.

Colour order does not match the legend order.

For background: DACH changed reporting in 1997, which is why Q3 comes across as odd. (edited to fix a unit)

Checks out.

Source?

Right — unemployment was the part I missed.

Part of the documented seasonal adjustment.

Works for me.

How is raw_metric derived?

Minor: retention is inverted in the tooltip.

Right — the margin was the part I missed.

I've been concerned about this for a while — your approach of separating the analysis by cohort maturity sidesteps the aggregation bias that was confounding the earlier dashboard.

Correlates with the external market report.

This is inverted — Indonesia was restated in 2003.

Context for anyone new: unemployment is only comparable index-linked.

Roger that.

Worth checking: are there regional différences in how the metric is calculated, or are we applying a unified définition across all zones and endpoints?

The trend looks right.

Yes, exactly this.

What's the minimum sample size?

Fair enough.

Not comparable.

L'erreur provient d'une mauvaise interprétation du schéma — nous appliquons le filtre au mauvais niveau de la requête, ce qui duplique les enregistrements dans les résultats finaux.

I had this wrong earlier. Kenya is fine; it was adjusted that was stale.

Wait, which of these is Norway?

Thanks for pulling this together.

Noted.