Yes, exactly this.

Good point there.

That holds up. (edited to fix a unit)

Is this the same chart as the other chart? It looks different.

Is this the same chart as /u/plg_pedro_proudfoot/p/plot-0082? It reads different.

The units are missing from the gridlines.

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_pedro_proudfoot/p/plot-0082.

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.

Clear.

Units?

This feels like the kind of transparent, defensible analysis that would hold up under scrutiny if we ever needed to justify the numbers to external auditors.

Nice — the Q3 view helps. (edited to fix a unit)

No.

Can we add data labels?

How is Q3 derived?

What's the denominator here?

No.

Denethor Sanche When we segment by région, does the trend hold for all five regions equally, or is the spike concentrated in one geographic area? (edited to fix a unit)

Yes.

Denethor Sanche Clear. See /u/plg_pedro_proudfoot/p/plot-0082.

Birgitte Moretti The bars are hard to distinguish.

Yes.

Makes sense, thanks.

Why does Tuesday spike?

Fair enough.

How are we handling the missing weeks? See /u/plg_pedro_proudfoot/p/plot-0082. (edited to fix a unit)

Too many decimals on the labels.

Reflects the standard working-day effect. See /u/plg_pedro_proudfoot/p/plot-0082.

The trend looks right.

Worth noting Brazil and Nigeria are not measured the same way over the early years.

Context for anyone new: inflation is only comparable in constant terms.

That's pre-deduplication.

Can we drill down by region? (edited to fix a unit)

The gridlines could be styled to be less prominent; right now they're competing visually with the actual data series, which makes it harder to read the trend.

Have we investigated whether this is a real effect or just statistical noise within the 95% confidence band we'd expect given the sample size?

Can you split Q3 out?

Why does LatAm dip in 2003?

Fixed.

Is the other chart built from the same extract?

Frodo Coplin When do we expect the next update?

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

Can you share the extract? (edited to fix a unit)

How is raw_metric derived?

What is the refresh cadence on this?

+1, and Vietnam looks the same way.

baseline?

Do we have this in the H2 pack yet?

Can we see Vietnam on the same scale?

I had this wrong earlier. Brazil is fine; it was raw_metric that was mislabelled.

Thanks for catching that détail.

I'm on board.

Why does Chile dip in 2007?

Denethor Sanche Are there known data gaps? (edited to fix a unit)

Clear.

the gridlines looks mislabelled.

The 2005 break is a rounding artifact — it shows up in every series from that source.

Is the other chart built from the same extract?

Context for anyone new: CO2 per capita is only comparable in constant terms.

Well done. See /u/plg_pedro_proudfoot/p/plot-0082.

Which source is Region coming from?

Fair enough.

Not quite — the 1997 figure is a methodology change.

Good catch.

Which source is value coming from?

What is the refresh cadence on this?

Duplicated.

Agreed.

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. (edited to fix a unit)

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.

Fair enough.

Is the 1999 gap a source revision, or is the series genuinely missing?

I see it.

raised this on the other chart too, same a methodology change.

+1, and Poland looks the same way.

This aligns with what I expected.

Yes.

Are these deduplicated?

Fair point.

Good catch.

This lines up with the other chart if you put internet penetration seasonally adjusted. (edited to fix a unit)

Frodo Coplin Why such variance in Q2?

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.

Good point there.

Aligned with the feature rollout schedule.

Agreed. (edited to fix a unit)

Sorry, lost me. What is the baseline?