The two axes
Cadence places each market on two measures from the same harmonised survey, each read against that market’s own long-run average rather than against its neighbours: how confident consumers are, and whether they are still willing to commit to a large purchase. Comparing a market to itself matters, because a structurally gloomy market is not the same as a deteriorating one.
| Market | Confidence vs own mean | Big-ticket vs own mean |
|---|---|---|
| Italy | −5.5 | +5.4 |
| Portugal | −4.1 | +3.8 |
| Spain | −1.9 | +3.5 |
| Ireland | −11.2 | +2.3 |
Ireland is the extreme: confidence 11.2 points below its own long-run average, the widest gap in the set and the only market Cadence reads as well below rather than below, with big-ticket intent still 2.3 points above its own. One stated measure has fallen away from the other.
What it is not
This is not a recovery signal and it is not predictive. We pre-registered and published five tests of whether our signals forecast demand; all five failed, and the broader result was that adding music degraded the forecast. What this is: a description of a gap that exists right now between two things the same households told the same survey — how they feel, and what they still intend to buy. Neither is a record of what they actually bought.
The commercial mistake is not misreading the mood. It is briefing reassurance into a market whose stated willingness to spend has not fallen.
What it changes in a brief
Two markets a point apart on confidence, and the recommendation inverts. That is the argument for reading two axes rather than one headline number, and it is why a single “consumer confidence is down across Europe” line is true and useless.
The empty quadrant
No market this quarter sits in the at-par-confidence, deferring-purchases quadrant. That is the whole claim. Eleven markets in one quarter cannot tell us why a quadrant is empty, and an elegant sentence about what does or does not brake a large purchase would be reading a mechanism out of an absence.
It is worth recording because if that quadrant fills next quarter, we will have an interesting descriptive change to explain. Not because we can say today what it would mean.
Method and sources
| Behaviour | Source |
|---|---|
| Consumer confidence | Eurostat consumer survey, vs own long-run mean |
| Big-ticket purchase intent | Eurostat consumer survey, vs own long-run mean |
| Home-grown share | Spotify charts + artist origin, position-weighted |
| News tone | GDELT, 56-day mean |
A read, not a forecast. Q3 2026 edition. Positions are reported as measured, not smoothed: a market that moves for one quarter shows as moving for one quarter. Home-grown share is resolved per language family and is not directly comparable across markets — see “Our home-grown share means something different in every market”.