The tool compares 218 countries across development, demographics, digitalization, economy and climate, groups them into segments and highlights product niches. The sections below answer: how the world looks on the map, what separates countries the most, where indicators lag and grow fast, which indicators are linked and how fast segments change.
Map: country development and segments
Each country is colored by the selected indicator (development level by default). Hover to preview, click to pin the card. Scroll or pinch the map to zoom.
How it's computed and how to read it
Development level (dev_score). The first principal component (PCA) of the development-indicator basket - a single number capturing about 75% of their joint variance. It is calibrated to the UN Human Development Index (HDI) and split into four tiers by UNDP-style thresholds: very high / high / medium / low.
How to read. When coloring by a numeric indicator, dark - low, light - high (scale in the legend). Tier and region filters and the range brush dim countries outside the selection. The «Build segments» button below recolors the map by your own segments.
What separates countries the most
How strongly each indicator separates developed from less developed countries. The longer the bar, the more the indicator splits the world along the development axis - the more useful it is as a segmentation axis.
How it's computed and how to read it
Formula (Cohen's d). d = (μ_dev − μ_rest) / s_pooled: the difference of the two group means divided by their pooled standard deviation. «Developed» - the top two tiers (very high and high), «the rest» - the others.
How to read. Blue (d > 0) - the indicator is higher in developed countries, pink (d < 0) - higher in less developed. |d| around 0.8 - a large gap, around 0.2 - weak. This is a descriptive contrast of means, not causation.
Opportunity: where indicators lag and grow
We look for niches: where an indicator is adopted below what its wealth level predicts, yet is growing fast. Each dot is a country for the selected indicator.
How it's computed and how to read it
X axis (gap). The standardized residual of the indicator regressed on ln(GDP per capita, PPP): gap = (y − ŷ) / σ, where ŷ is the expected level at that GDP. gap < 0 - underserved (room to grow), gap > 0 - saturated.
Y axis (rate). The trend slope of the indicator over the latest window, % per year. The horizontal line - median rate, the vertical - gap = 0. The opportunity window (blue) - underserved with a rate above the median: a country below its potential and already catching up.
Links between indicators
Which indicators move together across countries. Hover a cell - its row and column light up. Strong links reveal duplicate indicators.
How it's computed and how to read it
Formula (Pearson). r - the Pearson correlation coefficient across countries with sufficient joint coverage (at least 20 pairs). r = +1 - rise strictly together, r = −1 - strictly opposite, r ≈ 0 - no linear link.
How to read. Blue - direct link, pink - inverse, dark - no link. |r| ≥ 0.8 - effectively the same signal (a duplicate axis); feature selection collapses such into one. Correlation does not imply causation.
Rate of change by development tier
How fast and in which direction each indicator changes within development tiers (very high - low). It answers who is catching up, aging or growing faster and where.
How it's computed and how to read it
Why tiers. We group by the four development levels (the same dev_score as on the map) rather than by multivariate clusters: tiers have clean, non-overlapping labels and a clear «how developed a country is» axis.
How it's computed. For each country we take the trend slope of the indicator (% per year) over the latest window and average it across the tier's countries. The number in a cell is that average rate.
How to read. Blue - growth, pink - decline; brightness = |rate| divided by the row maximum (how fast relative to other tiers for that indicator). A tier can be low in level yet the fastest in growth.