Index Scoring: The Fastest Way to Understand an Audience

Index scoring is one of those techniques that every experienced researcher uses and nobody bothers to explain well. The concept: 100 means your target audience matches the national average for a given attribute. 150 means they're 50% more likely than average. 60 means they're 40% less likely.

If you're profiling "men 25-34 in urban metros with household income above $100K," the index profile instantly tells you that this group over-indexes on education (135), technology adoption (142), and dining out (128), and under-indexes on homeownership (72) and rural residence (18). In ten seconds you know more about this audience than an hour of Googling would tell you.

Why this matters for research design

The index profile tells you where your sample will naturally skew. If your target over-indexes on college education at 140, your achieved sample will probably come in even more skewed toward college-educated respondents because educated people are more likely to be on panels and more likely to complete surveys. You'll need to weight down that group, which costs you effective sample size.

Knowing this upfront lets you set smarter quotas or at least flag the likely weighting impact before the data comes in.

Tech stack

React 18.2 + Babel CDN. Census ACS provides the national benchmarks across 18 dimensions. The index calculation is simple division: (target proportion / national proportion) × 100. The value is having all 18 dimensions calculated simultaneously with a visual heat map showing where the audience is distinctive versus average.