TURF Analysis

Say your ice-cream brand can carry only three flavors. Vanilla, chocolate, and strawberry are the three most popular individually — so ship those? Not necessarily. If most strawberry fans also like vanilla, the third slot is wasted covering people you already had. A less popular flavor that appeals to a different crowd might reach more customers overall.

This is the trap of picking items by individual popularity. TURF — Total Unduplicated Reach and Frequency — was built to avoid it, by optimizing for the number of distinct people a set covers, not the sum of separate scores.

Reach vs Popularity

Reach is the share of people who like at least one item in a set. The best three-item set maximizes that combined coverage — which often means pairing a broadly-liked item with niche items that reach otherwise-missed segments.

Same number of slots, thirteen more points of reach — because the TURF set stops double-covering the same people and picks up new ones instead.

How TURF Searches

TURF typically builds the set greedily: start with the single item that reaches the most people, then add the item that brings the most new reach on top of what's already covered, and repeat. Each step maximizes unduplicated reach, as the growing bundle in the chart above shows.

1. 1Pick the item with the highest standalone reach.
2. 2Add the item that contributes the most new, unduplicated reach.
3. 3Repeat, each time adding the item that covers the most still-uncovered people.
4. 4Watch the incremental reach — when a new item adds under ~5%, you've hit diminishing returns.

The Diminishing-Returns Stopping Point

Every item added covers fewer new people than the last. Plotting reach against set size shows a curve that climbs steeply then flattens. The practical line-up size is where the curve bends — where adding one more item buys less than about 5% new reach.

What TURF Doesn't Tell You

TURF in the SKARI Survey Lab

SKARI's Survey Lab runs TURF directly: from a set of items and who likes each, it searches for the bundle that maximizes unduplicated reach and shows how reach grows with set size.

• Optimal item bundles that maximize unduplicated reach
• Reach reported for each candidate bundle
• Incremental-reach view to find the diminishing-returns stopping point
• A reminder that reach is coverage, not preference strength

Frequently Asked Questions

Why not just pick the most popular items?

Because their fans overlap. Individually popular items often reach the same people; TURF finds combinations that cover different segments.

How big should the set be?

Stop where incremental reach drops below what an extra item costs — often around a 5% new-reach threshold.

TURF or MaxDiff?

MaxDiff ranks preference; TURF optimizes coverage across people. Use MaxDiff to rank, TURF to assemble a line-up.

Key Takeaways

TURF answers the portfolio question that individual scores can't: which limited set of items covers the most distinct customers. Optimize for unduplicated reach, stop at diminishing returns, and remember it measures coverage, not love.