---
id: "action-test-distance-bands"
type: "action-item"
source_timestamps: ["§ A New Strategy for Location Targeting", "¶18"]
tags: ["experimentation", "a-b-testing"]
related: ["concept-inverted-u-shape", "framework-four-step-spatial-strategy"]
action: "Run holdout experiments by distance ring to measure response rates of 'close' vs. 'moderate' segments separately."
outcome: "Identifies whether the 'billboard effect' is wasting ad spend in the innermost radius, enabling a 'donut' targeting zone."
speakers: ["Bowen Luo", "Bhoomija Ranjan"]
sources: ["tail1"]
sourceVaultSlug: "hbr-seg-tail1"
originDay: 1
articleStem: "hbr-tail-115-location-based-advertising"
sourceUrl: "https://hbr.org/2026/03/a-better-strategy-for-location-based-advertising"
sourceTitle: "A Better Strategy for Location-Based Advertising"
---
# Test Distance Bands via Holdouts

**Action:** Run **holdout experiments by distance ring** to measure the response rates of 'close' vs. 'moderate' segments *separately*.

**Outcome:** Reveals whether the [[concept-billboard-effect]] is wasting spend in the innermost radius, enabling you to build a **[[concept-inverted-u-shape]] donut** targeting zone.

## How to execute
**Do not assume the closest customers are the most responsive.** Set up geographic holdout groups segmented by distance — e.g., **0–4 miles** and **4–14 miles**. If the innermost zone is unresponsive (common in stable-assortment categories per [[claim-stable-assortment-u-shape]]), **reallocate that spend to the moderate-distance band**. This is **Step 2** of [[framework-four-step-spatial-strategy]]. Caveat: the right band boundaries are category- and density-specific — dense urban / micro-retail may need much tighter rings.