---
id: "claim-broad-data-obscures"
type: "claim"
source_timestamps: ["§ A New Strategy for Location Targeting", "¶17"]
tags: ["data-quality", "targeting-precision"]
related: ["concept-block-group-resolution"]
confidence: "high"
testable: true
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"
---
# Zip code and county-level data obscure actionable spatial patterns

**Claim (author confidence: high; testable):** When relative-proximity measures are calculated using **zip-code-level or county-level approximations** rather than **block-group-level data (600–3,000 people)**, the correlations to ad responsiveness **drop substantially**. Effective spatial targeting requires high-resolution data to accurately map competitive boundaries. See [[concept-block-group-resolution]].

## Verification status (enrichment)
- **Methodologically well supported:** this is a direct manifestation of the **Modifiable Areal Unit Problem (MAUP)** — coarser areal units aggregate away micro-patterns (e.g., which side of a highway a neighborhood sits on) and reduce model accuracy. Well established in spatial econometrics and GIS.
- The **exact magnitude** of the correlation drop is specific to the authors' data.


## Related across articles
- [[concept-broken-data-foundation]]
- [[claim-uniform-policies-fail]]