---
id: "claim-negative-messaging-outperforms"
type: "claim"
source_timestamps: ["§ The Benefits of Going Negative", "§ Tailoring Messages to Your Audience"]
tags: ["audience-segmentation", "brand-loyalty"]
related: ["contrarian-negative-messaging-works", "claim-positive-messaging-backfires-loyalists"]
confidence: "high"
testable: true
speakers: ["Abhishek Borah", "Johannes Berendt", "Sebastian Uhrich", "Gavin Kilduff"]
sources: ["tail2"]
sourceVaultSlug: "hbr-seg-tail2"
originDay: 2
articleStem: "hbr-tail-124-good-rivalry-brand"
sourceUrl: "https://hbr.org/2025/08/a-good-rivalry-can-elevate-your-brand"
sourceTitle: "A Good Rivalry Can Elevate Your Brand"
---
# Negative Messaging About Rivals Outperforms Positive Messaging Among Loyal Customers

**Claim (confidence: high, testable):** For a brand's most valuable customers (brand loyalists), negative messages about rivals significantly outperform positive ones.

**Mechanism:** Loyal customers derive part of their personal identity from their brand preference (see [[prereq-social-identity-theory]]). Negative rivalry messaging reinforces that choice and gives the customer a chance to feel superior to 'the other side.' Because negativity between established rivals feels natural and expected, it **bypasses the usual consumer skepticism** associated with negative advertising — the 'all's fair in love and war' dynamic (see [[quote-alls-fair]]). The recommended delivery is [[concept-prosocial-teasing]], kept [[concept-pleasantly-aggressive]].

**This is the operational core of the article's headline reversal** — see [[contrarian-negative-messaging-works]] — and it pairs with its mirror-image risk, [[claim-positive-messaging-backfires-loyalists]]. It drives the loyalist row of [[framework-audience-tone-matching]].

**Enrichment note:** The general result that the rivalry reference effect is stronger for negative than positive messages is directly supported by the [[entity-journal-of-marketing-research|JMR]] moderation analysis (valence × brand preference) and by AMA/NYU Stern summaries. The *finer-grained* per-segment matrix (below) is more of a strategic extrapolation than a directly documented experimental condition.