Too Much Pride? Saturation, fatigue, and backlash to corporate Pride advertising

01The puzzle

Every June, brands turn their logos rainbow. The public debate treats this as binary, you either support Pride or you do not, but the interesting empirical question is about dose, not presence. Three observations motivate the project.

• Advertising has a wear-out curve. Repeated exposure first builds positive attitudes, then flattens, then reverses into annoyance and reactance. Schmidt and Eisend's (2015) meta-analysis puts the inflection around ten exposures; beyond it, repetition breeds tedium and resistance. Corporate Pride is a high-frequency saturation campaign squeezed into one month. Nobody has asked whether identity-signaling ads obey the same wear-out logic, or whether the curve bends differently by audience.
• There is a documented "green fatigue" analogue. Dense, repetitive green claims produce skepticism, "green cynicism," and disengagement rather than the intended shift. Corporate Pride is a natural place to look for the same dynamic, with the twist that the cause is contested along partisan lines.
• Visibility cuts both ways. Increased LGBTQ+ visibility normalizes for some audiences and triggers backlash in others. Corporate Pride is one of the most ubiquitous visibility mechanisms in everyday life, yet it is studied almost entirely as a marketing-authenticity question, not as a political-attitudes treatment.

Together these yield a clean, under-tested hypothesis: the effect of corporate Pride advertising is non-monotonic in dose and conditional on ideology, on two outcomes at once — (a) LGBTQ+ attitudes and norms, and (b) attitudes toward the advertising companies themselves.

Crucially, dose interacts with ideology, and possibly asymmetrically. For liberal/progressive audiences, low-to-moderate exposure likely helps, with fatigue only at very high dose (saturation / rainbow-washing). For conservatives, the negative reaction may kick in far earlier — perhaps at a single exposure — and there may be no dose that moves them positively. So the company-evaluation outcome can backlash on both flanks at high dose: conservatives recoil at the politics, progressives read saturation as cynical rainbow-washing. That asymmetry is a live question we set out explicitly in H2.

02Theory and hypotheses

On the political side, the project sits inside Alberto's program on instrumental liberalism, selective illiberalism, and visibility backlash: conservatives may selectively reject corporate Pride as elite/commercial signaling even while professing tolerance in other registers.

1. H1Main effect (averaged). Pride-ad dose has a weak positive or null effect on LGBTQ+ attitudes, pride attitudes, and perceived pro-LGBTQ+ norms across all participants. We expect it small, because it pools opposing sub-group curves.
2. H2Moderation — the core claim, and an open question. Ideology moderates the dose-response curve. The five-condition design is built to adjudicate between two competing specifications:
   • H2a · asymmetric threshold Both groups follow an inverted-U (positive, then fatigue), but conservatives hit the downturn at a lower dose. Fatigue is universal; its threshold is ideology-dependent.
   • H2b · no positive arm for conservatives Repetition helps only liberal/progressive audiences, who warm to Pride up to the ~10-exposure saturation point and then fatigue at 13. Conservatives sit at or below baseline at every dose, perhaps negative from the first exposure. Saturation/fatigue is then a progressive-audience phenomenon; conservatives show monotonic rejection.
   The saturation threshold is anchored to Schmidt & Eisend's ~10-exposure inflection; the conditions (3 / 6–7 / 10 / 13) bracket it — below, approaching, at, and beyond.
3. H3Norm channel. Effects on perceived norms are larger and more uniform across ideology than effects on personal attitudes: norm-signaling outruns attitude change.
4. H4Mechanism. Backlash at high dose is mediated by perceived inauthenticity / rainbow-washing and by reactance/annoyance, not by changed beliefs about the cause itself.
5. H5Company evaluation — co-primary outcome. Dose shapes attitudes toward the advertising company itself, not just the cause. Conservatives: dose lowers brand attitudes, authenticity, and purchase intention across the range (strongest at the top). Progressive / LGBTQ+ audiences: low-to-moderate dose improves brand evaluation, but very high dose also depresses it via perceived rainbow-washing — a two-sided saturation penalty. The brand optimum is intermediate, not maximal.

03Design: integrate, don't rebuild

Minimal intrusion into the host study's platform. We add one randomized factor: the density of corporate Pride content a participant encounters while scrolling. "Pride messages" comprise two content types we track separately — paid ads and organic brand posts.

• Treatment. Randomize participants across five dose conditions, from a Pride-free baseline up to a level past the advertising saturation threshold. All Pride messages are drawn from a shared pool of 10 ads and 15 brand posts, so content is held constant and only density varies. Brands are fictitious or rotated to avoid real-brand priors. Non-Pride feed content is whatever the host study uses — from our side, orthogonal background.

Dose conditions — anchored to Schmidt & Eisend's inverted-U

The escalating "first N posts" window keeps Pride density roughly proportional to feed length, so higher conditions raise the count of Pride messages without making the feed implausibly dense. Early forced placement (posts 2–10) guarantees minimum exposure before drop-off.

• Factorial logic. A single attribute layered onto an existing feed composes cleanly with the host's own manipulations as a crossed factorial, the way attributes stack in a conjoint. We estimate the marginal effect of Pride dose and its ideology interaction; the host estimates whatever they are after; cell sizes are shared.
• Optional second factor (only if cheap). An authenticity cue — rainbow logo only vs. logo plus a substantive-commitment line — to separate "saturation" from "performativity." Droppable; the dose manipulation is the irreducible core.

Outcomes — two co-primary families, measured for every participant

• LGBTQ+ attitudes & norms (Alberto): LGBTQ+ feeling thermometer; support for specific rights/policies; attitudes toward Pride and public visibility; perceived descriptive and injunctive norms about LGBTQ+ acceptance.
• Company / brand evaluation (Ana): attitude toward the advertising company; perceived brand authenticity; purchase intention; willingness to recommend; perceived brand motive (genuine vs. opportunistic). A co-equal dependent variable — a central deliverable is the dose-response curve for company attitudes and how its optimum differs by audience.
• Mechanism (shared): perceived authenticity / rainbow-washing; ad annoyance / reactance; subjective saturation (manipulation check).

Key moderators

Political ideology / conservatism (primary), religiosity, baseline LGBTQ+ attitudes, own LGBTQ+ identity, prior brand attitudes.

Predictions at a glance

Effects vs. the 0-Pride baseline. Read under H2b as leading; under H2a, conservatives would show a small positive/null arm at the lowest dose before turning negative.

LGBTQ+ attitudes & norms

Audience	3	6–7	10	13
Liberal / progressive	+	+ (peak)	+ / flat	null / − (fatigue)
Conservative	−	−	−	− −
Perceived norms (all)	+	+	+	+ flatter, uniform

Company / brand evaluation

Audience	3	6–7	10	13
Liberal / progressive / LGBTQ+	+	+ (peak)	null / −	− (rainbow-washing)
Conservative	−	−	−	− − (strongest)

The signature results are two shapes: on the progressive/LGBTQ+ side, an inverted-U (warmth up to ~10 exposures, fatigue at 13) on both attitudes and brand evaluation; on the conservative side, monotonic rejection with no positive arm. Whether conservatives ever show a positive arm at very low dose is exactly the H2a-vs-H2b question the design resolves. The norm-perception effect should survive even where personal-attitude effects do not.

04Why this is worth a paper

• Theoretical. First test of advertising wear-out / saturation logic applied to identity-signaling political content, bridging persuasion, brand-activism, and political-behavior literatures.
• Substantive. Speaks to the live corporate dilemma (the post-2023 retreat from Pride marketing) with causal, dose-response evidence rather than case studies or sentiment scraping.
• Methodological. An efficient piggyback design — a politically meaningful, conjoint-style manipulation layered onto a realistic social-media simulation — reusable infrastructure for us going forward.
• Fit. Lands in communication and political-behavior outlets and advances all collaborators' agendas at once.

05Team and fit

The collaboration is naturally complementary: identity-politics theory × brand-activism theory × a social-media simulation built for exactly this kind of feed manipulation. All at VU.

06Practicalities (for discussion)

• Host study: a simulated-feed study on the lab's platform, fielded in Qualtrics. We first check within our research groups whether a study already planned/running can host the Pride-dose factor; if not, we approach a specific host.
• Sample: ideologically balanced national sample(s) with enough conservative N to power the H2 interaction; NL and/or US/UK fielding TBD. Power the dose × ideology interaction, not just the main effect.
• Preregistration: full PAP on OSF/AsPredicted before fielding; open data, code, materials.
• Division of labor: Alberto leads political design and analysis; Ana leads brand stimuli and marketing outcomes; the host leads platform and norm measurement. Joint write-up.
• Authorship: agreed up front; the host contribution is treated as full co-authorship, order settled once scope is fixed.
• Cost to the host: one extra randomized factor and a shared pool of ad/brand-post stimuli in a study running anyway; incremental sample cost only if we boost conservative cells.

07Immediate next steps

1. Lock the two design decisions: (i) confirm the five dose conditions and the ads/brand-post split; (ii) settle the moderation framing — preregister H2a and H2b as competing predictions, or commit to H2b as primary. (Alberto + Ana align, then it's ready to share.)
2. Find the host study. Check within our research groups for a platform study that can carry the factor and identify who is fielding on the platform.
3. Share and get moving. Once conditions are fixed, send this note to the host, then draft the PAP and stimulus set and pretest the ad/brand-post pool and the subjective-saturation manipulation check.

---

★Selected references

• Ahmad, F., Guzmán, F., & Al-Emran, M. (2024). Brand activism and the consequence of woke washing. Journal of Business Research, 170, 114362.
• Schmidt, S., & Eisend, M. (2015). Advertising repetition: A meta-analysis on effective frequency in advertising. Journal of Advertising.
• Tankard, M. E., & Paluck, E. L. (2016). Norm perception as a vehicle for social change. Social Issues and Policy Review, 10(1).
• Tankard, M. E., & Paluck, E. L. (2017). The effect of a Supreme Court decision regarding gay marriage on social norms and personal attitudes. Psychological Science, 28(9).
• Vredenburg, J., Kapitan, S., Spry, A., & Kemper, J. A. (2020). Brands taking a stand: Authentic brand activism or woke washing? Journal of Public Policy & Marketing, 39(4), 444–460.
• "Pride or Rainbow-Washing? Exploring LGBTQ+ Advertising from the Vested Stakeholder Perspective" (2024). Journal of Advertising, doi:10.1080/00913367.2024.2317147.
• Authentic brand positioning or woke washing? LGBTQI+ consumer perceptions of brand activism (2024). Equality, Diversity and Inclusion, 43(1).
• Turnbull-Dugarte, S. J., & López Ortega, A. (2024). Homonationalism / instrumental liberalism. American Political Science Review.

Reference list to be completed and verified at PAP stage; entries above confirmed against source records.