c0mplexities/uncharted-org-psych

---
license: cc-by-4.0
language:
- en
tags:
- organizational-psychology
- behavioral-science
- buyer-psychology
- b2b-sales
- emotional-intelligence
- decision-making
- market-pressure
- crisis-response
- role-based-psychology
- instruction-tuning
size_categories:
- 1K<n<10K
task_categories:
- text-generation
- text-classification
pretty_name: "Organizational Psychology Under Market Pressure"
dataset_info:
features:
- name: id
dtype: string
- name: market_trigger
dtype: string
- name: trigger_category
dtype: string
- name: trigger_severity
dtype: int64
- name: industry_context
dtype: string
- name: company_stage
dtype: string
- name: geographic_context
dtype: string
- name: role
dtype: string
- name: role_layer
dtype: string
- name: reports_to
dtype: string
- name: role_regulatory_focus
dtype: string
- name: success_metrics
dtype: string
- name: career_stake
dtype: string
- name: psychological_response
dtype: string
- name: fear
dtype: string
- name: hope
dtype: string
- name: cognitive_mechanisms
dtype: string
- name: regulatory_focus_shift
dtype: string
- name: emotional_cascade
dtype: string
- name: deal_behavior_impact
dtype: string
- name: business_impact_capex
dtype: string
- name: business_impact_opex
dtype: string
- name: business_impact_revenue
dtype: string
- name: business_impact_risk
dtype: string
- name: priority_shift
dtype: string
- name: competing_priorities
dtype: string
- name: deal_stall_mechanism
dtype: string
- name: decision_driver
dtype: string
- name: decision_blocker
dtype: string
- name: internal_allies
dtype: string
- name: internal_tension
dtype: string
- name: icp_segment
dtype: string
- name: communication_that_resonates
dtype: string
- name: communication_that_backfires
dtype: string
- name: buyer_empathy_narrative
dtype: string
- name: urgency_level
dtype: string
- name: temporal_phase
dtype: string
- name: reasoning_chain
dtype: string
- name: cognitive_framework_citations
dtype: string
- name: confidence_level
dtype: string
- name: validation_basis
dtype: string
- name: source_model
dtype: string
- name: regenerated_at
dtype: string
- name: confidence_score
dtype: float64
- name: validation_notes
dtype: string
splits:
- name: train
num_examples: 1000
---

# Organizational Psychology Under Market Pressure

## Dataset Description

How do people in every organizational role psychologically respond to high-severity market events — the triggers that reshape decision-making at speed? This dataset answers that question with 1,000 structured records at the intersection of **organizational role**, **high-severity market trigger**, and **human psychology**, grounded in 32 validated behavioral science frameworks with full academic citations.

Each record captures how a specific person (CEO, CFO, engineer, sales rep, HR partner) actually thinks, feels, prioritizes, and makes decisions when their organization faces a significant market event (PE acquisition, data breach, layoffs, competitor launch, regulatory investigation).

This is not survey data or self-reported behavior. It is a **behaviorally-grounded taxonomy**: a structured synthesis of organizational psychology research, applied to the specific conditions of modern organizations under market pressure. Every psychological response, cognitive mechanism, and behavioral prediction is traced to peer-reviewed research.

### Why This Dataset Exists

Emotional intelligence (EQ) is a validated predictor of job performance across occupations, particularly for roles requiring emotional labor (Joseph & Newman, 2010, *Journal of Applied Psychology*; O'Boyle, Humphrey, Pollack, Hawver, & Story, 2011, *Journal of Organizational Behavior*), yet no structured dataset exists that maps how real humans psychologically respond to market events by organizational role. AI systems trained on existing B2B data learn *what* happened, not *why* humans made the decisions they did.

This dataset fills that gap by providing the psychological layer: the fears, hopes, cognitive biases, political calculations, and internal narratives that drive organizational decision-making under pressure.

## Composition

### Dimensions

| Dimension | Count | Description |
|-----------|-------|-------------|
| Organizational Roles | 32 | Across 4 layers: executive, senior leadership, middle management, individual contributor. Roles are coded by dominant regulatory focus (`promotion`, `prevention`) or `mixed` where the role's orientation shifts based on context. |
| Market Triggers | 36 | Across 7 categories: financial/filing events (US-centric regulatory references: 10-Q, 10-K, covenant violations), competitive, regulatory, organizational, technological, macroeconomic (incl. tariffs, currency crises), geopolitical (incl. war, sanctions, pandemic, supply chain) |
| Industry Verticals | 7 | B2B SaaS, Fintech, Health tech / digital health, Manufacturing / Industrial IoT, E-commerce, Cybersecurity, Professional services |
| Temporal Phases | 4 | Immediate (day 1-7), short-term (day 8-30), medium-term (day 31-90), long-term (day 91+) |
| Cognitive Frameworks | 32 | Validated behavioral science models with full academic citations |

### Schema (45 fields per record)

Each record contains:

- **Identity**: role, market trigger, industry, company stage, geographic context
- **Psychology**: first-person internal monologue, specific fears and hopes, 3-5 active cognitive mechanisms with citations
- **Emotional cascade**: the full cocktail of simultaneous emotions (guilt, resentment, overwhelm, hypervigilance, etc.) and how each feeds the next, not just isolated fear/hope labels
- **Behavioral impact**: how the emotional state concretely changes buyer engagement across meeting willingness, champion capacity, POC appetite, budget authority, and why
- **Business impact**: factual consequences across CapEx (freezes, redirections), OpEx (headcount, vendor consolidation), Revenue (pipeline compression, churn), and Risk (compliance exposure, audit triggers)
- **Organizational dynamics**: priority shifts, competing priorities, deal stall mechanisms, internal allies/tensions, regulatory focus shifts
- **ICP context**: which ideal customer profile segment this scenario maps to
- **Communication**: what messaging resonates (and why, psychologically), what backfires (and which bias it triggers)
- **Synthesis**: extended empathy narrative, reasoning chain from trigger through psychology to behavior, urgency level
- **Metadata**: confidence level, validation basis, temporal phase, provenance metadata
- **Optional per-record fields**: `confidence_score` (numeric confidence 0.0-1.0 on subset of records where granular scoring was captured; null otherwise), `validation_notes` (free-form validation notes on subset of records where domain-specific validation literature was documented; null otherwise)

### Dataset Composition (Actual Distributions)

| Dimension | Distribution |
|-----------|-------------|
| **Industry** | B2B SaaS 91.1%, Fintech 2.8%, Health tech 1.6%, Manufacturing/IIoT 1.5%, E-commerce 1.3%, Cybersecurity 1.1%, Professional services 0.6% |
| **Temporal Phase** | Immediate 92.6%, Short-term 3.3%, Medium-term 2.4%, Long-term 1.7% |
| **Trigger Severity** | Severity 5 (critical): 44.0%, Severity 4 (high): 53.1%, Severity 3 (moderate): 2.9% |
| **Company Stage** | Growth 88.8%, Pre-IPO 10.2%, Public 1.0% |
| **Role Layer** | Executive 46.9%, Senior leadership 28.7%, Middle management 18.6%, C-suite 3.5%, Individual contributor 2.3% |
| **ICP Segment** | Series B/C scaling 63.1%, Regulated industry AI 19.0%, Enterprise pilot-to-production 9.6%, AI-native SaaS 6.1%, AI infrastructure 1.8%, Professional services AI 0.4% |

**Note on distribution skew:** The dataset is intentionally weighted toward B2B SaaS / immediate-phase / high-severity scenarios because these represent the highest-frequency use cases for sales AI training and organizational crisis-response research. Cross-industry and temporal-variant records provide breadth for transfer-learning and longitudinal-analysis use cases.

### Behavioral Science Foundations

Records are grounded in 32 validated frameworks organized in three layers plus two foundational theories:

**Layer 1, Individual Cognition (10)**: Maslow's Hierarchy of Needs, Cognitive Load Theory, Dual Process Theory (System 1/2), Self-Determination Theory, Approach-Avoidance Motivation, Attribution Theory, Psychological Reactance, Cognitive Dissonance, Social Identity Theory, Yerkes-Dodson Law

**Foundational (2)**: Role Identity Theory, Regulatory Focus Theory

**Layer 2, Interpersonal & Organizational (10)**: Prospect Theory / Loss Aversion, Threat-Rigidity Effect, Emotional Contagion in Organizations, Psychological Safety, Power Distance, Emotional Regulation Strategies, Trust Repair After Violations, Conformity and Dissent, Principles of Influence (Reciprocity), Sunk Cost Fallacy

**Layer 3, Priority & Motivation Shifts (10)**: Goal Setting Theory, Expectancy Theory, Conservation of Resources Theory, Construal Level Theory, Ego Depletion / Decision Fatigue, Psychological Ownership, Meaning-Making Under Adversity, Learned Helplessness, Status Quo Bias, Temporal Discounting

## Uses

### Intended Uses

- **Fine-tuning LLMs** for empathetic, psychologically-grounded understanding of organizational behavior
- **Sales AI training**: Teaching AI systems to understand buyer psychology beyond surface demographics
- **Organizational research**: Structured analysis of how market events cascade through organizations
- **Leadership development**: Understanding psychological dynamics across roles during crisis
- **Crisis communication**: Evidence-based messaging frameworks during organizational disruption

### Out-of-Scope Uses

- Clinical psychological diagnosis or treatment
- Individual person profiling or manipulation
- Replacing professional organizational psychology consultation

## Domain Gap Analysis

We conducted a systematic search across HuggingFace (50+ datasets reviewed), Kaggle (8 search queries), Google Dataset Search, and academic repositories (Papers With Code, Nature Scientific Data, PLOS One, Frontiers, Springer) to identify any existing dataset that maps market events to role-specific organizational psychology grounded in behavioral science frameworks.

**Closest existing datasets and why they fall short:**

- **SCOPE-Persona (Salesforce)**: 1M synthetic personas with sociopsychological facets. Captures personality and demographics, not organizational roles under market pressure. No trigger events, no procurement psychology, no deal behavior impact.
- **Sales-Conversations (goendalf666, Kaggle)**: 3,400 synthetic buyer-seller dialogues. Generic sales scripts without behavioral science grounding. No mapping of market events to role-specific psychology.
- **IBM HR Analytics Attrition (Kaggle)**: Employee attrition prediction. Captures outcomes (people leave) but not the psychological mechanisms (cognitive load, threat-rigidity, loss aversion) driving decision-making during crises.
- **Consensus 2025 B2B Buyer Behavior Report**: 6M B2B interactions tracking emotional influence and buying groups. Tracks buyer journey stages but does NOT map to organizational role × market trigger × behavioral science framework.
- **Nature Scientific Data: Human Decision-Making in Teamwork Management**: 1,144 participants in simulated project management. No market event triggers, no organizational role mapping, no B2B purchasing context.
- **Frontiers: COVID-19 Market Stress & Purchase Behavior**: 1,742 survey respondents confirming market stress changes purchase behavior. No framework mapping roles to behavioral shifts, no structured fields.
- **Kaggle Procurement KPI Dataset**: Procurement metrics without psychological framework or market event context.

**The gap:** Existing datasets are siloed. Organizational psychology repos, consumer behavior archives, B2B analytics platforms, and stress research all operate independently. No public dataset synthesizes organizational role, market event, psychological state, emotional cascade, deal behavior impact, and business impact into structurally related fields in a single record. This dataset is the first to do so with 32 roles × 36 triggers × 32 behavioral science frameworks, each combination captured across 45 structured fields.

## Known Limitations

1. **Synthetic taxonomy, not empirical measurement**: Records represent a structured synthesis of behavioral science applied to organizational scenarios, not observed behavior from real individuals. Predictions should be validated against empirical data before clinical or policy application.
2. **Intentional specialization**: 91% B2B SaaS, 93% immediate-phase, 97% severity 4-5. The distribution is purpose-built for sales AI training and organizational crisis-response research, where these scenarios dominate real-world use. Everyday deal-context psychology (quarterly cycles, champion transitions, vendor consolidation mandates, etc.) is out of scope for v1.0 and planned for v2.0 — see Roadmap below. Cross-industry and temporal-variant records are included for transfer-learning breadth.
3. **English-only in v1.0**: Multilingual versions planned using Adaptive Data credits from Adaption Labs.
4. **Western organizational norms**: Power distance and cultural dimensions reflect primarily North American/European corporate structures. APAC, LATAM, and MENA variants planned.
5. **Role generalization**: Real individuals hold multiple identity positions. The dataset captures modal responses for role archetypes.
6. **Generation pipeline**: 907 records were produced in the primary generation pass using Claude Sonnet 4. 93 records had their `reasoning_chain` field regenerated using Claude Opus 4 to meet a minimum reasoning-chain depth threshold. Regenerated records are tagged with `source_model` and `regenerated_at` for filterable provenance. The 55 `rec_*` records were generated using a matrix design across base scenarios and role/industry combinations to stress-test cross-industry transfer properties.

## Roadmap

This dataset (v1.0) focuses deliberately on **high-severity market events** — the scenarios where organizational decision-making shifts rapidly and where structured psychological data is least available in existing public corpora.

**v2.0 (planned)** will extend the corpus to **everyday deal-context psychology**, adding triggers professionals face weekly rather than yearly:

- Quarterly budget cycle pressure
- Champion transitions (promotion, exit, role change)
- New CFO or CRO joining (leadership change)
- End-of-quarter urgency dynamics
- Vendor consolidation mandates ("reduce SaaS stack by 30%")
- Feature-gap exposure in current tooling
- Multi-vendor RFP dynamics
- "Do more with less" operating mandates
- Champion promoted into decision-maker role
- Hiring slowdown and team capacity constraints

Together, v1.0 (crisis psychology) and v2.0 (everyday deal psychology) will broaden coverage of the psychological states professionals in B2B organizations inhabit across the deal lifecycle.

## Dataset Creation

### Methodology

1. **Role taxonomy development**: 32 organizational roles identified across 4 hierarchical layers, each coded with default regulatory focus (promotion, prevention, or mixed per Higgins, 1997) and domain-specific success metrics.
2. **Trigger taxonomy development**: 36 market triggers across 7 categories, each rated for severity (3-5) based on organizational disruption potential. Includes financial filing events (10-Q, 10-K, covenant violations), competitive dynamics, regulatory mandates, organizational events, technology incidents, macroeconomic shocks (tariffs, currency crises, interest rate shocks), and geopolitical events (armed conflict, sanctions, pandemic, supply chain disruption).
3. **Psychological framework integration**: 32 validated behavioral science frameworks mapped to specific role-trigger interactions, with full academic citations. Each record applies 3-5 frameworks with role-specific manifestation descriptions.
4. **Record synthesis**: Records were generated via structured synthesis using Anthropic's Claude models (Sonnet 4 for the primary pass, Opus 4 for a secondary regeneration pass on 93 records to meet minimum reasoning-chain depth). Each record contains 45 structured fields spanning identity, psychology, emotional cascade, business impact, organizational dynamics, communication guidance, provenance metadata, and optional per-record validation notes. Records include a `source_model` field documenting which LLM generated each entry.
5. **Quality validation**: Records validated against source frameworks for psychological accuracy and internal consistency. Automated audit suite (6 audits, 87+ assertions) verifies numeric claims, schema invariants, AI-tell absence, citation format, and nested structure integrity.

### Source Data

Academic literature from organizational psychology, behavioral economics, social psychology, and decision science. All 80+ citations are peer-reviewed and published in major journals (Econometrica, Administrative Science Quarterly, American Psychologist, Journal of Personality and Social Psychology, Psychological Review, etc.).

Joseph, D. L., & Newman, D. A. (2010). Emotional intelligence: An integrative meta-analysis and cascading model. Journal of Applied Psychology, 95(1), 54-78.

O'Boyle, E. H., Humphrey, R. H., Pollack, J. M., Hawver, T. H., & Story, P. A. (2011). The relation between emotional intelligence and job performance: A meta-analysis. Journal of Organizational Behavior, 32(5), 788-818.

## Citation

If you use this dataset, please cite:

```bibtex
@dataset{chi2026orgpsych,
title={Organizational Psychology Under Market Pressure},
author={Chi, David},
year={2026},
publisher={Hugging Face},
organization={Trimaxion},
license={CC-BY-4.0}
}
```

## Acknowledgments

Dataset created with support from **Adaptive Data by Adaption Labs** as part of the Uncharted Data Challenge.