Brand & Competitive Research / Marken- & Wettbewerbsanalyse

eliaskouloures
4 days ago
7 min read

# BLACK FOREST LABS & FLUX – DEEPRESEARCH PROMPT

## ROLE & MANDATE

Act as a world-class advertising strategist, a hands-on AI product CEO, a competitive-intelligence lead, and a forensic research librarian. Your mission is to produce the most structured, detailed, and comprehensive deep-research analysis of Black Forest Labs (BFL) and its FLUX image generation models, benchmarked against all relevant competitors and adjacent substitutes, and to output a clean "ground-truth" dataset suitable for downstream analysis and prompting.

### Meta-constraints

"**Rigor first:** prioritize verifiable facts over opinions; when uncertain, output “Unknown / Not publicly disclosed” and log the gap."

"**Separation of concerns:** clearly separate Facts (with citations), Interpretation, and Assumptions."

"**Evidence ledger:** every non-trivial claim must carry a source citation and retrieval date."

"**No chain-of-thought exposure:** provide concise Reasoning Briefs (bullet summaries of logic) and all external artefacts (tables, datasets, quotes <= 25 words, code used for parsing), but do not reveal step-by-step internal deliberations."

---

# SCOPE & COVERAGE ✅

## 1. Company Core (BFL)

**Founders, funding, investors, HQ, incorporation details, timeline** (major releases, hires, partnerships, product updates).

**Product & service map** with deep focus on FLUX family (FLUX.1, FLUX 1.1 Pro, Pro Ultra, Kontext, Fill, and any other variants or deployment forms: hosted app, API, on-prem, open weights).

**Business model** (**B2C, B2B, API pricing, enterprise, licensing, integrations**), go-to-market, channels, support model, partner ecosystem.

**Brand reputation** (local 🇩🇪, EU, global), community sentiment, developer traction, enterprise trust signals.

**Risk, compliance & policy** (content safety, watermarking, data usage, IP stance, EU AI Act/GDPR posture).

## 2. Competitive Landscape (Direct & Adjacent)

**Direct T2I/Image tools**: Midjourney, OpenAI (DALL-E / GPT-Image), Adobe Firefly, Ideogram, Stability (Stable Diffusion/SDXL lineage), Leonardo, Playground, Krea, Canva Magic Media, Google (Imagen family), Microsoft Designer (DALL-E powered), Shutterstock & Getty AI, etc.

**Adjacent substitutes / "sources of business"**: full-stack design suites (Adobe CC, Figma), stock marketplaces, photo editors (Photoshop/Generative Fill, Affinity), creative suite AIs, enterprise content platforms, on-device/OEM AI imaging, video/multimodal creation tools that divert budget.

For each: **products, models, licensing, pricing, capabilities, guardrails, distribution, target segments**, and USPs vs. FLUX.

## 3. Customer Profiles & Journeys

**Segments** in AI / Image Gen / T2I / I2I / Manipulation, B2C & B2B.

**JTBDs**, purchase triggers, evaluation criteria, integration needs, budget range, procurement path, champions vs blockers.

**Journey maps** from Awareness → Evaluation → Trial → Adoption → Expansion → Renewal/Churn, with touchpoints, KPIs, friction points, and content/tools needed.

---

# REQUIRED DELIVERABLES 📦

1. Executive Summary (≤ 500 words) — terse, board-ready; 5–10 definitive takeaways; 5 key risks; 5 blue-ocean angles.

2. BFL Deep Dossier (bullet-dense):

* History & Trajectory (dated milestones)

* Product/Service Overview (feature matrix)

* Performance Review (releases, reliability, model quality)

3. Frameworks Pack (each as its own subsection):

* SWOT

* Porter's Five Forces

* BCG Growth-Share Matrix (define segment axes & share proxies)

* Business Model Canvas

* PESTEL

* Balanced Scorecard

* McKinsey 7S

4. FLUX Model Family Deep-Dive

* Version-by-version: purpose, architecture family, inputs/outputs, strengths/limits, editing & inpainting/outpainting, upscaling, fine-tuning/LoRA, controls (ControlNets, IP-Adapter), compositionality, typography, photorealism, style diversity, safety filters, NSFW policy, watermarks, hardware/VRAM needs, throughput, latency.

* Benchmark table (if public) + practical quality proxies (prompt fidelity, text rendering, hands, faces, consistency).

* Pricing & licensing by channel (consumer app, API, enterprise).

* Release Timeline with dates & links.

5. Competitor Landscape

* Feature-parity & USP matrix across all rivals.

* Pricing tables (dated), rate limits, API terms, commercial rights.

* Positioning map (axes: quality vs control; consumer vs enterprise; openness vs closed).

6. Customer Intelligence

* ICP dossiers (B2C & B2B), journey maps, objections, win/loss patterns, playbooks.

5. Competitor Landscape

* Feature-parity & USP matrix across all rivals.

* Pricing tables (dated), rate limits, API terms, commercial rights.

* Positioning map (axes: quality vs control; consumer vs enterprise; openness vs closed).

6. Customer Intelligence

* ICP dossiers (B2C & B2B), journey maps, objections, win/loss patterns, playbooks.

7. Blue-Ocean Strategy Kit

* White-space opportunities, category design angles, wedge GTMs, partnerships, moats (data, distribution, workflows, ecosystem).

8. Ground-Truth Dataset (primary output):

* Provide both CSV and JSON (inline code blocks) following the schemas below.

* Include Data Dictionary and Provenance Ledger (source URL + title + date + claim IDs).

9. Appendices

* Glossary, Methodology, Limitations, Open Questions.

---

DATA SCHEMAS (authoritative) 📦

A. Company Schema (JSON)

{

"company_id": "string",

"name": "string",

"hq_country": "string",

"founded_year": 0,

"founders": ["string"],

"funding_total_usd": 0,

"investors": ["string"],

"business_models": ["B2C","B2B","API","Enterprise","On-prem","Open-weights"],

"products": ["string"],

"brand_notes": "string",

"compliance_notes": "string",

"last_updated_utc": "YYYY-MM-DD"

}

B. Model Schema (T2I/I2I)

{

"model_id": "string",

"company_id": "string",

"name": "string",

"version": "string",

"release_date": "YYYY-MM-DD",

"modalities": ["text-to-image","image-to-image","inpainting","outpainting","upscaling"],

"controls": ["ControlNet","IP-Adapter","Depth","Pose","Sketch","None"],

"fine_tuning": {"supported": true, "methods": ["LoRA","DreamBooth","None"]},

"capabilities": {

"photorealism": "Low/Med/High",

"typography": "Low/Med/High",

"style_diversity": "Low/Med/High",

"prompt_fidelity": "Low/Med/High"

"safety": {"nsfw_policy": "string", "watermarking": "Yes/No/Unknown"},

"perf": {"throughput_img_per_min": null, "latency_sec": null, "vram_gb_min": null},

"pricing": [{"channel":"API|App|Enterprise","unit":"image|credit|seat|token","price":"string"}],

"license_rights": "Personal|Commercial|Enterprise|Restricted",

"notes": "string",

"sources": ["url1","url2"],

"last_updated_utc": "YYYY-MM-DD"

}

C. Pricing Snapshot Schema

{

"product_id": "string",

"date": "YYYY-MM-DD",

"plan_name": "string",

"price_period": "monthly|annual|paygo",

"price_amount": 0,

"currency": "USD|EUR|... ",

"quota": "string",

"rate_limits": "string",

"rights": "string",

"source": "url"

}

D. Customer Profile Schema

{

"segment_id": "string",

"jobs_to_be_done": ["string"],

"evaluation_criteria": ["quality","control","speed","IP/safety","price","integration"],

"budget_range_usd": "string",

"stack_integrations": ["Adobe","Figma","Notion","Slack","GCP","AWS","Azure"],

"key_objections": ["string"],

"success_metrics": ["time_saved","content_quality","cost_per_asset"],

"personas": [{"role":"string","influence":"Low/Med/High"}]

}

E. Journey Event Schema

{

"journey_id": "string",

"segment_id": "string",

"touchpoints": ["Docs","Gallery","Discord","X","YouTube","App","API","Sales"],

"kpis": ["signups","MAU","images/day","API calls","NPS","ARPA"],

"frictions": ["string"],

"content_needed": ["Guides","Benchmarks","Case studies","Templates"]

}

---

# METHOD & QA 🔬

*Collection**: use official docs, pricing pages, TOS/licensing, release notes, engineering blogs, reputable news, conference talks, and community posts (flagged as anecdotal). Record URL + title + publication date + retrieval date.

*Triangulation**: require ≥2 independent sources for critical facts. If not available, mark as Single-source.

*Normalization**: map all competitors to the above schemas; align units (USD, monthly) and date-stamp snapshots.

*Red teaming**: log contradictions; resolve or mark Contested with both views.

*Region & time sensitivity**: capture EU/DE vs US differences (pricing, rights, safety).

*Output hygiene**: deduplicate entities; consistent IDs; human-readable and machine-parsable.

---

# ANALYTICAL FRAMEWORKS (explicit instructions) 🧩

**SWOT**: 6–10 bullets each; tag Fact, Inference, or Assumption.

**Porter's**: rate each force Low/Med/High with 2–3 drivers.

**BCG**: define segments (e.g., Prosumer T2I App, Enterprise API, Open-weights) with relative share proxies (search interest, traffic, API adoption) and growth rates (market reports).

**Business Model Canvas**: one tight table per company (BFL + top 5 rivals).

**PESTEL**: EU-centric with global deltas.

**Balanced Scorecard**: objectives, KPIs, targets, initiatives per quadrant.

**McKinsey 7S**: bullet assessment + alignment risks.

---

# FLUX-SPECIFIC CHECKLIST 🧪

* Versions & variants (e.g., FLUX.1, 1.1 Pro, 1.1 Pro Ultra, Kontext, Fill, etc.) with release dates, intended use, notable improvements.

* Capabilities (photorealism, text rendering, consistency, prompt adherence), controls, fine-tuning, editing, in/out-painting, upscaling.

* Safety & rights (NSFW policy, watermarks, commercial usage).

* Perf & infra (latency, throughput, VRAM, cloud partners).

* Pricing & channels (consumer app, API, enterprise).

* Comparative prompts (standardized prompt set) to evaluate fidelity & style.

---

# OUTPUT FORMAT (strict) 📤

1. Markdown report with the section headers below. Use bold for keywords and one emoji in each H2 title. Only bullet points; no walls of text.

* Executive Summary 🚀

* BFL: History & Trajectory 🗺️

* Products & Services 🧭

* FLUX Family Deep-Dive 🔬

* Performance Review 📈

* Frameworks Pack 🧩 (SWOT, Porter, BCG, BMC, PESTEL, Balanced Scorecard, 7S)

* Competitive Landscape 🥊

* Customer Profiles & Journeys 👥

* Brand Reputation 🌐

* Blue-Ocean Opportunities 🌊

* Risks & Compliance ⚖️

* Ground-Truth Dataset 📊 (CSV + JSON + Data Dictionary + Provenance Ledger)

* Appendices 📎 (Glossary, Methodology, Limitations, Open Questions)

2. Code blocks for:

* `ground_truth.json` (entities: companies, models, pricing, segments, journeys).

* `ground_truth.csv` (wide or tall; include header row).

* `data_dictionary.md` (field names, types, allowed values).

* `provenance_ledger.csv` (claim_id, entity_id, statement, url, pub_date, retrieved_utc).

3. Reasoning Briefs (bullets) at the end of each major section summarizing why the data supports the conclusion (no hidden step-by-step reasoning).

---

# QUALITY BAR & GUARDRAILS 🛡️

*Accuracy**: never guess. Prefer Unknown over speculation.

*Completeness**: if a field is N/A, state Not applicable and why.

*Timeliness**: include "Data current as of: {UTC date}".

*Comparability**: use the same prompts/criteria for model comparisons.

*Reproducibility**: any reader should be able to re-locate each fact via the ledger.

*Ethics/IP**: summarize each tool's IP & content policy as it affects commercial use.

---

# OPTIONAL FILTERS / KNOBS (set if provided)

* `focus_regions` : ["EU","US","APAC"]

* `target_audiences` : ["Prosumer","Agency","Enterprise IT","Creative Ops"]

* `depth_limit_per_competitor` : 200–400 words (analysis) + full dataset rows

* `priority_competitors` : e.g., ["Midjourney","OpenAI","Adobe","Ideogram","Stability","Leonardo","Krea","Playground","Shutterstock","Getty"]

* `omit_adjacent` : true|false

---

# START

1. If any critical inputs are missing (filters/knobs above), proceed with sensible defaults and log the assumption.

2. Produce outputs in the exact order and formats specified.

3. End with a Checklist of Open Questions for follow-up research.

Click to open PDF