AI summary by TwinMind
65. Stock Photos | The Economics of Everyday Things
Freakonomics Radio Network
23m 7s
Introduction to Stock Photography
- Generic images surround daily life: customer service chatbots, investment brochures, gym advertisements
- Yuri Arkers: CEO/founder of People Images, most prolific stock photographers globally
- Claims: walk 10 minutes in any big city, see 5-6 of his shots; all major brands use their files
- Current database: 600,000 images
- Challenges: shrinking commissions, new competition making money harder
Historical Evolution of Stock Photography
- 120+ years ago: newspapers/magazines sent photographers for generic shots
- Mid-20th century: companies created photo archives for licensing fees
- Ordering process: big catalogs, page-by-page selection, call with file numbers (e.g., XY 2467), receive film for scanning
- 1990s: physical photos transferred to computer servers, CDs with pre-loaded images
- Elite photographers made massive money at very high prices with rights management
- Additional fees for size changes, brochure usage - "very tightly managed and incredibly tedious"
Microstock Revolution (2000s)
- Websites allowed anyone to upload/sell photographs
- Price shift: $1 per download vs $600-$1,200 previously
- Opened door for amateur photographers like Arkers
- Arkers' background: psychology student in Denmark, started 2006 on Shutterstock
- Revenue progression: $500 → $44,000 → $112,000 monthly within 2 years
- Peak: $40,000/month, more than potential psychologist salary
Current Operations
- 20,000 sq ft production studio, Cape Town, South Africa
- 3-4 shoots daily, 100+ employees (photographers, stylists, editors, lighting crew)
- Archive: 600,000 stock photos, 250,000 stock videos
- Distribution platforms: Adobe Stock, Shutterstock, Dreamstime, 123RF, iStock, Deposit Photos
- Search volume examples: "marriage" = 3.4M results, "sunset" = 14M+ results
Business Model and Economics
- Customer payment: monthly subscription for set downloads OR individual purchases
- Standard license: broad worldwide rights forever
- Photographer royalties: 15-40% depending on annual sales volume
- Tier systems designed to minimize payouts
- Most photographers earn ~$0.30 per license
- Arkers (top tier): ~$1 average payout, 10,000+ daily sales
- Portfolio performance: 10% of images generate 90% of sales
- Return Per Image (RPI): revenue per image monthly across entire portfolio
- High RPI: $1 per asset/month; average: cents to dollar
Content Strategy and Trends
- Can't upload random shots; must predict what sells
- Trend monitoring essential: Elon Musk-Trump handshake → politics shots
- Visualization of abstract concepts crucial
- Factory example: staff lineup, detail shots, machinery focus
- Best-selling asset: woman taking running break, hands on knees, blue sky, hopeful gaze
- Multiple applications: bank commercials, insurance, vacation, fitness
Production Process
- 2+ days research per concept series
- Realism priority: real code on monitors, actual bands for concerts, genuine presentations
- Shoot planning: strict schedules due to expense
- Recent office shoot: $5,000 cost, 106 images produced
- Crew requirements: location, lights, catering, 4-6 models, producer, focus puller, camera operator
- Break-even timeline: 2-3 years typically
- Industry entry barrier: requires substantial loan
Model Selection and Casting
- Traditional agencies provide wrong model types (high fashion/runway)
- Skinny models problematic - "not good asset"
- "Weekend test": would you spend weekend on stranded island with this person?
- Trustworthy appearance sells better than conventional beauty
- Demand for all ethnicities, races, sizes, ages
- Senior models "sell extremely well"
- Casting methods: social media calls, studio walk-ins, street scouting
- Farmers markets: "amazing really stylish trendy healthy people"
- Payment: day rates (no royalties), occasional celebrity status
Model Experiences and Issues
- Models don't know final usage locations
- Example problem: German HIV awareness campaign using "sickly" looking model
- Legal issues require takedown notices, lawyer letters
- Viral success: 2009 "Good Girl Gina" meme (Emma Catherine)
- Meme usage typically unlicensed but not worth pursuing legally
SEO and Discoverability
- Indexing crucial: specific keywords for search visibility
- Team of indexers attach extremely specific/abstract keywords
- Examples: flower identification requiring botany research, weapon specifications
- Specific keywords better than common ones due to less competition
- Keyword accuracy essential for surfacing in search results
Industry Challenges
- Free stock photo websites disrupting traditional model
- Business model: "disrupt enough to force buyout"
- Limited quality behind attractive front pages
- AI threat: Dolly, Mid Journey generating realistic images
- Nvidia, OpenAI, Meta partnerships with stock libraries
- AI limitations: 200-300 generations for one shot (~2.5 hours work)
- Professional precision changes difficult for AI
- Current assessment: not refined enough for finicky art designers
- Income hasn't decreased due to AI yet
Personal Satisfaction
- Joy of seeing work in the wild
- People's reactions: "look at me like I was some kind of Looney Tunes"
- Finds the recognition "pretty hilarious sometimes"
