4,000 Runs in 7 Days: The Overnight Pattern That Proved Enterprise Demand for Korean Data

Seven days ago, I flipped the monetization switch on 13 Korean web scrapers. The first day brought 19 runs. Today, we crossed 4,000. But the number itself isn't the story. The story is in a 12-hour gap where nothing happened — and what that silence revealed. The Overnight Test Between 6 PM and 6 AM KST on the night of March 18-19, my 14 actors generated exactly 6 runs. Total. Here's what the per-actor delta looked like during that 12-hour window: Actor Runs (18:00) Runs (06:00) Δ naver-news-scraper 1,521 1,521 0 naver-blog-search 391 394 +3 naver-place-search 609 610 +1 naver-blog-reviews 591 592 +1 naver-place-photos 22 23 +1 All others — — 0 Zero. The actor that drives 47% of all my runs produced exactly zero runs between 6 PM and 6 AM. The day before, I'd identified a corporate automation pattern: naver-news-scraper fires at 53 runs/hour during Korean business hours (10-18 KST) and goes silent overnight. The overnight data didn't just confirm this pattern — it confirmed the user typ