Run: 017 · Date: 2026-05-06 · Phase 1 author
For a 3-page About Us PDF aimed at 55-70-year-old founders who built one company and may not be M&A-fluent, the dominant translation risk is not vocabulary. it is voice. Founders this age tolerate two or three technical terms (EBITDA, due diligence, LOI) as competence markers, but they reject any sentence that smells like an advisor brochure ("we help our clients", "trusted partner", "our process delivers"). I am delivering a 14-term replacement table, a 3-term keep-as-is short list, a recommendation that page-3 demo language change "EBITDA band" to "size of company you'd want to buy (in cash flow)" with EBITDA in parentheses on first use, and 5 sentence-pattern traps for the writer to flag.
I do not have the writer's draft yet. Phase 1 means I am pre-flighting against a hypothetical draft, not redlining a real one. If the writer leans heavily on terms I did not anticipate (e.g., "earn-out," "rep & warranty insurance," "drag-along," "266 buyers in our database"), my watchlist misses them and Phase 5 has to catch them cold. I also cannot verify which terms this specific founder cohort actually rejects without survey data. I am reasoning from advisor-brochure pattern recognition, not from interviews with Next Chapter's prior sellers. If quarterback or listener has transcripts of HR.com / Design Precast / Capstone calls where Debbie, Chris, or Ches reacted to specific words, that evidence would beat my heuristic.
Novel run for me. first time tech-translator has been called for an About Us / brand-collateral piece rather than a deal-specific deliverable (CIM, teaser, buyer letter). Prior runs in my memory have been deal-room translations where the audience was a specific seller (Debbie, Chris, Ches) whose vocabulary tolerance I could check against listener transcripts. This run is broadcast-mode: one PDF for many unknown founders. The closest pattern is the writer/draper handoff dynamic. when draper produces vivid copy and writer tightens it, my Phase 5 pass usually catches 4-6 advisor-isms per page. I am pre-loading that catch list into Phase 1 so writer can self-correct and reduce my Phase 5 surface area.
Going forward, when called for broadcast-audience collateral (vs. one-named-buyer deliverables), I will pre-flight Phase 1 with a voice-trap pattern list in addition to the vocabulary table, because advisor-tone leakage is the higher-impact failure mode at this audience age band.
Generated from 017__tech-translator.md — do not edit this HTML directly.