Writing on PR, journalism, and the craft of pitching.
Long reads and practical research on PR, journalist discovery, media outreach, pitching, and building better media lists with Medialyst. New posts most weeks, with data cuts and operator notes from inside Medialyst.
33 posts
AI
Showing 12 of 33Sorted by date, newest first
PR
Introducing Multi-Angle Media Lists and Beat Sweep Agents
Medialyst now splits your news into distinct story angles and runs a Beat Sweep Agent on each one. The Carly Score jumped 42.2 to 51.3.
PR
Medialyst Database Quality: Carly Score Week 2, 33.7 → 42.2
Medialyst database quality, graded in public: the Carly Score went from 33.7 to 42.2 after we overhauled the entire media database.
PR
How to Audit a Pitch With 37 Claude Subagents
A practical method for pressure-testing pitches, press releases, and launch announcements with isolated Claude subagents, plus the limits of synthetic personas.
PR
Carly Martinetti Joins Medialyst as Founding Strategic Advisor
Carly Martinetti joins Medialyst as founding strategic advisor. Elvis is embedding inside Notably to build and test an AI-native PR workflow end to end.
PR
Pricing update — effective August 1, 2026
Subscribe before 11:59pm ET on July 31 to lock in up to 20% more credits every month. Plan prices don't change, nothing costs more, and 1 credit now buys 1 researched journalist.
PR
How to turn Claude into a media desk: the 11 integrations that run a full PR workflow
Connect Claude to your actual PR stack via MCP: research angles with Perplexity, find journalists with Medialyst, run follow-ups in Gmail, and prove ROI with Search Console - all in one window.
PR
How to Install Newsjack and Turn Claude Into a PR Associate That Never Sleeps
Step-by-step walkthrough of installing Newsjack's 18 open-source PR skills in Claude, getting verified journalist emails, and setting up a 24/7 news monitor with Claude Code.
Eval
Claude Fable 5 vs Opus 4.8: Which Model Writes Better PR Story Angles? (50-Brand Blind Eval)
We ran 50 real brands through Claude Fable 5 and Opus 4.8, had GPT-5.5 judge every pair blind in both orderings. Fable won 67 of 100 judgments and led 6 of 7 quality dimensions — but the position-bias finding is the part to take home.
PR
Ex-TechCrunch: How she sends pitches now
Rita Liao explains how she uses Medialyst and Simular's Sai to research journalists faster while keeping judgment, writing, and relationship-building human.
MTMedialyst TeamApr 15, 20264 min read
PR
HARO Is Dying: The Data, The Reasons, and What's Replacing It
Journalist request platforms have lost 50-64% of their activity since 2024. Here's why, and what's replacing HARO.
Product
How We're Building the World's First Media Database powered by AI-agents
A journalist database that's always fresh—because AI agents can enrich live