Hey there!
Here’s the seventeenth edition of Workflows Weekly, with a few workflows, tools, and ideas you want to rely on
Let's get into it.
1. Featured workflow
Account prioritization sounds simple until you look at how reps actually decide who to go after.
It's usually a mix of gut feeling and whatever sits at the top of the CRM.
That does not scale.
To fix this, Anabel from our team put together the HubSpot Account Tiering & Scoring Playbook.
Here's what it covers:
Pull your closed won and closed lost deals and find what your best accounts actually have in common.
Define segments, scoring logic, and a tier structure based on your ICP.
Build and enrich the data model in Clay and sync it into HubSpot.
Set up workflows that keep tiers accurate over time.
This way, every rep knows which accounts to go after and why.
2. Which AI tool you should look into
I write a lot every day. LinkedIn posts, emails, Slack messages, prompts inside Clay and Claude.
At some point I realized I was spending more time typing than thinking.
So I started using Wispr Flow.
It's a voice-to-text tool that works inside any app. You just speak, and it writes.
Best parts about it:
Works everywhere; Notion, Gmail, Slack, WhatsApp, Cursor, anything with a text box
Uses context to spell names and terms correctly so you're not fixing things after
It automatically adds structure to an email
Pro Tip: Use it for all your prompts inside Claude Code.
→ Try Wispr Flow
3. What GTM resource we found useful
Dan built an AI-native GTM operating system using Claude Code and GitHub.
The whole system runs on a Company OS repo where your team's context, skills, and workflows live in one place.
He then packaged the entire setup into a starter kit so you can build your own version.
The OS Starter Kit includes:
Guide to getting started on Cursor and Claude Code
Full Company OS blueprint with folder structure and .md guides
Base set of GTM skills: outbound copywriter, LinkedIn post writer, ICP modelling, GTM strategy, discovery prep
GTM engineering workflows plugin
Safety hooks and a checklist to get everything running
Want access?
4. AI news of the week
Anthropic just announced Claude Mythos Preview.
It's a new frontier model, but they're not releasing it to the public
The reason: it's too good at finding and exploiting software vulnerabilities…
A few examples:
A 27-year-old vulnerability in OpenBSD that could remotely crash any machine
A 16-year-old flaw in FFmpeg that automated tools had tested 5 million times without catching
A chain of Linux kernel vulnerabilities that gave full control of a machine from a regular user account
Instead of a public launch, Anthropic created Project Glasswing and giving access to companies like Microsoft, Google, Apple, Cisco, and NVIDIA so they can use Mythos to find and fix vulnerabilities in their own systems.
They're committing $100M in usage credits and $4M in donations to open-source security orgs.
This is the FIRST time since OpenAI withheld GPT-2 in 2019 that a major AI company has held back a model over safety concerns.
That’s a wrap for this week.
What part of your content or GTM workflow feels hardest to manage right now?
Hit reply and let us know. We’ll break it down in a future issue.