Hey there,
Here’s the #39 edition of Workflows Weekly, with a few workflows, tools, and ideas to scale your go-to-market systems.
Let's get into it.
1. Featured workflow
Managing replies from automated cold outbound gets messy fast.
Every reply from Instantly, HeyReach, and other tools sits in a separate inbox, and reps sort through each one manually.
The fix: a reply management system that captures every reply, classifies sentiment with AI, and posts it to Slack with one-click actions.
Here's how it works:
Enrich the lead with Clay for firmographics, BetterContact for phone, and OpenAI for a lead summary.
Clean the raw reply text: strip HTML, signatures, footers, and irrelevant URLs.
Classify every reply into Interested, Neutral, Not interested, or Out of office (auto re-enroll OOO back into the campaign).
Push sentiment tags back into Instantly and HeyReach through n8n.
Post the reply, lead summary, and sentiment tag to the right Slack channel.
Add one-click buttons for LinkedIn profile, company site, Instantly/HeyReach conversation, and CRM record.
2. Which AI tool you should look into
We build outbound lists for clients every week and the list quality determines everything downstream: bounce rate, reply rate, positive reply rate.
AI Ark is one of the most reliable databases we pull from because it refreshes every 30 days and verifies every email at export.
AI Ark lets you:
Search 400M+ people and 68M+ companies with data refreshed every 30 days
Get emails validated in real-time at export (only pay for verified ones)
Find lookalike companies and use AI semantic search to build ICP lists
Filter by job roles, skills, or keywords inside LinkedIn profiles and company websites
Enrich records through the API for CRMs, automation tools, and custom integrations
3. What GTM resource we found useful
I just dropped my new YouTube video breaking down how we built Workflows to $250K/month in 12 months.
The video covers the offer, channels, org structure, and the specific plays we ran to get there.
The video walks through:
The offer we sold and how it evolved
The channels that drove the first $250K/month (LinkedIn content, outbound, referrals)
How we structured the team from day one
The exact plays we ran on each channel
The mistakes we made and what we'd do differently
4. AI news of the week
OpenAI just launched GPT-6 Astra, its most intelligent and aligned model yet.
Astra is now the best model out there for computer use, coding, complex math, and cybersecurity work. It solves problems that previous models couldn't.
A few things to know:
Costs $10 per million input tokens and $50 per million output tokens (same as GPT-5.6 Sol)
Best model for computer use: fills forms, updates CRMs, drafts summaries, runs QA checks on websites autonomously
Astra respects the rules you set for it: GPT-5.6 Sol ignored built-in approval requirements 48% of the time, Astra ignored them 0%
OpenAI is also releasing an update to the Codex harness that makes Astra 1.9x faster on real workflows