The Infinite Garden & Agentic AI Z-Axis Hack... plus some news!

A New Chapter for the Weekly Grind

Before we get into this week’s Coffee Chat, a quick (and bittersweet) announcement:

This will be the final edition of The Weekly Grind as you know it. But don’t worry—we’re not going away. In fact, we’re evolving.

We’re excited to officially relaunch the Frankly IT brand—this time as a LinkedIn newsletter built to spark real conversations, share practical MSP insights, and connect more directly with the community.

I will still be contributing from time to time, but I’m thrilled to introduce Amanda Doucette-Lachappelle as the new voice behind Frankly IT. Amanda brings sharp perspective, deep industry experience, and a fresh energy that’s going to level up everything we’ve been doing here.

Why LinkedIn? We wanted a space that’s easier to find, easier to share, and—most importantly—better for building relationships between readers, authors, and peers across the MSP ecosystem.

Now, back to this week’s Coffee Chat...

👉 Subscribe to Frankly IT on LinkedIn

Coffee Chats Recap: Growing Your ICP Garden with AI: Ben Holley’s Vision for Signal-Based Outreach

What if you could nurture every ideal customer at once, using AI to signal when and where to focus? Ben Holley, Head of Growth at Cyft.AI, breaks down the “Unlimited Garden” concept and shows how it transforms GTM. Find practical tips and how-tos below

(Watch the episode)

🔍 What we learned:

Ben’s “Unlimited Garden” philosophy challenges the traditional GTM approach of chasing endless leads. Instead, Ben emphasizes putting your entire ideal customer profile (ICP) universe into your “garden” (your CRM) while harnessing AI to deeply analyze, filter, and measure engagement signals at scale. By treating your prospect landscape as a living garden, you can nurture every opportunity and know whom to water, when, and why. This way, your effort is focused, relevant, and driven by actionable triggers that matter. As Ben said during our chat, “If you don’t know they exist, they’ll never know you exist.”

💡Biggest Takeaways:

1. CRM as a Living Garden: “You plant these plants, you just focus on these plants and do your best until they die or they bear fruit.” Put every potential ICP in your CRM and let AI help you know when or if to engage.

2. Signal-Based Outreach: AI makes it possible to monitor triggers, like hiring patterns, website tech stacks, or operational changes, so your team always acts on real events rather than generic surface-level cues. “Find those unique signals that maybe you can dig into, not just the obvious ones.”

3. AI Magic Wand: Tech only amplifies what already works. “AI will only amplify what's already working.” Start by refining your processes and outcomes before automating.

4. Less Is More in Lead Gen: Rather than acting like gophers collecting endless leads, identify fewer, higher-fit targets and use AI to engage them at the right moment. “You want leads to convert; too many leads can kill your time.”

✅ Recommendations:

• Get everything in your ICP domain into your CRM, even those you aren’t ready to target yet. Use AI tools like Clay or custom scripts in Cursor to continuously enrich and qualify your ICP with creative filters.

• Build out automatic, signal-based workflows. Consider triggers like staff changes, tech adoption, M&A, or unique site markers such as tech in BuiltWith, changes in portals, or job postings. Let AI surface daily “top opportunities” directly to your reps.

• As Always. Don’t chase AI hype. Focus implementation where it accelerates real business outcomes rather than vanity metrics.

• Experiment with both general AI models like GPT-4 and Claude and real-time discovery models like Perplexity for data gathering and validation.

Huge thanks to Ben for sharing his “Unlimited Garden” vision… a compelling new path for GTM teams to nurture and measure every ICP at scale with AI.

The Secret to Frictionless Agentic AI: Why the Z-Axis Matters More Than You Think

Last week, I wrote about using friction as a third dimension—the Z-axis—in the classic value–effort matrix for evaluating agentic AI use cases.

Traditionally, we’ve mapped tasks based on:

• Value they deliver (Y-axis)

• Effort they take (X-axis)

It’s a helpful framework when prioritizing what’s worth doing next. But when it comes to agentic AI (the kind that acts autonomously and completes real work without human supervision), we need more than a 2D model.

That’s where friction comes in.

Why Friction Matters

Agentic AI is no longer theoretical. It’s already here, and it’s capable of doing real jobs that used to be reserved for people. This isn’t just about augmenting your workflow with tools like ChatGPT. We’re talking about AI that can:

• Pick up a task

• Make decisions

• Deliver outcomes with little to no human involvement

And that’s where the fear creeps in.

Because if the AI can do the job without you… where does that leave you?

The Friction Lens

By layering friction onto the value–effort matrix as a Z-axis, we get a clearer picture of where agentic AI can do the most good with the least resistance.

Instead of automating high-value, high-enjoyment tasks (and potentially stripping away the soul of someone’s role), we can focus on low-friction tasks—things that create drag.

These are the repetitive, tedious, or annoying jobs that still need to get done. If an agent can take those off someone’s plate, that’s a win for everyone.

10 Low-Friction, High-Potential Tasks for Agentic AI

1. Email follow-ups – Chasing people down after meetings or sales calls

2. Status update aggregation – Collecting updates across teams or systems

3. Data entry and cleanup – Fixing broken spreadsheets or cleaning up CRM entries

4. Meeting scheduling & rescheduling – Coordinating calendars across time zones

5. Project handoffs – Creating clean transitions with documentation and next steps

6. Sales journey transitions – Moving leads from one stage to the next with all the right info

7. Report generation – Auto-compiling weekly or monthly reports from raw data

8. Compliance checklists – Running through routine legal or procedural checks

9. Helpdesk triage – Categorizing incoming tickets and assigning them to the right person

10. Reminder nudges – Sending internal pings or prompts for recurring tasks

The Bottom Line

These aren’t flashy use cases, but they’re the kind of things that eat up time and energy—and nobody misses them when they’re gone.

So next time you’re evaluating agentic AI opportunities, don’t just chase value and efficiency. Add the Z-axis. Look for friction.

You’ll not only build smarter systems. You’ll build more human ones, too.

🎙️ I’m Live Next Week

MSP Customer Prospecting & Sales Done Right (Thursday Sept 11th, 9am ET)

Brad brings a wealth of experience transforming how MSPs engage prospects. His insights are actionable and refreshingly no-nonsense, so they’re perfect for anyone who wants to elevate their sales tactics.

• The biggest prospecting mistakes MSPs make and how to fix them

• Non-obvious prospecting channels that get results

• Strategies for moving maybes into yeses while closing deals with integrity

Thursday, September 11th, 2025, at 9:00 AM EST on LinkedIn Live. Come ready to add some serious power to your sales strategy

Comment below or DM me for a reminder, and bring your questions for Brad to tackle!

🎧 Register here 

🔗 Stuff Worth Clicking

• IT departments face huge Windows 10 support bill

• UK government trial of M365 Copilot finds no clear productivity boost

• Smarter Alerts, Better SaaS Insights: Auvik Rolls Out New Platform Upgrades for MSPs

📬 Hit Me Back

Ben Holley gave us a new mental model for GTM this week—nurture everyone, chase no one, and let signals tell you when it’s time to engage. Paired with smarter agentic AI thinking, we’re seeing a shift: from hustle to precision. So tell us:

🌱 How are you planting and tending your “ICP garden”?
📡 What’s one unique signal you track (or wish you could) to drive timely outreach?
🧩 And where are you applying the Z-axis lens—those tedious, low-friction tasks AI can finally take off your plate?

If you're rethinking how you prospect, prioritize, or build agentic systems that don’t steal the joy from your team, we want to hear it. Drop a reply, share your Z-axis wins (or fails), and let’s grow smarter together.

Less noise. More nurture. The garden’s infinite—let’s tend it well.