The Leverage Ratio: The Only Productivity Metric That Matters When You Work With AI

Why Traditional Productivity Metrics Break Down With AI

Here's the problem: every productivity framework I've ever used assumes that you are doing the work.

Tasks completed? Meaningless when an AI agent completes 40 tasks and you completed zero. Are you productive or idle? Both, somehow.

Hours worked? Irrelevant. I can work 12 hours and produce less than my agent team produces while I sleep. Do the 12 hours count for anything? Not really.

Revenue? Lagging indicator. Revenue this month reflects work done 2-3 months ago. It tells you nothing about whether your current system is working.

Pomodoros, deep work blocks, inbox zero — all of it assumes a world where your output is a direct function of your input. That world is over for anyone seriously using AI.

The question isn't "how much did I produce today?" It's "how much was produced today, and what was my role in it?"

That shift — from measuring personal output to measuring system output relative to personal input — is what the Leverage Ratio captures.

The Leverage Ratio: Definition and Scoring

The formula is simple:

Leverage Ratio = Total Output Value / Gregor Hours Invested

I self-assess this daily on a 1-5x scale. Here's what each score means in practice:

The daily question that drives this: "Did I do this, or did I direct something else to do this?"

That's it. No elaborate tracking system. No time-tracking app. Just one honest question at the end of every day.

My Baseline and What It Revealed

When I started measuring in late March 2026, my baseline was roughly 1.5x. I was stunned.

I had a 10-agent AI team operational. I had automation infrastructure. I had a Mac Mini running scheduled tasks. And my leverage was 1.5x — barely above "doing everything myself."

The reason was obvious once I measured it: I was still doing most things manually out of habit. The agents existed, but I wasn't using them. I'd get an email from a prospect and type a reply instead of having an agent draft it. I'd see a content gap and start writing instead of briefing Prax. I'd check CRM data manually instead of waiting for Holden's revenue brief.

This is the insight the Leverage Ratio forces: having AI tools doesn't make you leveraged. Using them consistently does.

My targets tell the story of where I'm trying to go:

The gap between 1.5x and 3.0x isn't a technology gap. It's a behavior gap. The tools exist. The habit of using them is what's missing.

The "Should I Be Doing This?" Filter

The Leverage Ratio revealed a problem. I needed a system to fix it. So I built a 4-step filter that I run every time I'm about to do something:

1. Can an AI agent do this right now? — Delegate immediately. Don't touch it. Don't even look at it. Brief the agent and move on.
2. Can an AI agent do 80% of this? — Agent drafts, I review. This is the 2-3x leverage zone. Most of my work should live here.
3. Does this require my relationships or judgment? — I do it, but AI preps everything. Discovery calls, strategic decisions, client relationships. These are my Gregor-only activities. But even here, an agent can prep the briefing doc, draft the follow-up email, and update the CRM after.
4. Am I doing this out of habit, not necessity? — Stop. Find the alternative. This is the killer question. I caught myself manually checking Apollo dashboards every morning — something Amos can report on. I caught myself writing LinkedIn posts from scratch — something Dawes can draft. Habit, not necessity.

The filter produces a shrinking list of "Gregor-only activities." Right now that list is: discovery calls, key client relationships, strategic decisions (pricing, partnerships), and vision/direction-setting. Everything else should be flowing through agents.

I keep the filter visible. It's in my weekly planning template. Anna includes it in every W-plan. If I'm doing something that isn't on the Gregor-only list, the system flags it.

Real Examples: 1x vs 3x vs 5x Activities

Abstract frameworks are useless without specifics. Here's what different leverage scores look like in my actual week:

1x Activities (I'm the bottleneck)

• Manually writing a proposal from Apple Notes after a discovery call — 2-3 hours per proposal
• Researching a prospect's tech stack by clicking through their website — 30 minutes per company
• Formatting and scheduling social media posts — 45 minutes per batch
• CRM data entry after calls — 15 minutes per entry, adds up to hours per week

3x Activities (I direct, agents execute)

• "Amos, build a Shopify D2C sequence, value-first angle, $100K+ revenue tier" — I spend 10 minutes briefing, he spends 2 hours building, Bobbie reviews. Total Gregor time: 10 min input + 5 min approval.
• "Anna, draft the W15 plan based on Holden's pipeline data and Alex's research" — I spend 5 minutes scanning the output and making 2 edits. Total Gregor time: 5 minutes for a complete weekly plan.
• "Prax, write the next 3 PostHog frustration keyword articles" — I review the first one for voice (10 min), then Prax self-gates the rest. Total Gregor time: 10 minutes for 3 published articles.

5x Activities (System runs without me)

• Naomi deploys infrastructure updates at 3am via scheduled tasks on the Mac Mini. I find out when it's done.
• Prax publishes SEO content to growthanalyticsengine.com on schedule. I batch-review monthly.
• Holden generates the Monday revenue brief from live CRM data. It's in my inbox when I start the day.

The pattern is clear: 1x activities are me doing execution. 3x activities are me making decisions. 5x activities are me not being involved at all.

Every week, I try to move at least one 1x activity into the 3x column. That's the game.

How to Calculate Your Own Leverage Ratio

You don't need my agent team to use this framework. You need one thing: honesty about how you spend your time.

Step 1: Track for One Week

At the end of each day, write down 3-5 things you accomplished. Next to each one, score it:

• 1x — I did this myself, start to finish
• 2x — AI or an assistant drafted it, I reviewed and finalized
• 3x — I gave a brief, something else executed, I approved the output
• 5x — This happened without my involvement

Step 2: Average It

Take the average score across the week. That's your Leverage Ratio baseline. If you're like me when I started, it's probably between 1.0x and 1.5x — even if you think you're using AI extensively.

Step 3: Identify the 1x Activities

Look at everything you scored 1x. For each one, ask: could this be a 2x or 3x with a different approach? Usually the answer is yes, but you've been doing it manually out of habit.

Step 4: Move One Thing Per Week

Don't try to revolutionize your workflow overnight. Take one 1x activity and figure out how to make it 2x or 3x. Next week, take another one. In two months, you'll have fundamentally changed how you work.

The Metric That Matters Alongside It

I track four supporting dimensions:

• Delegation Rate — percentage of completed tasks done by AI or agents (target: >60%)
• Bottleneck Count — number of tasks blocked waiting on me (target: <3 at any time)
• Pipeline per Gregor-hour — EUR of pipeline moved per hour I invest in sales
• Gregor-only tasks — the list of things only I can do. This list should be shrinking.

The Leverage Ratio is the headline number. These four tell you why the number is what it is.

The Uncomfortable Truth the Leverage Ratio Reveals

Here's what I didn't expect when I started measuring this: the biggest obstacle to higher leverage isn't technology. It's me.

I wrote this in my personal planning document: "Standing in my own way shows up everywhere. Business: doing instead of directing. Wealth: working hard, not accumulating. Health: knowing what to do, not doing it."

The Leverage Ratio made this painfully visible. Every 1x activity is me choosing to do instead of direct. Every time I draft an email instead of briefing an agent, every time I manually check a dashboard instead of waiting for the automated brief, every time I tinker with infrastructure instead of focusing on discovery calls — that's me standing in my own way.

The metric doesn't lie. And it doesn't accept excuses. "I could do it faster myself" — maybe, but that's not the point. The point is freeing your time for work that only you can do. "The agent won't do it as well" — probably true for the first draft, but is your 1x-quality manual work really better than 2x-quality agent work plus the time you save?

Traditional productivity metrics let you hide. You can work 12 hours, complete 47 tasks, and feel productive — while your business stalls because you're doing execution instead of strategy.

The Leverage Ratio doesn't let you hide. It asks one question: did the system produce more because of how you directed it, or did you just do everything yourself again?

That question, asked honestly every day, changes how you work. Not through discipline — through a system that measures whether you're actually leveraged or just busy.