Fundamentals

This week establishes the foundation for everything that follows. Before diving into AI tools and automation, you need to understand your current work patterns and where improvement is actually possible.

The three improvement paths

There are fundamentally three ways to improve how work gets done:

1. Design work to be more intentional and predictable

Most founders operate reactively. Work happens to them rather than being shaped by them. The first lever is simply being more deliberate about what work exists, how it flows, and when it happens.

2. Improve via upskilling

This splits into two categories:

• Soft skills: Communication, delegation, prioritization, time management
• Tech skills: AI usage, automation, tool proficiency

Many founders skip straight to tech skills without addressing soft skill gaps that would give them faster wins.

3. Mindset improvement through a deliberate Founder OS

Your operating system as a founder should match your business stage. Early-stage founders need different habits than those running mature, scaled operations.

Work capacity framework

Work capacity is the personal or external potential to get work done at excellent quality and reliability.

The key insight: different types of work output require different approaches.

Predictable outcomes require rule-based workflows or processes. If you know exactly what needs to happen and the output should be consistent, build a process.

Semi-predictable outcomes are where AI shines. When you need variation, judgment, or handling of edge cases, AI can help - but expect some variability in results.

The mistake most founders make is trying to use AI for predictable work (where simple automation works better) or using rigid processes for work that needs flexibility.

Work classification as a soft skill

Before automating anything, you need to understand what you are actually doing. Work falls into four categories:

Chores: Repetitive, manual tasks that feel like busywork. Checking email, data entry, scheduling. These are prime automation candidates.

Repeatable tasks: Work that follows a pattern and can be systematized. Client onboarding, invoicing, reporting. These need documented processes.

Ad-hoc projects: Emergent, responsive work that cannot be predicted. Crisis response, unique client requests, new opportunities. These need flexible frameworks, not rigid processes.

Broad work: Research, thinking, analysis, strategy. This is where your highest-value contribution lies - and often where founders spend the least time.

The soft skill is learning to classify incoming work quickly and route it appropriately. Even ad-hoc work can be enriched with context and prioritized effectively.

Founder OS by business stage

Your operating approach should match where your business actually is:

Early stage (pre-revenue and early revenue)

Focus on a scrappy loop of experiments with clear structure:

1. Hypothesis: What do you believe will work?
2. Setup: What is the minimum you need to test it?
3. Execution: Run the experiment
4. Evaluate: What did you learn?

Be organized and spot early signs of success. Prioritize ruthlessly - at this stage, it might NOT be tech or AI that matters most.

Mature stage (high revenue, established)

Shift to metrics-driven operations:

• North Star metrics that guide all decisions
• OKRs that cascade through the organization
• Repeatable processes that scale
• Weekly reporting rhythms
• Input metrics, not just output metrics

The danger is trying to operate like a mature company when you are still early stage (creating overhead) or operating like a startup when you have scaled (creating chaos).

Tech skills framework for founders

Your existing tech background shapes how you should approach AI and automation:

No prior tech skills:

• Focus on organizing and documenting first
• Learn to spot patterns before automating
• Delegate technical implementation early
• Your value is in direction, not execution

Some tech skills:

• Decide early if tech is critical to your business
• Recognize what can be automated vs what needs custom solutions
• Be careful not to build when you should buy
• Your value is in smart tool selection

Strong tech skills:

• Watch for tech bias - not every problem needs a technical solution
• Focus on market and traction, not elegant systems
• Spot automation opportunities but do not over-engineer
• Your value is knowing when NOT to build

AI vs automation

These are different tools for different jobs:

Workflows and automation are for predictable, defined work. If you can write a complete decision tree, use automation. It is faster, cheaper, and more reliable.

AI is for semi-predictable outcomes. When you need to handle variation, apply judgment, or work with unstructured data, AI makes sense. But expect some variability.

The practical test: Can you write exact rules for every scenario? Use automation. Do you need something that can adapt? Consider AI.

Why context matters for AI

When you do use AI, the quality of output depends heavily on input quality:

• Fresh, specific data produces better answers than generic prompts
• Cleaned, organized context outperforms raw data dumps
• More relevant context equals better answers

This is why understanding your work patterns (the foundation of this week) matters for effective AI use later. You cannot feed AI good context if you do not understand your own operations.

Key takeaway

Do not jump to AI and automation until you understand what work actually looks like in your business. Classification and intentional design come first. The tools are only as good as your understanding of what they are being applied to.