Stop Doing What Machines Can Do for You

The Five-Minute Task That Eats Your Day

Let’s do some math.

You send a weekly status update email every Friday. Takes about 5 minutes to gather the data, format it, and send it. No big deal, right?

5 minutes per week. 52 weeks per year. That’s 260 minutes – over 4 hours annually on a single email.

Now count all the other “quick” tasks: the follow-up emails, the data entry, the meeting scheduling, the report formatting, the Slack reminders, the invoice processing, the contact list updating.

Those 5-minute tasks add up to weeks of your year. And every one of them is a candidate for automation.

What to Expect

This course is broken into focused, practical lessons. Each one builds on the last, with hands-on exercises and quizzes to lock in what you learn. You can work through the whole course in one sitting or tackle a lesson a day.

What You’ll Learn in This Lesson

By the end of this lesson, you’ll have a framework for identifying which tasks in your daily work are worth automating, understand the true cost of manual repetitive work, and have a prioritized list of automation opportunities to tackle throughout this course.

Your Workflow Audit

Before we can automate anything, we need to find what’s worth automating. Let’s start with a simple exercise.

The Task Diary (do this now or after the lesson):

For one workday, note every task that fits this description:

• You’ve done it before (at least 3 times)
• It follows a predictable pattern
• It doesn’t require creative judgment
• You could explain the steps to someone else in under 2 minutes

Here’s what a typical task diary looks like:

Total: 175 minutes per week. That’s nearly 3 hours – every week – on tasks that follow predictable patterns.

Quick Check

Think about your last two hours of work. Can you identify at least one task that was repetitive, rule-based, and predictable? That’s your first automation candidate.

What Makes a Task Automatable?

Not everything should be automated. Here’s how to tell the difference:

Automate when:

• The task follows consistent rules (“If X, then Y”)
• It happens regularly (daily, weekly, on a trigger)
• It involves moving or transforming data between systems
• Human judgment isn’t required for each instance
• Mistakes are costly or common

Don’t automate when:

• The task requires creative thinking or nuanced judgment
• It happens once and never again
• The rules change constantly and unpredictably
• The relationship value of doing it personally outweighs the time saved
• The task involves sensitive decisions about people

Gray area (automate partially):

• Customer service responses (automate common answers, escalate complex ones)
• Data analysis (automate data gathering, do the analysis yourself)
• Content creation (automate distribution, create the content yourself)

The Automation Value Formula

Not all automatable tasks are worth automating. Prioritize using this formula:

Automation Value = (Time saved per instance x Frequency) - Setup cost

High value: Tasks that take significant time and happen frequently

• Weekly 15-minute report = 13 hours/year saved

Medium value: Tasks that are quick but very frequent, or slow but occasional

• Daily 2-minute check = 8.7 hours/year saved
• Monthly 1-hour process = 12 hours/year saved

Low value: Tasks that are quick and infrequent

• Quarterly 5-minute task = 20 minutes/year saved (probably not worth automating)

But time isn’t everything. Also consider:

• Error reduction: If the manual task produces errors that cause downstream problems, automation’s value is higher than pure time savings
• Consistency: If the task needs to be done the same way every time, automation guarantees that
• Timeliness: If delays in the manual task cause bottlenecks, automation removes the delay
• Mental load: Some tasks aren’t long but they occupy mental space (“I need to remember to send that email”)

Common Automation Opportunities by Role

Operations

• Status report generation and distribution
• Data synchronization between tools
• Invoice processing and reminders
• Inventory tracking and reorder alerts
• Meeting notes distribution

Marketing

• Social media post scheduling
• Email sequence sending (welcome, nurture, follow-up)
• Lead routing based on criteria
• Campaign performance reports
• Content distribution across channels

Sales

• Lead follow-up sequences
• CRM data entry and updates
• Meeting scheduling with prospects
• Deal stage notifications
• Win/loss report generation

Project Management

• Task creation from templates
• Status update collection and compilation
• Deadline reminders and escalations
• Resource allocation notifications
• Sprint report generation

Personal Productivity

• Email triage and auto-labeling
• Calendar management and reminders
• File organization and backup
• Expense tracking and categorization
• Reading list curation

The Rule of Three

Here’s a simple heuristic: if you do something manually more than three times, investigate automation.

The first time you do a task, you’re learning how to do it. The second time, you’re confirming the pattern. The third time, you’re wasting time.

This doesn’t mean every third-time task deserves a full automation. Sometimes a template or checklist is enough. But it means you should at least ask: “Could this be automated?”

AI’s Role in Automation

Throughout this course, you’ll use AI in two ways:

1. Designing automations. AI is excellent at helping you think through workflow logic: “Given these inputs and desired outputs, what steps and conditions do I need?” It can help you map complex processes and identify edge cases you’d miss.

2. Powering automations. AI can be a step within your automation itself: processing text, categorizing data, generating responses, and making simple decisions based on rules you define.

For now, focus on the first role. Before you can automate anything, you need to clearly understand the workflow.

Your First Automation Candidate

Let’s pick one task from your task diary (or from the common opportunities above) and prepare it for automation.

Answer these questions about your candidate task:

1. What triggers it? (What event or condition starts this task?)
2. What are the inputs? (What information do you need?)
3. What are the steps? (Walk through exactly what you do)
4. What’s the output? (What do you produce or change?)
5. Where does it go? (Who receives the output, or what system gets updated?)
6. What could go wrong? (What exceptions or errors have you encountered?)

Example:

Task: Weekly team status update email

1. Trigger: Every Friday at 2 PM
2. Inputs: Task completion data from project management tool, hours logged, blockers
3. Steps: Pull data from PM tool, summarize progress by project, note blockers, format as email, send to distribution list
4. Output: Formatted status email
5. Destination: Team + manager distribution list
6. What could go wrong: PM tool data is incomplete, a project has no updates, someone was added or removed from the team

Those answers are the blueprint for your automation. In Lesson 3, you’ll turn this blueprint into a real design.

Exercise: Build Your Automation Opportunity List

Create your own version of the task diary table. List at least 5 repetitive tasks from your work, estimate the time and frequency for each, and rank them by automation value.

For your top 3, answer the six blueprint questions above. You’ll use these throughout the course.

Key Takeaways

• Repetitive, rule-based tasks are automation candidates – creative, judgment-heavy tasks aren’t
• The Rule of Three: if you do it manually more than three times, investigate automation
• Calculate automation value: (time saved x frequency) minus setup cost
• Don’t forget hidden costs: errors, inconsistency, mental load, and opportunity cost
• AI helps both in designing automations and powering them as processing steps
• Start by documenting your workflows clearly – you can’t automate what you don’t understand

Next lesson: the building blocks of every automation – triggers, actions, and conditions.