How to identify which processes are actually worth automating

Wrong question

"What should we automate?"

Better question: "Which painful processes are actually worth the investment to automate?"

Not everything that's painful should be automated. Some processes are painful because they're broken. If you automate a broken process, you just automate the pain.

The three-axis check

Before you automate anything, check three axes:

Axis 1: ROI

Will this automation pay for itself?

Low ROI (red): No automation.
Medium ROI (yellow): Only if it's simple.
High ROI (green): Automate this.

You calculate ROI the standard way:

(Hours saved per year × hourly cost) / (implementation cost) = payback period in years.

Example: 10 hours saved per week × $25/hour × 52 weeks = $13,000 annual benefit.

If the automation costs $5,000 to build, you pay it back in 5 months. Green. Automate it.

If the automation costs $30,000 to build, you pay it back in 2+ years. Yellow. Only if it's simple.

If it saves 2 hours per year and costs $10,000, don't even think about it.

Axis 2: Complexity

How complicated is the workflow?

Simple (green): Clear rules, predictable inputs, simple logic. Invoice approval that always routes the same way.
Medium (yellow): Some edge cases, some judgment calls, but mostly rules-based. Lead scoring with a few custom fields.
Complex (red): Lots of exceptions, lots of judgment, fuzzy rules. Sales coaching.

Rule of thumb: If a computer can't understand the rules consistently in English, the automation will be expensive and fragile.

Axis 3: Technical stability

Can you automate this without it breaking when you change something else?

Stable (green): The systems it touches rarely change. The data format is consistent. Integration is straightforward.
Unstable (yellow): One of the systems changes frequently or the data format is unpredictable. You'll need to maintain it.
Fragile (red): The workflow depends on systems you don't control, or the data is inherently inconsistent. Automation will break constantly.

Example: Automating invoice approval by pulling data from an API that changes quarterly is fragile. Automating email processing is stable (email hasn't changed format in 20 years).

Reading the matrix

You need all three axes to be green or yellow.

If even one axis is red, don't automate yet. Fix the workflow first.

Green/Green/Green: Automate immediately. High ROI, simple, stable.

Green/Green/Yellow: Automate. Accept ongoing maintenance cost.

Green/Yellow/Green: Automate. Accept increased complexity.

Yellow/Green/Green: Automate only if it's part of a larger wave (justifies more setup).

Red/anything/anything: Stop. Don't automate broken things. Fix the process first, then automate.

How to use this in practice

When someone says "we should automate X," ask three questions:

ROI: "How many hours would this save per year? At what cost to build?" Run the math.

Complexity: "What are the exceptions? Can you write the rules in plain English?" If you can't explain it clearly, it's too complex.

Stability: "If this system changes, will the automation still work?" If the answer is "no," it's fragile.

If all three land green or yellow, you have a candidate. If any lands red, you have pre-work to do.

Do the pre-work first. Fix the broken process. Then automate it.

Administrative:

• Data entry across systems
• Meeting scheduling
• Status reporting
• File organization
• Email management

What You're Looking For

At the end of Week 1, you'll have a raw list of tasks. Now categorize them:

High automation potential (AUTOMATE THESE):

• Repeated frequently (daily/weekly)
• Rules-based (if X happens, do Y)
• Involves multiple systems or data transfers
• Takes significant time
• High frustration scores
• Low judgment required

Medium automation potential (CONSIDER THESE):

• Repeated occasionally (monthly/quarterly)
• Mostly rules-based with some exceptions
• Takes moderate time
• Could benefit from templates or assistance

Low automation potential (KEEP HUMAN):

• Requires judgment and nuance
• Relationship-building activities
• Creative or strategic work
• Unpredictable inputs
• Happens rarely

Step 2: Apply the Automation Criteria Filter

Now that you have your list, evaluate each task against these criteria:

The 6 Automation Criteria

1. Rule-Based Decision Making

Can the task be expressed as a set of clear rules?

✅ Good for automation:

• "If invoice is under $500, auto-approve. If over $500, route to manager."
• "When a new lead submits a form, create a contact, assign to rep based on territory, send welcome email."

❌ Not good for automation:

• "Decide if this customer complaint is worth escalating."
• "Determine if this proposal needs custom pricing."

The test: Can you write it as an "if-this-then-that" flowchart? If yes, automate it.

2. Repetition Frequency

How often does this task occur?

Scoring:

• Multiple times per day: ⭐⭐⭐⭐⭐ (AUTOMATE IMMEDIATELY)
• Daily: ⭐⭐⭐⭐
• Weekly: ⭐⭐⭐
• Monthly: ⭐⭐
• Quarterly or less: ⭐ (low priority)

Why it matters: A task that takes 10 minutes daily = 43 hours per year. A task that takes 2 hours monthly = 24 hours per year.

3. Time Consumption

How much time does this task consume?

Calculate the annual burden:

Annual Hours = (Time per instance) × (Instances per year)

Examples:

• Invoice entry: 10 min × 200/month = 400 hours/year
• Weekly report generation: 2 hours × 52 weeks = 104 hours/year
• Monthly reconciliation: 4 hours × 12 months = 48 hours/year

Prioritization rule: Start with tasks consuming 50+ hours annually.

4. Error Rates & Consequences

Does this task have quality issues?

Red flags for automation:

• High error rates (>2% of the time)
• Errors are costly (financial, customer satisfaction, compliance)
• Errors require significant time to fix
• Inconsistency between different people doing the task

Example: Manual invoice entry with a 3% error rate. Each error takes 30 minutes to discover and fix. On 200 invoices/month:

• Errors per month: 6
• Fix time: 3 hours
• Annual error correction: 36 hours
• Plus: Late fees, vendor relationship damage, accounting headaches

Automation impact: Error rates drop to <0.5%. Massive ROI just from error reduction.

5. Number of System Touch Points

How many different software tools does this process involve?

Scoring:

• 3+ systems: ⭐⭐⭐⭐⭐ (HIGH automation value)
• 2 systems: ⭐⭐⭐
• 1 system: ⭐⭐ (maybe workflow improvements vs integration)

Why it matters: Humans are terrible at being middleware. If your process is "copy from System A to System B," that's exactly what automation excels at.

Examples:

• New customer: Form → CRM → Billing → Project Management → Email
• Order processing: E-commerce → Inventory → Shipping → Customer notification
• Expense approval: Photo → Expense app → Accounting → Manager notification

Each handoff is time spent and an error opportunity. Automation eliminates handoffs.

6. Business Impact

What happens if this process runs faster, more accurately, or without manual intervention?

High-impact indicators:

• Customer-facing (speed = competitive advantage)
• Revenue-generating (faster = more deals)
• Compliance-critical (errors = legal risk)
• Blocks other work (bottleneck elimination)

Examples:

Customer-facing:

• Lead response time (5 min vs 2 hours = 9x conversion difference)
• Order confirmation (instant vs next day = customer confidence)
• Support ticket routing (immediate vs queued = satisfaction)

Revenue-generating:

• Quote generation (same day vs 3-5 days = win rate improvement)
• Proposal follow-up (automatic vs "I forgot" = close rate boost)
• Contract renewal reminders (proactive vs reactive = retention)

Compliance-critical:

• Data retention policies (automated = no violations)
• Audit trail generation (automatic = always ready)
• Security access reviews (scheduled = never miss one)

The Automation Scorecard

For each process, score it on these criteria:

| Process | Rule-Based (0-5) | Frequency (0-5) | Time (0-5) | Errors (0-5) | Systems (0-5) | Impact (0-5) | TOTAL | |---------|------------------|-----------------|------------|--------------|---------------|---------------|-----------| | Invoice entry | 5 | 5 | 4 | 4 | 3 | 3 | 24 | | Lead assignment | 5 | 4 | 2 | 2 | 3 | 5 | 21 | | Weekly reporting | 4 | 2 | 3 | 2 | 2 | 2 | 15 |

Prioritization:

• 20-30 points: Automate immediately (high ROI)
• 12-19 points: Strong candidates (plan for next quarter)
• 6-11 points: Consider (may be worth it for other reasons)
• 0-5 points: Keep manual (automation overhead > benefit)

Step 3: Estimate ROI & Prioritize

Now you have scored processes. Let's calculate the business case.

Simple ROI Formula

Annual Savings = (Hours Saved per Year) × (Loaded Hourly Cost)
Implementation Cost = (Complexity) × ($500 to $5000)
ROI Months = Implementation Cost / (Annual Savings / 12)

Complexity Estimates

Simple automation (2-5 hours, $500-$1500):

• Native integrations between two tools
• Simple form → database workflows
• Email notifications based on triggers
• Basic Zapier/Make workflows

Medium automation (10-20 hours, $2000-$5000):

• Custom API integrations
• Multi-step workflows with conditional logic
• Document generation from templates
• Data synchronization across 3+ systems

Complex automation (30-60 hours, $5000-$15000):

• Custom applications or portals
• AI/ML components (intelligent document processing)
• Complex business logic and exception handling
• Legacy system integration

Example ROI Calculations

Example 1: Invoice Processing Automation

• Current time: 10 min/invoice × 200/month = 400 hrs/year
• Automation reduces to: 1 min/invoice = 40 hrs/year
• Time saved: 360 hrs/year
• Loaded cost: $30/hour
• Annual savings: $10,800
• Implementation: Medium complexity = $3,500
• ROI timeline: 3.9 months

Verdict: DO IT.

Example 2: Monthly Report Generation

• Current time: 3 hours/month = 36 hrs/year
• Automation reduces to: 0 (auto-generated)
• Time saved: 36 hrs/year
• Loaded cost: $40/hour
• Annual savings: $1,440
• Implementation: Simple automation = $1,200
• ROI timeline: 10 months

Verdict: Still worth it, but lower priority.

Example 3: Quarterly Board Report

• Current time: 8 hours/quarter = 32 hrs/year
• Could automate data gathering: Save 5 hrs/quarter = 20 hrs/year
• Time saved: 20 hrs/year
• Loaded cost: $75/hour (executive time)
• Annual savings: $1,500
• Implementation: Medium complexity = $4,000
• ROI timeline: 32 months

Verdict: Probably not worth it (unless you value executive time much higher).

Step 4: Build Your Automation Roadmap

You now have: ✅ A scored list of processes
✅ ROI calculations for each
✅ Implementation complexity estimates

Now create your roadmap.

The 3-Horizon Approach

Horizon 1: Quick Wins (Next 30-90 Days)

Pick 2-3 automations that are:

• High scores (20+ points)
• Fast ROI (<6 months)
• Simple or medium complexity
• High frustration (team will love you)

Goal: Build momentum and prove value.

Example Horizon 1:

1. Automated lead assignment & email sequences
2. Invoice processing workflow
3. Customer onboarding automation

Horizon 2: Strategic Improvements (3-6 Months)

After quick wins prove the value:

• Medium-to-high scores
• ROI 6-12 months
• Medium or complex implementations
• May require process redesign

Goal: Tackle deeper efficiency gains.

Example Horizon 2:

1. Integrated order-to-cash workflow
2. Automated month-end close procedures
3. Customer portal for self-service

Horizon 3: Transformational Changes (6-12 Months)

Long-term, high-impact projects:

• Complex implementations
• ROI 12-24 months
• Require organizational change
• Competitive differentiators

Goal: Fundamentally change how your business operates.

Example Horizon 3:

1. Fully automated customer lifecycle management
2. Intelligent document processing for all contracts
3. Predictive analytics and automated reporting

The Automation Opportunity Worksheet

Here's a practical worksheet you can use today:

Part 1: Process Inventory

List your top 10 time-consuming or frustrating processes:

(etc.)

Part 2: Scoring Matrix

For each process, score 0-5:

| # | Process | Rule-Based | Frequency | Time | Errors | Systems | Impact | Total | |---|---------|------------|-----------|------|--------|---------|--------|-----------| | 1 | | | | | | | | | | 2 | | | | | | | | |

Part 3: ROI Estimation

For top 5 scored processes:

| Process | Annual Hours | Hourly Cost | Annual Savings | Est. Implementation | ROI Months | |---------|-------------|-------------|----------------|---------------------|------------| | | | | | | |

Part 4: Prioritization

My Quick Wins (start here):

My Next Wave (plan for Q2/Q3):

Common Automatable Process Patterns

If you're still stuck identifying opportunities, here are the most common patterns I see:

Pattern 1: "Swivel Chair" Integration

What it looks like:

• Open System A
• Copy data
• Open System B
• Paste data
• Repeat 47 times per day

Examples:

• CRM → Accounting
• E-commerce → Inventory
• Support tickets → Project management
• Form submissions → Database

Automation: Direct integration via API or middleware.

Typical ROI: 2-6 months

Pattern 2: "The Status Update Hamster Wheel"

What it looks like:

• Check status in System A
• Email status update to stakeholders
• Update status in System B
• Add note to System C
• Repeat whenever anyone asks "what's the status?"

Examples:

• Project status reporting
• Order fulfillment updates
• Ticket resolution notifications
• Deal pipeline updates

Automation: Event-driven notifications and dashboards.

Typical ROI: 3-8 months

Pattern 3: "The Document Factory"

What it looks like:

• Receive request
• Open template
• Copy information from email/form
• Fill in blanks
• Save, name, and file document
• Email to recipient

Examples:

• Contracts and agreements
• Proposals and quotes
• Invoices
• Reports
• Onboarding packets

Automation: Template generation from database triggers.

Typical ROI: 2-5 months

Pattern 4: "The Approval Bottleneck"

What it looks like:

• Request sits in someone's email
• Chaser emails: "Did you see my request?"
• Escalation when it's forgotten
• Manual routing: "Actually, send this to [other person]"

Examples:

• Expense approvals
• PTO requests
• Purchase orders
• Contract reviews
• Content publishing

Automation: Workflow routing with auto-escalation and visibility.

Typical ROI: 4-10 months

Pattern 5: "The Data Treasure Hunt"

What it looks like:

• "Let me pull that report for you..."
• Export from System A
• Export from System B
• Combine in Excel
• Clean data
• Create charts
• Send to requestor
• Next week: same request, repeat process

Examples:

• Sales performance reports
• Financial dashboards
• Customer health scores
• Inventory reports
• KPI tracking

Automation: Automated reporting and dashboards.

Typical ROI: 6-12 months (depends on frequency)

Real-World Automation Discovery Examples

Case Study 1: Professional Services Firm

Discovery process:

• 1-week audit revealed 23 potential automations
• Scored and prioritized to top 5
• Estimated combined savings of 780 hours/year

Top 5 they implemented:

1. Client intake forms → CRM → project setup (8-month ROI)
2. Automated proposal generation from templates (4-month ROI)
3. Time tracking → invoicing integration (6-month ROI)
4. Project status updates to clients (10-month ROI)
5. Monthly financial reporting (11-month ROI)

Results after 12 months:

• Actual time saved: 712 hours/year (91% of estimate)
• Annual value: $49,840
• Total implementation: $18,500
• Net first-year savings: $31,340
• Ongoing annual savings: $49,840

Bonus impact: Team satisfaction scores improved dramatically ("I don't feel like a data entry robot anymore").

Case Study 2: E-Commerce Business

Discovery process:

• Founder personally mapped order workflow
• Identified 18 manual handoffs from order to delivery
• Prioritized customer-facing processes

Top 3 they implemented:

1. Order → inventory → shipping integration (3-month ROI)
2. Automated customer notifications (2-month ROI)
3. Returns processing workflow (7-month ROI)

Results:

• Order processing time: 15 min → 90 seconds
• Customer inquiry reduction: 37% (they got proactive updates)
• Same-day shipping rate: 45% → 94%
• Freed 1.5 FTEs for customer service and growth initiatives

Revenue impact: Faster fulfillment and better communication led to 28% improvement in repeat customer rate (worth far more than labor savings).

Your Next Steps: The 30-Day Automation Discovery Sprint

Ready to identify what processes can be automated in your business? Here's your action plan:

Week 1: Audit & Document

• Have team track repetitive tasks
• Focus on time, frequency, frustration
• Collect 50-100 task examples

Week 2: Score & Analyze

• Apply the 6 criteria scorecard
• Calculate ROI for top 10
• Estimate implementation complexity

Week 3: Prioritize & Plan

• Select 2-3 quick wins for Horizon 1
• Map out Horizon 2 and 3
• Get team buy-in

Week 4: Start Implementing

• Tackle your #1 quick win
• Document the before/after
• Measure the results

Ongoing:

• Re-audit quarterly (new processes emerge)
• Continuously add to your automation backlog
• Celebrate wins and share savings with the team

The Bottom Line

Every business has business automation opportunities—you just need a systematic way to find them.

The framework is simple:

1. Map what you're doing
2. Score based on automation criteria
3. Calculate ROI
4. Prioritize and execute

Most businesses discover 15-25 high-ROI automation opportunities within their first audit. The question isn't "what can we automate?"—it's "which of these 25 things do we automate first?"

Start with the framework. Discover your opportunities. Build your roadmap. And stop wasting time on work computers should be doing.

Want help identifying your highest-ROI automation opportunities? Schedule a free automation discovery session and I'll walk you through this framework for your specific business. Or use our ROI calculator to estimate the value of automating your top processes.

Next steps: Check out our guide on automation quick wins you can implement this month to see specific examples and implementation tactics. For plant-floor and shop-floor process discovery, see our manufacturing playbook.