Three Tests. 15 Documents. 48 Hours. $1,500.

Phase 1 Feasibility Sprint Report

This is what you receive: Field-level validation on 15 of YOUR "worst-case" documents across three rigorous tests. If we hit ≥90% accuracy, you proceed to Phase 2 with confidence. If we don't, full refund—you just saved $7,500+ on a deep dive that wouldn't have worked.

What You Get for $1,500
  • Three Critical Tests on YOUR Documents: We process 15 of your actual "worst-case" documents through three rigorous tests: Field Extraction Accuracy (≥90% target), Straight-Through Processing Rate (≥60% target), and Exception Handling & Safety validation.
  • No Cherry-Picking: You select your most complex, problematic documents—handwritten annotations, poor scans, variable formats. If AI can handle these, it can handle anything.
  • Category-Specific Configuration: Configure extraction fields for 3 different document categories (5 documents each). Each category can have completely different requirements—invoices vs. schedules vs. manifests.
  • 48-Hour Turnaround + Money-Back Guarantee: Complete field-level validation report delivered in 48 hours. If we don't hit ≥90% accuracy, full refund—you just saved thousands on a Phase 2 that wouldn't have worked.

Examples below show industry-specific scenarios. Your actual Phase 1 report will be customized with YOUR documents, YOUR extraction fields, and YOUR ROI calculations based on your staff count and loaded cost rates.

$522k Year-1 Efficiency* (Example)
$1.31M Capacity Upside* (Example)
48hrs Guaranteed Turnaround

*Example: 50-person firm at $140/hr. Your numbers calculated from YOUR data.

The Three Tests: What We Validate

Every Phase 1 sprint validates technical feasibility through three mandatory tests. All three must pass to proceed to Phase 2.

Test 1: Field Extraction Accuracy

Question: Can AI accurately extract the required fields from YOUR document formats?

✅ Success Criteria:
  • ≥90% overall accuracy across all fields
  • Critical fields (IDs, amounts, dates) >95%
  • Low-confidence fields properly flagged
❌ Failure Indicators:
  • <90% accuracy = System unreliable
  • Critical fields consistently missed
  • Document format/complexity too high

Test 2: Straight-Through Processing Rate

Question: What % of documents can be processed with ZERO human intervention?

✅ Success Criteria:
  • ≥60% STP rate (varies by industry)
  • Consistent pattern across categories
  • Remaining docs properly routed to review
❌ Failure Indicators:
  • <40% STP = Too many exceptions
  • Inconsistent across document types
  • Automation ROI insufficient

Test 3: Exception Handling & Safety

Question: Does the system fail safely when encountering ambiguous data?

✅ Success Criteria:
  • Genuine exceptions correctly identified
  • Low false positive rate
  • Safe fail-over to human review
❌ Failure Indicators:
  • High false positive (over-flags valid data)
  • Missed exceptions (risky automation)
  • Inconsistent edge case handling

Pass All Three Tests → Proceed to Phase 2

Failing any single test triggers a full refund. This protects you from investing $7,500+ in a Phase 2 deep dive when the technology isn't ready for your document types.

Sample Size: 15 documents (3 categories × 5 documents)
Document Selection: Your "worst-case" documents (not cherry-picked)
Validation Method: Manual ground truth verification against AI results
Turnaround: 48 hours from document receipt

Interactive Report Preview

Select your industry to see sample validation results. Your actual report will use YOUR documents and YOUR configured extraction fields.

Curam-Ai

Curam-Ai Protocol™

Phase 1 – Feasibility Sprint

Engineering Sector Report

Generated: January 2025 | Sample Client

Assumed Variables

Variable Value
Total Staff50
Documentation Staff (85%)43
Hours/Week per Staff4.5
Loaded Cost Rate$140/hr
Current ProcessManual / Email-based

*Loaded cost includes salary + superannuation + overhead.

Test Results Summary

Test 1: Field Extraction
94.2%
✅ PASS (Target: ≥90%)
Test 2: STP Rate
73%
✅ PASS (Target: ≥60%)
Test 3: Exception Safety
PASS
✅ 2 exceptions properly flagged
✅ RECOMMENDATION: PROCEED TO PHASE 2

All three validation tests passed. System demonstrates reliable field extraction (94.2%), strong automation efficiency (73% STP), and proper exception handling. Technical feasibility confirmed for your document types.

Document Sample: 15 CAD schedules (5 beam, 5 column, 5 mixed)
Processing Time: 47 hours from document receipt
Validation Method: Manual ground truth comparison (field-by-field)

Preliminary ROI Model

Baseline Calculation

Documentation Staff (85% of 50)43
Hours per Staff per Week4.5
Total Weekly Hours193.5
Loaded Cost Rate$140/hr
Working Weeks per Year48

Annual Documentation Cost

193.5 hrs × $140 × 48 weeks$1,306,080

Conservative Automation Opportunity (40%)

Proven low-hanging fruit tasks$522,432

Note: This baseline uses industry-average assumptions. Phase 1 testing on YOUR documents may reveal higher or lower automation potential based on document complexity and format standardization in your specific practice.

Technical Deep Dive: CAD Schedule

Analysis of Drawing S-105 (Roof Beam Schedule) reveals high structured data extraction with proper handling of variable-length members.

Field-Level Validation

Field Extracted Value Confidence Status
Mark B-101 99.8% PASS
Profile 460UB82.1 99.2% PASS
Quantity 4 98.5% PASS
Length VARIES 1200-2400 42.0% FLAG
Grade 300PLUS 97.8% PASS
Remarks Hot Dip Galv, See D5/S-500 88.0% PASS
👷 Engineer's Note:

The system correctly identified that member STR-1 contains a variable length range and flagged it for engineering review. This "Safety Flag" prevents invalid fabrication orders—expected behavior, not an accuracy failure.

Feasibility Verdict: Document S-105 is Passed. The variable-length exception demonstrates proper safety handling.

Ready to Validate YOUR Documents?

What Happens Next

1

You Send 15 Documents

Your worst-case documents across 3 categories. Handwritten notes, poor scans, variable formats—whatever challenges you face.

2

We Configure & Process

Category-specific extraction fields configured. AI processes all documents. Manual ground truth validation performed.

3

48-Hour Report Delivery

Field-level validation report with three test results. Pass all three → proceed to Phase 2. Fail any test → full refund.

Start Your Feasibility Sprint

Money-Back Guarantee: If we don't hit ≥90% field extraction accuracy, full refund. No questions asked.

Curam-Ai

Curam-Ai Protocol™

Phase 1 – Feasibility Sprint

Accounting Sector Report

Generated: January 2025 | Sample Client

Assumed Variables

Variable Value
Total Staff50
Documentation Staff (85%)43
Hours/Week per Staff4.5
Loaded Cost Rate$140/hr
Current ProcessManual / Practice Management System

*Loaded cost includes salary + superannuation + overhead.

Test Results Summary

Test 1: Field Extraction
97.1%
✅ PASS (Target: ≥90%)
Test 2: STP Rate
87%
✅ PASS (Target: ≥60%)
Test 3: Exception Safety
PASS
✅ 1 exception properly flagged
✅ RECOMMENDATION: PROCEED TO PHASE 2

All three validation tests passed. System demonstrates excellent field extraction (97.1%), strong automation efficiency (87% STP), and proper exception handling. Technical feasibility strongly confirmed for structured accounting documents.

Document Sample: 15 vendor invoices (5 standard, 5 multi-line, 5 complex)
Processing Time: 46 hours from document receipt
Validation Method: Manual ground truth comparison (field-by-field)

Preliminary ROI Model

Baseline Calculation

Documentation Staff (85% of 50)43
Hours per Staff per Week4.5
Total Weekly Hours193.5
Loaded Cost Rate$140/hr
Working Weeks per Year48

Annual Documentation Cost

193.5 hrs × $140 × 48 weeks$1,306,080

Conservative Automation Opportunity (40%)

Proven low-hanging fruit tasks$522,432

Note: This baseline uses industry-average assumptions. Phase 1 testing on YOUR documents may reveal higher or lower automation potential based on document complexity and format standardization in your specific practice.

Ready to Validate YOUR Documents?

Start Your Feasibility Sprint

Money-Back Guarantee: If we don't hit ≥90% field extraction accuracy, full refund.

Curam-Ai

Curam-Ai Protocol™

Phase 1 – Feasibility Sprint

Logistics Sector Report

Generated: January 2025 | Sample Client

Assumed Variables

VariableValue
Total Staff50
Documentation Staff (85%)43
Hours/Week per Staff4.5
Loaded Cost Rate$140/hr
Current ProcessManual / TMS Integration

*Loaded cost includes salary + superannuation + overhead.

Test Results Summary

Test 1: Field Extraction
94.8%
✅ PASS (Target: ≥90%)
Test 2: STP Rate
78%
✅ PASS (Target: ≥60%)
Test 3: Exception Safety
PASS
✅ 2 exceptions properly flagged
✅ RECOMMENDATION: PROCEED TO PHASE 2

All three tests passed. System handles bills of lading, customs forms, and freight manifests effectively.

Preliminary ROI Model

Baseline Calculation

Documentation Staff (85% of 50)43
Hours per Staff per Week4.5
Total Weekly Hours193.5
Loaded Cost Rate$140/hr
Working Weeks per Year48

Annual Documentation Cost

193.5 hrs × $140 × 48 weeks$1,306,080

Conservative Automation Opportunity (40%)

Proven low-hanging fruit tasks$522,432

Note: This baseline uses industry-average assumptions. Phase 1 testing on YOUR documents may reveal higher or lower automation potential based on document complexity and format standardization in your specific practice.

Ready to Validate YOUR Documents?

Start Your Feasibility Sprint

Money-Back Guarantee: If we don't hit ≥90% accuracy, full refund.

Curam-Ai

Curam-Ai Protocol™

Phase 1 – Feasibility Sprint

Financial Planning Sector Report

Generated: January 2025 | Sample Client

Assumed Variables

VariableValue
Total Staff50
Documentation Staff (85%)43
Hours/Week per Staff4.5
Loaded Cost Rate$140/hr
Current ProcessManual / CRM Integration

*Loaded cost includes salary + superannuation + overhead.

Test Results Summary

Test 1: Field Extraction
92.3%
✅ PASS (Target: ≥90%)
Test 2: STP Rate
71%
✅ PASS (Target: ≥60%)
Test 3: Exception Safety
PASS
✅ 2 exceptions properly flagged
✅ RECOMMENDATION: PROCEED TO PHASE 2

All tests passed. System handles SOAs, client statements, and compliance documentation effectively.

Preliminary ROI Model

Baseline Calculation

Documentation Staff (85% of 50)43
Hours per Staff per Week4.5
Total Weekly Hours193.5
Loaded Cost Rate$140/hr
Working Weeks per Year48

Annual Documentation Cost

193.5 hrs × $140 × 48 weeks$1,306,080

Conservative Automation Opportunity (40%)

Proven low-hanging fruit tasks$522,432

Note: This baseline uses industry-average assumptions. Phase 1 testing on YOUR documents may reveal higher or lower automation potential based on document complexity and format standardization in your specific practice.

Ready to Validate YOUR Documents?

Start Your Feasibility Sprint

Money-Back Guarantee: If we don't hit ≥90% accuracy, full refund.

Curam-Ai

Curam-Ai Protocol™

Phase 1 – Feasibility Sprint

Insurance Sector Report

Generated: January 2025 | Sample Client

Assumed Variables

VariableValue
Total Staff50
Documentation Staff (85%)43
Hours/Week per Staff4.5
Loaded Cost Rate$140/hr
Current ProcessManual / Claims System

*Loaded cost includes salary + superannuation + overhead.

Test Results Summary

Test 1: Field Extraction
95.2%
✅ PASS (Target: ≥90%)
Test 2: STP Rate
82%
✅ PASS (Target: ≥60%)
Test 3: Exception Safety
PASS
✅ 1 exception properly flagged
✅ RECOMMENDATION: PROCEED TO PHASE 2

All tests passed. System handles claims forms, policy documents, and renewal paperwork effectively.

Preliminary ROI Model

Baseline Calculation

Documentation Staff (85% of 50)43
Hours per Staff per Week4.5
Total Weekly Hours193.5
Loaded Cost Rate$140/hr
Working Weeks per Year48

Annual Documentation Cost

193.5 hrs × $140 × 48 weeks$1,306,080

Conservative Automation Opportunity (40%)

Proven low-hanging fruit tasks$522,432

Note: This baseline uses industry-average assumptions. Phase 1 testing on YOUR documents may reveal higher or lower automation potential based on document complexity and format standardization in your specific practice.

Ready to Validate YOUR Documents?

Start Your Feasibility Sprint

Money-Back Guarantee: If we don't hit ≥90% accuracy, full refund.