Feasibility Factors in AI Prioritization
Explore the key factors that shape AI feasibility and learn how to assess practicality, improve adoption, and boost success by breaking big processes into manageable AI pilots.
Feasibility factors determine how realistic and straightforward it will be to transition a workstream into an AI-driven process. Understanding feasibility ensures you prioritize initiatives your firm can genuinely and smoothly implement, avoiding unnecessary complexity and delays.
1. Systems Involved
Evaluate how many and how complex the systems involved are. Fewer and simpler systems mean higher feasibility, whereas complex integrations can significantly complicate implementation.
Example:
• High Feasibility: Automating simple expense reporting using one or two standard applications.
• Low Feasibility: Integrating multiple legacy systems for complex audit processes.
2. Technical Complexity
This factor assesses how challenging it is to automate a workstream based on human judgment or interpretation required. Processes requiring minimal human judgment are more easily automated.
Example:
• High Feasibility: Automating straightforward categorization of routine financial transactions.
• Low Feasibility: Complex advisory tasks that require nuanced human decision-making.
3. Data Complexity
Data complexity covers the accessibility, structure, and quality of data needed for automation. Clean, structured data simplifies automation significantly.
Example:
• High Feasibility: Standardized data formats for invoice processing.
• Low Feasibility: Analyzing handwritten notes or complex, unstructured client emails.
4. Time to Implement
Estimate how long it will realistically take to complete a workstream's automation. Shorter implementation periods make for higher feasibility.
Example:
• High Feasibility: Automating straightforward internal workflows within weeks.
• Low Feasibility: Comprehensive firm-wide system upgrades taking months or longer.
5. Resources Required
Assess how much time, money, and human resources the transition demands. Less resource-intensive workstreams naturally rank higher in feasibility.
Example:
• High Feasibility: Small-scale improvements manageable by an individual or small team.
• Low Feasibility: Major initiatives needing significant cross-firm investment and effort.
Feasibility is Relative
Just like impact, feasibility must be seen within your firm's specific context. What seems straightforward in isolation may become complicated within larger processes or systems. For example, scheduling meetings individually appears easy until considering integration across numerous departmental calendars, complicating the implementation significantly.
Enhancing Feasibility with Granularity
If a high-impact workstream has low feasibility, consider increasing its granularity by breaking it down into smaller, more manageable sub-streams. Granularity can significantly improve feasibility, making it easier to identify segments ready for AI.
Practical Tip: Increase granularity by splitting complex processes like month-end close into smaller tasks such as expense management, payroll, and inventory management.
