How to Assess Financial Viability as an AI Product Manager
What I Learned After Over 100 Assessments - ROI, NPV, IRR—OMG 😱 From Facepalms to High Fives 🤦♂️🙌 A comprehensive Guide.
If only I had kept track of all the AI use case assessments I’ve conducted over the past decade, I could share some impressive statistics and comparisons over time. While I haven’t done that, I can tell you one thing in detail:
In the past six months alone,
- more than in all the previous years combined.
This surge is primarily due to one phenomenon: the hype around GenAI.
This trend highlights just how powerful GenAI is, significantly increasing the number of potential applications and new opportunities for companies.
Now that AI use cases are popping up everywhere, companies and AI Product Managers must swiftly determine which ones are worth pursuing. Otherwise, the risk of allocating resources to the wrong initiatives could result in wasted time, money, and effort.
Some of you might already know that I strongly advocate for "assessing the core before opening the door," adhering to the well-known desirability, feasibility, and viability assessments. This approach ensures that we focus on projects that align with the strategic goals and capabilities of the organization.
It’s more important than ever to stay grounded and methodical. The allure of GenAI can lead to a scattergun approach, quickly enabling AI use cases without a deep understanding of their overall consequences. However, disciplined assessment helps us identify the real game-changers from the mere AI use cases.
Today, this post is all about viability from a financial perspective, focusing on the potential return on investment. It may not be the most exciting and joyful topic and might feel like a pain at times 🤕, but do yourself a favor and understand the core principles, as this is where your potential impact as an AI Product Manager will be determined 🔥.
You will learn:
What is Financial Viability?
How Finance Departments would assess an AI Investment
Adapting Financial Assessment for AI Product Managers
Step-by-Step Example to Assess an AI Investment
Final Thoughts
Happy reading 🛋️
Before starting: If you’ve found my posts valuable, consider supporting my work. You can help by sharing, liking, and commenting on my LinkedIn posts introducing this issue. This helps me reach more people on this journey.
Thank you for supporting and motivating me to continue sharing my experience with you. Thanks for being part of this community ❤️.
#beyondAI
Every AI Product Manager needs to understand how to assess the financial viability of a proposed AI use case. In companies, potential AI use cases are proposed either by business teams to the AI team or vice versa. Every single idea is worth at least being heard.
However, when it comes to running a cost-benefit analysis and determining financial viability, AI teams and their Product Managers aren't domain experts. They often struggle to understand the overall implications of an AI solution on business processes and the KPIs it impacts, making it challenging to get the full picture.
Similarly, business teams may not fully grasp the financial implications of allocating people and resources to build an AI solution for their problem.
However, both perspectives are essential for a comprehensive analysis. So, no matter how sophisticated your formula is for calculating potential financial viability, your estimate will always be poor if done in isolation. You need to collaborate with business teams, and they need to be cooperative to come to a sound estimation of the financial impact.
Everything else is just half-baked.
As an AI Product Manager, it’s important to be transparent from the beginning with your business colleagues. Make it clear that you will need their support to get the numbers right and to justify the investment.
Having said this, let’s dive deeper into some financial concepts and terminology. We'll also fully derive the formula for a cost-benefit analysis to understand if investing in an AI solution for a proposed problem is financially sound.
What is Financial Viability?
Financial viability refers to the ability of a project or business to generate enough income to cover its costs and achieve a reasonable profit. In the context of an AI use case, financial viability means that the proposed AI solution should provide more financial benefits than it costs to develop and implement.
To determine financial viability, you need to consider various factors:
Costs: This includes all expenses associated with developing, implementing, and maintaining the AI solution.
Benefits: These are the financial gains expected from the AI solution. Benefits might include increased revenue, cost savings, improved efficiency, or other positive impacts on the business.
Return on Investment (ROI): This is a key metric that compares the benefits to the costs. A positive ROI means that the benefits outweigh the costs, indicating that the project is financially viable.
By assessing these factors, you can get a clearer picture of whether an AI project is worth pursuing from a financial perspective. If you have already read my post "AI Initiatives are Investments. Act like it," you might be curious about how finance departments decide when to invest and when not. We can learn a lot from their approach, so let’s delve into that.
How Finance Departments would Calculate the AI Investment Case
Finance departments use a structured approach to evaluate the potential return on investment (ROI) and ensure that the project aligns with the company's financial goals. Here’s how they typically calculate the AI investment case:
Cost Assessment: Finance teams meticulously identify and document all costs associated with the AI project. This includes direct costs like software and hardware, as well as indirect costs such as personnel, training, and ongoing maintenance.
Revenue Projections: They estimate the potential revenue generated by the AI solution. This involves analyzing how the AI project will impact sales, customer retention, market expansion, and other revenue-driving activities.
Cost Savings: Finance departments evaluate how the AI solution can reduce operational costs. This might include savings from automation, improved efficiency, reduced error rates, and lower labor costs.
Capital Budgeting Techniques: Finance departments use capital budgeting techniques such as Net Present Value (NPV), Internal Rate of Return (IRR), Return on Investment (ROI), and Payback Period to assess the investment. These techniques consider the time value of money and help in comparing the AI project against other potential investments.
Net Present Value (NPV): NPV calculates the present value of cash inflows and outflows over the project's life. A positive NPV indicates that the project is expected to generate more value than it costs.
Internal Rate of Return (IRR): IRR is the discount rate that makes the NPV of cash flows equal to zero. It represents the project's expected rate of return.
Return on Investment (ROI): ROI measures the profitability of an investment as a percentage of the initial cost. It gives a quick snapshot of the efficiency of the investment by comparing the net profit to the initial investment.
Payback Period: This is the time it takes for the project to generate enough cash flow to recover the initial investment. A shorter payback period is generally preferred as it indicates quicker recovery of the invested capital.
Risk Analysis: Finance departments conduct a risk analysis to understand potential uncertainties and their impact on the project. This involves scenario analysis and sensitivity analysis to see how changes in key variables affect the project's financial outcomes.
Stakeholder Consultation: They collaborate with other departments, in this case, it would be the AI and business teams, to validate assumptions and ensure that all relevant factors are considered. This ensures a holistic view of the project's financial viability.
Financial Reporting: Finance teams prepare detailed reports and presentations to communicate their findings to senior management and stakeholders. These reports include key metrics, financial projections, and risk assessments.
By following this structured approach, finance departments ensure that AI investments are thoroughly evaluated and aligned with the company's strategic and financial objectives. This collaborative process helps in making informed decisions and justifying the investment to all stakeholders.
Now, let’s see how this would look if we, as AI Product Managers, were to run such a calculation on our own. But first, let’s understand why we need three different metrics: ROI, NPV, and IRR to assess an AI investment.
Why isn’t ROI sufficient?
Each metric provides unique insights and has its limitations. It's essential to consider all of them together to get a well-rounded view (if that’s the aim) of the investment's financial viability. Here’s why:
Return on Investment (ROI)
What it shows: ROI measures the profitability of an investment as a percentage of the initial cost.
Why it's important: It gives a quick snapshot of the efficiency of the investment.
Limitations: ROI does not account for the time value of money or the duration of the investment.
Net Present Value (NPV)
What it shows: NPV calculates the present value of future cash flows minus the initial investment. It accounts for the time value of money by discounting future cash flows.
Why it's important: NPV provides a direct measure of the added value from the investment, considering both the magnitude and the timing of cash flows.
Limitations: NPV requires an accurate discount rate and can be sensitive to changes in this rate.
Internal Rate of Return (IRR)
What it shows: IRR is the discount rate at which the NPV of an investment is zero. It represents the expected rate of return.
Why it's important: IRR allows for comparison between projects with different sizes and durations, showing the efficiency of an investment.
Limitations: IRR assumes that future cash flows are reinvested at the IRR, which may not be realistic. It can also give multiple values for projects with alternating cash flows.
Why Consider All Metrics Together
Comprehensive Evaluation: Each metric highlights different aspects of the investment. Using them together provides a more complete picture.
Cross-Verification: Comparing NPV, IRR, and ROI helps cross-verify the attractiveness of an investment. For example, a high IRR should ideally correspond to a positive NPV.
Different Perspectives: NPV focuses on value addition, ROI on efficiency, and IRR on the rate of return. Together, they cover value, cost-effectiveness, and potential growth.
Risk Assessment: Combining these metrics helps assess risk better. NPV shows the value at risk, IRR the efficiency under different scenarios, and ROI the basic profitability.
While each metric has its strengths, relying on only one can lead to an incomplete or misleading assessment. For a robust evaluation, consider NPV, ROI, and IRR together to understand the full financial implications of an AI investment. This comprehensive approach is how finance departments typically evaluate investments.
🤔 But do we as AI Product Managers need to approach it like this?
Adapting Financial Assessment for AI Product Managers
As AI Product Managers, our role involves making quick and informed decisions to prioritize and advance AI initiatives. While adopting the thorough approach of finance departments can be beneficial, there are practical considerations to keep in mind:
Speed and Efficiency: AI Product Managers often need to make swift decisions. Simplified assessments using just one or two metrics (like ROI or NPV) can be sufficient for initial evaluations and prioritization.
Resource Constraints: AI teams may not always have the tools or expertise to conduct detailed financial analyses. Using basic metrics that are easier to calculate and understand can help in making timely decisions.
Iterative Process: AI projects often evolve, and initial financial assessments might need adjustments. Starting with simpler calculations allows for flexibility and quick iterations based on new data or insights.
Collaborative Approach: For critical decisions, AI Product Managers can collaborate with finance departments to leverage their expertise in conducting detailed analyses. This ensures a balanced approach without overburdening the AI team.
Practical Approach
Initial Screening with ROI: Use ROI to quickly gauge the profitability of potential AI use cases. This helps in filtering out less promising projects early on.
Deeper Analysis with NPV: For shortlisted projects, calculate NPV to understand the value addition considering the time value of money.
Consult Finance for IRR: For high-stakes projects, involve finance departments to calculate IRR and conduct comprehensive risk assessments.
Example: Assessing the Financial Viability of an AI Investment
⚠️ This is a simplified version of an assessment. Finance departments will conduct these evaluations in more detail and may use different assumptions. All figures and assumptions here are fictitious and used for demonstration purposes.
A company is considering investing in an AI solution to automate its customer service operations. The initial development cost is estimated at $500,000, and the annual maintenance cost is $100,000. The AI solution is expected to generate cost savings of $250,000 per year by reducing labor costs and improving efficiency. The project is expected to have a lifespan of 5 years.
Step-by-Step Calculation
Identify Costs
Initial Development Cost: $500,000
Annual Maintenance Cost: $100,000
Total Costs over 5 years:
\(Total Costs = $500,000 + $100,000 \times 5 = $1,000,000\)
Estimate Benefits
Annual Cost Savings: $250,000
Total Benefits over 5 years:
\(\text{Total Benefits} = \$250,000 \times 5 = \$1,250,000\)
Calculate Net Benefits
Net Benefits:
\(\text{Net Benefits} = \$1,250,000 - \$1,000,000 = \$250,000\)
Calculate ROI
ROI:
\(\text{ROI} = \left( \frac{\text{Net Benefits}}{\text{Total Costs}} \right) \times 100 \)\( \text{ROI} = \left( \frac{\$250,000}{\$1,000,000} \right) \times 100 = 25\% \)
Consider the Time Frame
Since the project spans 5 years, we also consider the annual ROI:
\(\text{Annual ROI} = \frac{25\%}{5} = 5\% \)
Calculate NPV
To calculate the NPV, we need to make some assumptions about the discount rate. The discount rate is the interest rate used to determine the present value of future cash flows. It reflects how much future money is worth today. For our example, let's use a discount rate of 10%. The NPV is calculated as follows:
\(\text{NPV} = \sum \left( \frac{\text{Annual Savings}}{(1 + \text{Discount Rate})^t} \right) - \text{Initial Cost} \)\(\text{NPV} = \left( \frac{\$250,000}{(1 + 0.1)^1} + \frac{\$250,000}{(1 + 0.1)^2} + \frac{\$250,000}{(1 + 0.1)^3} + \frac{\$250,000}{(1 + 0.1)^4} + \frac{\$250,000}{(1 + 0.1)^5} \right) - \$500,000 \)\( \text{NPV} = \$227,273 + \$206,611 + \$187,828 + \$170,753 + \$155,230 - \$500,000 = \$447,695 \)The discount rate is important because it helps us account for the fact that money today is worth more than the same amount in the future due to factors like inflation and risk. A higher discount rate means we value future cash flows less because of higher risk or uncertainty, while a lower discount rate means we value them more. Choosing the right discount rate is key to accurately assessing the NPV and understanding the financial viability of the project.
Why Use a 10% Discount Rate?
Using a 10% discount rate is a common practice for several reasons: Standard Industry Practice: Many companies and industries use a 10% discount rate as a standard benchmark for evaluating investments. It provides a consistent basis for comparison. Risk and Opportunity Cost: A 10% rate reflects a reasonable estimate of the opportunity cost of capital, which is the return that could be earned on an alternative investment of similar risk. Inflation: It takes into account the average rate of inflation over time, ensuring that future cash flows are adjusted for the decreasing purchasing power of money. Business Risk: For many businesses, a 10% rate appropriately reflects the risk associated with future cash flows, balancing the potential uncertainties and rewards.
Calculate IRR
For simplicity, we use financial software or a calculator to find that the IRR for this project is approximately 16%. You can use online tools or the =IRR() function in Excel. However, you will soon understand why I will not explain this further in detail 😉.
Calculate Payback Period
Payback Period is the time taken to recover the initial investment from the annual savings:
\(\text{Payback Period} = \frac{\$500,000}{\$250,000} = 2 \text{ years}\)
Summary
Total Costs: $1,000,000
Total Benefits: $1,250,000
Net Benefits: $250,000
ROI: 25%
NPV: $447,695
IRR: 16%
Payback Period: 2 years
✅ This example demonstrates that the AI investment is financially viable, with a positive NPV, a solid ROI of 25%, a favorable IRR of 16%, and a quick payback period of 2 years. These calculations help the finance department and other stakeholders make informed decisions about investing in the AI solution.
Final Thoughts
One might ask, if finance departments are conducting these assessments, why should an AI Product Manager do the work as well? Yeah, I know, it was also my first question in my early days as an AI Product Manager. But…
Not all AI teams need to go through the finance department approval process, as their teams are already pre-funded and just need to find AI use cases and implement solutions. For these teams, it is mandatory to understand how to assess financial viability on their own and prioritize accordingly.
Other teams do need to go through the approval process, but imagine having numerous AI use cases, each requiring approval. Typically, finance departments are not solely focused on assessing AI investments but handle all types of company investments. This means their backlog is already full. Thus, the approval process might take some time. And do we have time?
Surely, we don’t.
We never have time 🙂
So, why not request financial assessments only for those cases we know are good investments?
It’s not a difficult calculation, but it's one that I rarely see being done by AI teams. I regularly run these kinds of quick assessments, which help me determine exactly which cases make the most sense to focus on. Sometimes, I focus on the ROI, while other times I look at the NPV to quickly make an evaluation. If neither of these metrics is promising, I don’t bother with running the IRR. Maybe a finance colleague would do a facepalm 🤦♂️, but hey, it has worked so far.
So, if you are keen to invest your time wisely, think like a pragmatic investor 😉 I’m pretty sure it’s worth it.
Give it a try.
JBK 🕊
P.S and a friendly reminder: If you’ve found my posts valuable, consider supporting my work. You can help by sharing, liking, and commenting on my LinkedIn posts introducing this issue. This helps me reach more people on this journey.
Thank you for supporting and motivating me to continue sharing my experience with you. Thanks for being part of this community ❤️.