Once the clinical and operational priorities and objectives for AI have been defined in your AI strategy, you’re ready to start building an AI business case to help evaluate and justify the investment in budget and resources.
A value matrix is an extremely helpful tool for mapping objectives to the associated value potential of various AI solutions. As well, a value matrix helps you define the quantifiable measurement criteria that will be used to measure the bottom-line impact on ROI within your organization. This blog walks through the 3 steps for building an effective AI value matrix and business case.
Step 1: Define and organize your objectives
The first step in developing a value matrix is to identify how the desired outcomes for each objective will be achieved so that you can justify the financial budget, resource allocation, and other non-financial investments in a way that appeals to clinical, IT, administrative, and financial stakeholders.
When outcomes and value indicators for AI, consider the value they bring not only to the radiology department, but also the downstream impact they will have on other clinical departments and the entire health system. For each of the 4 value dimensions, keep in mind AI’s ability to contribute to:
- Clinical Value: AI’s clinical value is demonstrated through improvements to workflow efficiency, quality of care, improved outcomes, and other less tangible factors such as radiologist confidence and satisfaction.
- IT Value: With thoughtful planning and implementation, AI has the potential to significantly improve IT operations by reducing the complexity and effort associated with testing, integrating, and deploying AI across the enterprise and centralizing management and support activities.
- Administrative Value: AI has the potential to improve administrative processes to create efficiencies, increase productivity and performance, and accommodate greater patient throughput.
- Financial Value: AI can offer immense improvements to clinical workflow and outcomes, IT optimization, administrative efficiencies, and cost savings and revenue generation.
Step 2: Build measurable value indicators
Next, each objective must be mapped to measurable value indicators that will enable performance benchmarking of your current state and accurate assessment and tracking of value achieved both near and long-term.
To get a complete picture of overall ROI potential, both quantitative and qualitative factors must be included. For the intangible criteria that is challenging to peg directly to conventional metrics, you may need to consider more creative assessment methods such as referrer satisfaction or patient experience surveys. The following are some examples of quantifiable indicators for each of the 4 value dimensions:
- Quality-based indicators / KPIs
- Discordance reporting
- Number of uncertain or missed findings
- Reading volume (e.g. RVUs) and TATs
- Cost / complexity of infrastructure
- Number of IT hours / FTEs required to manage AI infrastructure and models
- Time to integrate new AI algorithms
- Average wait times
- Capacity / volume
- TATs and SLA adherence metrics
- Patient satisfaction
- RVUs, reimbursements, and revenue margins
- Number of referrals
- Average cost per encounter
- Amount of new quality-based incentives/payments
- ED utilization and readmission rates
- Financial penalties / payments
Step 3: Putting it all together
Once you have defined your desired outcomes and the associated value indicators, create a table so that you can easily compare the value propositions across different AI vendors and assess the potential ROI for each solution. To view a sample AI value matrix, click here.
Are you interested in learning more about building an AI business case and quantifying the value potential to inform ROI-based decision making? Click here to download our AI Buyer’s Guide, which provides an in-depth look at defining value potential and quantifying metrics to build an AI business case.