Developing the AI Business Case: Part 2

Defining project success

It’s a good idea to specify what success looks like at this stage. We can utilize the organization’s present process statistics—for example, how long something takes or how many units of a product are produced in a given time frame—as a baseline for comparison if the AI product replaces or supplements an existing process. For example, an e-commerce site may measure the change in the number of purchases made by customers by recording sales after adopting a recommendation engine and comparing it to the number of sales when the company didn’t use AI.

Defining the business vision

The data scientist may be able to provide an industrial perspective to the management team when they present their business use case. The following are some of the important questions to consider when presenting a new AI project:

• What are our competitors doing in the market?

• How can we use other frameworks like SWOT analysis (strengths, weaknesses, opportunities, and threats) to our advantage?

Depending on the organization and the importance of the AI evaluation, presenting use cases and a market overview of AI implementation in an organization’s sector to the leadership team may help deliver a clearer business vision.

Estimating implementation costs

The product owners also have to estimate the cost of the project, the resources needed, and the completion date.

Funding the project

The business owners must fund the AI projects, just as the product owner must present them to the business owners or management team.