Setting Up for AI Success

A Framework for Choosing the Right Use Cases

As hype around data science, machine learning, and AI continues to grow, more and more organizations are feeling the pressure to modernize their business by implementing advanced data solutions or else risk falling behind the technology curve.

But how to choose the right project?

This white paper has answers, providing:

A framework of questioning for choosing use cases.
A checklist of considerations associated with each question to answer before starting the project.

About the Author

Christina Hsiao

Evangelist and Sales Engineer at Dataiku

Christina Hsiao is a technical evangelist for Dataiku based in the US. In her role, Christina is able to share her passion for applied data science through writing and by speaking with customers, partners, and organizations interested in solving business problems with the powerful combination of people, data, and technology. Prior to joining Dataiku, she spent nearly a decade at SAS, mainly specializing in Natural Language Processing and text analytics. Christina holds a bachelor’s degree in Mechanical Engineering from Stanford University.

About the White Paper

An ideal AI project will have clear and compelling answers to each of these questions:

WHO will this project benefit?
HOW will it specifically improve experience or outcomes, and HOW can this be measured?
WHY is using AI for this purpose better than existing processes?
WHAT is the upside if it succeeds, and WHAT are the consequences if it fails?
WHERE will the data come from, and does it already exist?
WHEN should an initial working prototype and, subsequently, a final solution in production be delivered?

This white paper will go through each of these considerations in detail and provide examples, strategies, and follow-up questions to help guide the selection of AI use cases that will ultimately bring business value.

Extract: WHEN Does an Initial Working Prototype Need to Be Completed?

Finally: the all-important timeline. Like a do-it-yourself home construction project, it can be easy for an AI application build to stretch on and on; there always seems to be a bit more fine tuning to be done, another small feature to be added.

But in order to build credibility with internal stakeholders, it’s best practice to have a limited slice of the solution working from end-to-end in a short period of time rather than aiming for a fully baked solution in the first pass.

Real-life example: Take the example of a customer-facing chatbot; the first prototype can either focus on breadth (i.e., it can answer a wide variety of simple inquiries across diverse topics), or depth (it can answer very detailed questions asked in many different ways, but only about a few topics). However, it’s probably not realistic to strive for both of these goals in a first iteration. Once the chatbot preview earns internal user acceptance -- that is, stakeholders confirm that once built out, it will in fact deliver the desired business outcomes--only then is it appropriate to fully flesh out the application.

The Key to Successful AI Projects

Today's most successful organizations have embraced the idea that effectively leveraging data and technology can not only drive competitive advantage, but also improve the experiences of both their employees and customers. But it’s important to be strategic and deliberate when planning an organization's AI strategy.

Following the framework outlined in this white paper will help avoid false starts on AI projects that are ill-defined and create an environment for success.

Setting Up for AI Success

Get the White Paper

About the Author

More Recommended Content

About the White Paper

Extract: WHEN Does an Initial Working Prototype Need to Be Completed?

The Key to Successful AI Projects

Get the Full White Paper