Wyndham Hotels Technologist Improves AI Software Development
Defining project requirements before generating AI prompts is crucial for meeting business needs. This approach helps reduce wasted effort and improve results.
Original reporting · martech.org
Original: “Better prompts start before the prompt”
Read “Better prompts start before the prompt” at MarTech →
How this was written. GMS Agency Desk wrote this article with a language model from MarTech’s reporting (model: @cf/meta/llama-3.3-70b-instruct-fp8-fast); it describes that reporting rather than adding to it. The original, linked above, is the full story.
Key points from MarTech’s report
- Defining project requirements before generating AI prompts is crucial for meeting business needs.
- Established development practices, such as user stories and acceptance criteria, can help marketers provide AI with better guidance.
- AI can speed up the creation of artifacts, including business requirements documents and prompt frameworks, making the development process more efficient.
- Marketers must take on a more significant role in defining requirements and guiding the AI development process.
- Clear direction is necessary to ensure that faster code generation produces systems that meet business needs and do not introduce security and performance problems.
According to MarTech, the key to improving AI-built software lies in defining the user, task, and success metrics before generating any code. This approach enables marketers to give AI better direction and achieve their goals more efficiently.
AI can accelerate marketing technology development, but rapid code generation does not guarantee that the output will meet business requirements. To get useful results, marketers must define their needs clearly before prompting AI.
Established development practices, such as creating user stories and acceptance criteria, can help marketers provide AI with better guidance. These practices enable teams to define expectations upfront and avoid unnecessary iterations.
MarTech reported that various artifacts, including business requirements documents and prompt frameworks, can aid in defining project intentions. AI can also speed up the creation of these artifacts, making the development process more efficient.
Business requirements documents provide a detailed business vision for large initiatives, helping stakeholders to state the overall goal and success metrics. These documents keep the big picture in focus, enabling project participants to make decisions that align with the vision.
User stories break down projects into manageable tasks, defining the who, what, and why of a project or initiative. This approach empowers programmers to make decisions that align with the project's end goal, ensuring that the output meets business needs.
Acceptance criteria clarify user stories, specifying what success looks like without dictating how to accomplish the goal. Including acceptance criteria in a prompt ensures that it covers the necessary details, reducing the need for repeated revisions.
Questions and answers
What is the key to improving AI-built software?
Defining the user, task, and success metrics before generating any code is crucial for meeting business needs.
How can marketers provide AI with better guidance?
Established development practices, such as creating user stories and acceptance criteria, can help marketers provide AI with better direction.
What role do marketers play in the AI development process?
Marketers must take on a more significant role in defining requirements and guiding the AI development process to ensure that the output meets business needs.
