The Five-Part Formula for Getting AI to Actually Do What You Need

Anyone who has used AI for more than a few minutes has had the experience of getting an answer that’s technically responsive but not quite right. Maybe it’s too general. Maybe it’s missing context you thought was obvious. Maybe — as one DIG member discovered at the April 2026 Philly meeting — it’s working off last year’s tax tables instead of this year’s.

The Five-Part Prompt Formula

TJ then walked through a framework that addresses this kind of problem at the source — before the answer comes back wrong in the first place. A well-constructed prompt, he explained, has five components:

Role. Tell AI who you want it to be. Not just “an expert” — be specific. A seasoned real estate investor. A tax strategist specializing in small business. A COO who’s scaled a property management company. The more precisely you define the role, the more the response is filtered through the right lens.

Context. Give AI what it needs to work with. Here’s the deal. Here are the numbers. Here’s my situation and the relevant background. This is where most weak prompts fall apart — they ask a question without providing the information needed to answer it well.

Objective. Define what success looks like. What is the actual result you’re after? Not just “analyze this” but “tell me whether to pursue this deal and what would need to be true for it to make sense.” Clarity on the objective shapes everything that follows.

Constraints. Name the limits. Time horizon. Budget. Legal restrictions. Regulatory context — like, for example, the fact that you need the 2026 tax figures, not the 2025 ones. Constraints aren’t limitations on the answer; they’re information that makes the answer more accurate and more useful.

Output. Specify the format. Do you want a summary? A step-by-step breakdown? A pros and cons list? A draft email? Telling AI how you want the information delivered is as important as telling it what information you want.

Prompting Is Delegation

You would never hand a task to a real employee with no context, no defined objective, and no guidance on what a good outcome looks like. You’d brief them. You’d tell them what you need, why it matters, what the constraints are, and what the deliverable looks like. The clearer the brief, the better the work.

AI works exactly the same way. The quality of what you get back is a direct function of the quality of what you put in. A vague ask produces a vague answer. A well-constructed prompt — role, context, objective, constraints, output — produces something you can actually use.

Rob Coldwell and TJ Hock are the owners of Rentwell Property Management. They were the featured speakers at the April meeting of DIG Philly in King of Prussia. Hear their full presentation and many more on our podcast “Living Well with Rentwell”.

Subscribe today on Apple, Spotify, and Youtube!

For more information on property management in the Fishtown area, schedule a free owners’ consultation at Rentwell.com