Skip to content
Phase 4: Program Design

AI Prompt for SMART Objectives in a Grant Proposal

By Angel Reyes, DrPH(c), PMP, PMI-ACP, MPH, MCHES · Updated · Free to copy and use
Quick Answer

This AI prompt for SMART objectives in a grant proposal turns a program goal into objectives that are specific, measurable, achievable, relevant and time-bound, using your own baseline data and capacity. It checks each draft against the five criteria in a table and separates outcome objectives from process objectives. Where you have not supplied a baseline or target, it leaves a bracketed blank and asks you for the number. It never makes one up.

The prompt

Copy and paste
You are helping me write SMART objectives for a grant proposal. Use only the information I supply. Do not invent baselines, targets, percentages, numbers of participants or timeframes. Where a number is needed and I have not supplied it, leave a bracketed blank such as [TARGET %] or [BASELINE] and ask me for it.

Program goal: [PROGRAM GOAL]
Who the program serves and how many we can serve: [POPULATION AND CAPACITY]
Activities, with frequency and duration: [ACTIVITIES]
The changes we expect in participants or conditions: [EXPECTED CHANGES]
Baseline data we have, with source and date: [BASELINE DATA]
Targets our program staff consider realistic, if already agreed: [AGREED TARGETS OR "NOT YET SET"]
How we can measure each change with the tools we have: [MEASUREMENT TOOLS]
Grant period: [START AND END DATES]
Funder's requirements for objectives, if any: [PASTE FUNDER REQUIREMENTS OR "NONE"]

Step 1. Draft three to five objectives that follow from the goal. Label each one "outcome objective" (a change in people or conditions) or "process objective" (something the program will deliver). Include at least one outcome objective.

Write each in this form: By [date], [who] will [measurable change or delivery], from [baseline] to [target], as measured by [measure].

Step 2. Check every objective in a table with these columns:
Objective | Specific | Measurable | Achievable | Relevant | Time-bound | What is missing

For each criterion write Yes, No or Cannot tell, with a short reason. For "Achievable," judge only against the capacity and baseline I gave you. If I gave you no basis for judging it, write "Cannot tell: needs staff confirmation."

Step 3. List every bracketed blank and the question I need to answer to fill it.

Step 4. Point out any objective that only counts activity while claiming to show change, and any expected change of mine that has no objective.

Do not suggest typical targets or benchmarks from other programs.

Works in ChatGPT, Claude, Gemini and other AI chat tools. Replace each bracketed placeholder with your own details before you send it.

What to fill in

[PROGRAM GOAL]
The broad result the program is working toward. Goals are general. The objectives the prompt drafts are the specific, measurable steps toward the goal.
[POPULATION AND CAPACITY]
Who you serve and how many people you can realistically serve with the staff and budget planned. This is what "achievable" is judged against.
[BASELINE DATA]
Where things stand now, with the source and date: last year's completion rate, a pre-program survey result, a figure from a public data source. If you have none, write "none" and the prompt will show where you need one.
[AGREED TARGETS OR "NOT YET SET"]
Targets your program staff have already agreed are realistic. If they have not been set, say so. Setting them is a decision for your team.
[MEASUREMENT TOOLS]
How you could measure each change with what you have: attendance records, a pre and post survey, case notes summarized in a database.
[START AND END DATES]
The grant period, so every objective has a date that falls inside it.

Why this prompt works

Ask an AI tool for SMART objectives with no data and it will give you neat sentences with numbers in them, such as "increase literacy by 25%." The number came from nowhere. The prompt is designed to prevent exactly that.

  • Blanks instead of invented numbers. A bracketed [TARGET %] is honest. It shows you which decisions are still open, and each one comes back to you as a question.
  • A sentence frame. "By [date], [who] will [change], from [baseline] to [target], as measured by [measure]" builds the five criteria into the structure, so a missing element is obvious.
  • A criterion-by-criterion check. The table makes the AI evaluate its own drafts and explain each judgment. "Cannot tell" is an allowed answer.
  • "Achievable" is tied to your capacity. An AI cannot know what your program can deliver. The prompt limits that judgment to what you told it and otherwise sends the question to your staff.
  • Outcome and process objectives are labeled. Serving 100 people is a process objective. What changes for those people is an outcome objective. Funders usually want both, and step 4 catches a proposal that has only the first kind.
  • No borrowed benchmarks. A target that worked for another program in another place is not evidence of what yours can do.

Why it is appropriate to use

Write objectives during program design, after the logic model and before the evaluation plan and budget. Objectives are the hinge between them. The evaluation plan measures them and the budget pays for the work that achieves them.

AI is useful for the wording and for checking each objective against the criteria, which is mechanical. The substance is not mechanical. A target is a commitment. If you are funded, the funder may measure your performance against the numbers in your proposal, and you will report on them. That is why the prompt refuses to supply them and why the people who run the program must set and confirm them.

Do not use the prompt to produce targets, to make a weak design sound precise or to add objectives your program has no way of measuring.

Keep participant-level data out of the prompt and use summary figures. For the method behind the criteria, see the guide to SMART objectives and specific aims.

How to check and revise the output

  1. Fill every blank with a real number agreed by program staff, or rewrite the objective so it does not need one yet.
  2. Check each baseline against its source and note the date. An old or mismatched baseline makes the target meaningless.
  3. Ask the people who will do the work whether each target is achievable with the staff, time and budget planned.
  4. Confirm you can measure it. For each objective, name the tool, who collects the data and when.
  5. Check the dates fall within the grant period and allow enough time for the change to happen.
  6. Read them against the logic model and budget. Every objective should trace to an activity that is funded.
  7. Follow the funder's format if they specify how objectives must be written.

Frequently asked questions

What does SMART stand for in grant writing?
Specific, measurable, achievable, relevant and time-bound. A SMART objective states who will change, what will change, by how much, by when and how it will be measured.
What is the difference between a goal and an objective?
A goal is the broad result a program works toward. An objective is a specific, measurable step toward that goal with a target and a deadline. A proposal usually has a small number of goals, each with several objectives.
Can AI set the targets for my objectives?
It should not. An AI tool has no knowledge of your baseline or your capacity, so any number it supplies is a guess. Targets should come from your own data and be confirmed by the staff who will be responsible for meeting them.
What is the difference between a process objective and an outcome objective?
A process objective describes what the program will deliver, such as the number of sessions held or people served. An outcome objective describes the change that results for participants or conditions, such as a gain in skills or a change in behavior.

More prompts for program design

Related prompts and guides

← Back to all AI prompts for grant writing

Ready to Master Grant Writing?

The Complete Grant Architect is a free 16-week course that transforms you from grant writer to strategic grant professional. Learn proposal engineering, federal compliance, budgeting, evaluation design, and AI-powered workflows.

Enroll Free in The Complete Grant Architect