How to Write a Winning MIT THINK Proposal

By Dr. Harman Brah, MD · Judge, California Science & Engineering Fair

A strong MIT THINK proposal pairs one insightful idea with clearly defined goals and a step-by-step procedure that can really be finished in one semester on a $1,000 budget. Above all, it has to be the student’s own plan, because semifinalists are interviewed and have to defend every choice in it.

THINK is a STEM research and innovation program run by MIT students. Applicants submit a proposal for a project they have not done yet, so the whole application rests on the plan.

Start with the idea

THINK describes a good proposal as having “an insightful idea, clearly defined goals, and a well thought-out procedure for implementation.” The idea comes first for a reason.

An insightful idea does not have to be grand. It has to show that the student noticed a real problem and has a specific reason to think their approach might help.

Write the problem in one or two sentences. Then write what you would build or test, and why you expect it to work.

If you cannot say it that briefly, the idea is probably still too broad. One clear question is stronger than a list of ambitions.

Make the goals measurable

A goal is clearly defined when a reader could check, at the end of the semester, whether you reached it.

“Improve” or “explore” leaves that open. A goal that names what you will measure, and what result would count as success, closes it.

It also helps to separate the main goal from stretch goals. Reviewers can then see what you will finish even if the extra parts run out of time.

Lay out a procedure that fits one semester

This is where most of the thinking goes. Every project has to be doable in one semester, so the procedure is really a schedule.

List the steps in order. For each one, say what you will build or test, how you will know it worked, and roughly when it happens.

Then look for the step most likely to go wrong. Say what you would do if it does. A plan with a backup shows you have thought it through.

In our experience judging student research, one clear question with real data beats an ambitious project with no result. The same holds for a proposal: a narrow plan that can finish is easier to trust than a large one that cannot.

Build a budget that adds up

The project has to fit a $1,000 budget, and finalists receive up to $1,000 in funding to carry it out.

List each item you need and why. Tie every line to a step in the procedure, so a reader can see where each dollar goes.

If the total runs over, cut scope rather than quality. A careful budget is one of the clearest signs that the plan is real.

Write it for a reader outside your field

The proposal is read by people who may not know your topic well. Explain the problem plainly before the technical details.

Short sections, named steps and a simple budget table make it easy to follow. Clear writing tells the reader the student understands the plan.

The student does the work

The idea, the plan and the writing should all be the student’s. That is what the interview tests, and it is what makes the semester of building go well if the student is chosen.

Judges see through a project that looks parent- or mentor-built quickly. A student who wrote the plan can explain why each step is there without a script.

A mentor can help the student pressure-test the plan: is the scope right for a semester, is each step realistic, does the budget add up. The student still makes the choices and writes every word.

Prepare for the interview while you write

Semifinalists are interviewed, so write the proposal as if you will be asked about every line.

For each step, be ready to say why you chose it, what could go wrong and what you would try next. Those answers come easily when the plan is yours.

Timing

The 2026-27 dates had not been posted as of October 31, 2026. Last cycle, applications opened November 1, 2025 and were due January 1, 2026.

Check the THINK site for this year’s dates, and use the time now to settle the idea and draft the procedure and budget.

Our physician-led mentors help students shape and pressure-test research plans while the student does the work. See our science fair and research mentorship, or read about doing high school research without a lab.

FAQ

What makes an MIT THINK proposal stand out?

A focused idea, goals a reader can check, a realistic step-by-step procedure for one semester, and a budget within $1,000 that ties to each step. It should read as the student’s own plan.

Does my child need research experience or results to apply?

No. THINK asks for a proposal for a project that has not been done yet, so it rewards a clear plan rather than past results. A younger student with a well-defined question can compete.

How detailed should the budget be?

Detailed enough that each item connects to a step in the procedure and the total stays within $1,000. If it runs over, narrow the project.

Can a mentor or parent help write the proposal?

They can ask questions and help the student test whether the scope, steps and budget are realistic. The student should choose the idea and write the proposal, because semifinalists are interviewed on it.

What happens if my child becomes a finalist?

Finalists get mentorship from MIT professors and researchers, check-ins with the THINK team, up to $1,000 in project funding and, if circumstances permit, a trip to MIT. They finish the project by June and present their results.