If you’re serious about ISEF, the 3M Young Scientist Challenge, Thermo Fisher JIC, or Genius Olympiad, plan for 6–12 hours a week during the school year and 20–40 hours a week in summer crunch periods. That’s the range our science mentors quote to every student who walks into my office asking, “How bad is this, really?” Over years of mentoring middle and high schoolers through these competitions, our science mentors have learned that the biggest reason promising projects collapse isn’t talent — it’s a bad time estimate made in September. Getting this number right is the difference between finishing strong in April and abandoning a half-finished poster in March. Let’s look at what the real data and real finalists say, so you can plan like someone who’s done this before.
What the numbers actually say
Let us start with the most-cited figure in our world. Former Intel ISEF finalist Amber Hess, writing for Science Buddies, polled her ISEF and Science Talent Search peers and reported that top competitors invest between 400 and 1,600 hours on their projects, with the higher numbers reflecting multi-year continuations. Her own total was “about 800 hours over two years, which is about 8 hours per week, though I put more time in over the summers and less during the school year.” That’s the single best ballpark we can give you for an ISEF-level project.
An ISEF finalist on Quora broke it down by phase in a way that matches what we see with our students: “During the research and planning phase, you might only spend 10 hours a week reading reference material and writing up plans. During the active experimental or development phase, it’s probably a 4-hours-a-day kind of thing, after or before school, and some time on weekends.” Another ISEF/STS winner on the same thread reported their cohort averaged “12–15 hours per week for 4–5 months of active research, plus another 2 months of grueling writing at 8–10 hours per week.”
Math those out. A typical competitive ISEF project lives in the 300–500 hour range for a first-year student going from school fair to state. A top-tier, novel, Grand-Award-caliber ISEF project sits in the 800–1,600 hour range, often spread across two summers. Society for Science’s own rules allow you up to 12 continuous months of data collection — and the students who win usually use most of it.
Where are those hours actually spent
Students consistently underestimate the non-glamorous phases. From the Langley High School ISEF timeline (a well-documented, representative school-year model) and the academic-research guides we hand out, here’s roughly how the hours split on a full-year project:
| Phase | Typical share of total hours | What eats the time |
|---|---|---|
| Literature review & topic refinement | 15–20% | Reading 30+ papers, refining the research question |
| Research plan, forms, IRB/SRC approval | 10–15% | ISEF Forms 1, 1A, 1B; revisions take weeks |
| Experimental design & method validation | 10–15% | Piloting, buying/borrowing equipment, failed runs |
| Data collection | 30–40% | The phase that always takes longer than planned |
| Analysis & statistics | 10–15% | Stats rework, re-running experiments |
| Writing, poster, presentation prep | 15–20% | Abstract, paper, board, practicing the pitch |
Notice that data collection alone is typically a third of your total hours. That’s why mentors obsess about starting early — if you’re still refining your hypothesis in December, you’ve already lost.
How the tier changes the clock
A school or district fair is a different from ISEF. The planning guide at Make It Solar estimates a realistic student has about 6 available hours per week after school, sports, homework, and chores are subtracted — enough to finish a solid 60-hour school-fair project over 10 weeks. That’s your baseline.
A regional or state ISEF-affiliated fair usually demands 150–300 hours. Expect 4–8 hours per week during the school year with heavier weekends.
Regeneron ISEF finalists — the ~1,800 students who emerge from roughly 175,000 annual competitors — almost all worked through at least one full summer at close to full-time pace. In simple words: if you haven’t put in a dedicated summer, you’re competing against students who did. Our mentors noticed that, “A lot of ISEF winners spent a summer working on their projects.”
Compare that with structured research programs and the numbers line up neatly. MIT PRIMES officially expects “at least 10 hours per week during the school year,” rising to 25–30 hours in summer. Stony Brook’s Simons Summer Research Program requires “a minimum of 4 hours per day” with a mentor — around 20–30 hours per week. The Research Science Institute at MIT runs 40–60+ hours per week for 6 straight weeks. Those aren’t exotic numbers; they’re what serious original research actually costs.
Genius Olympiad sits just below ISEF
Genius Olympiad — hosted each June in upstate New York, and focused on environmental themes — attracted 2,400 submissions in 2024 with 727 projects accepted as finalists. The acceptance rate hovers around 50%, and is slightly less competitive than the ISEF Regeneron fair.
Many of our students double-dip, taking the same project to both their ISEF-affiliated fair and Genius Olympiad with added sustainability framing. The submission paper runs up to 15 pages, and Grand/Gold/Silver/Bronze winners in Science are expected to submit to the International Journal of High School Research. Translation: plan for 150–400 hours for a competitive Genius Olympiad entry, with the writing phase stretching further than you’d expect because your paper is being reviewed for publication, not just for judging.
Middle schoolers, the clock runs differently
The 3M Young Scientist Challenge is unique — and honestly, for most middle schoolers, it’s the most manageable national-level competition on this list. Entry is a single 1–2 minute video due by April 30, judged on creativity (30%), scientific knowledge (30%), communication (20%), and presentation (20%). A motivated 6th–8th grader can prepare a strong video in 15–30 total hours over 6–8 weeks: roughly 3–5 hours a week brainstorming, researching the problem, sketching the solution, filming, and editing.
The real time commitment kicks in only if you make the top 10 finalists. Then you enter the summer mentorship program — June through October — working virtually with a 3M scientist and building a working prototype for the October Final Event in St. Paul. Based on finalist vlogs, the cadence is roughly weekly video calls plus 5–10 hours per week of independent work, ramping to 15+ hours per week in August and September. As 2025 finalist Kiyara Gunawardena described her experience building her CORAL underwater robot, “Every week we meet. We set goals and we also talk about different products we can use so that CORAL can be optimized.“
The Thermo Fisher Junior Innovators Challenge (formerly Broadcom MASTERS) works differently. You can’t apply directly — you have to first place in the top 10% at a Society-affiliated fair, which itself typically costs 80–200 hours of project work across several months. The online application (due mid-June) and essays/video supplement add another 15–25 hours on top of your fair project. For the 30 finalists invited to Washington D.C., Finals Week includes team-based STEM challenges, so plan for additional prep in September and October.
For middle school in general, we tell parents the realistic target is 2–5 hours per week during the school year, with one or two intense 6–8 hour weeks near deadlines. More than that and you risk burning a kid out before high school, when the workload genuinely ramps.
The four time-management mistakes we see every year
After watching hundreds of students cycle through this process, the failure modes are depressingly predictable.
First, underestimating approvals. Society for Science explicitly warns that projects involving human subjects, vertebrates, or hazardous agents need SRC or IRB approval before experimentation begins. The Langley High School ISEF guide warns, “You may be asked to revise [your Research Plans] several times before they are approved. This can lead to difficulty meeting subsequent deadlines, and your experimentation will likely be delayed.” Build in 3–6 weeks of buffer here, not three days.
Second, a shallow literature review. A 2019 meta-synthesis of 113 studies on high school research found students “resorted to becoming internet dependent” under deadline pressure, producing shallow or copy-pasted work. If you’ve read fewer than 15–20 real papers on your topic, your project isn’t ready to start — and the judges will know in about 90 seconds.
Third, scope creep. Dr. Frederick Grinnell’s PLOS One series (five published papers from 2018–2025 surveying thousands of science fair students) repeatedly shows that students with mentor access “had an easier time getting their research idea” and “less difficulty getting resources.” Translation: a mentor forces you to scope realistically. Don’t try to cure cancer in a semester.
Fourth, skipping the summer. Every spring, we watch students who competed at the regional level decide in October that they want to make ISEF. It’s almost always too late. The students who advance used the preceding summer. If you’re reading this in June or July, that’s the single most valuable piece of information in this blog.
Fast Forward
A competitive research project is not a weekend assignment; it’s closer to a part-time job you keep for a year. Budget 6–10 hours a week during school, 20–30 hours a week across at least one summer, and build in a month of buffer for approvals and failed experiments. That schedule — which matches what Amber Hess, MIT PRIMES, Simons, and dozens of ISEF finalists independently report — is enough to do real science and still have a life.
The best students we’ve worked with weren’t the ones who put in the most total hours. They were the ones who started earliest, respected the literature, and protected a weekly research block like it was a class on their schedule. If you treat your project that way from October onward, the hours take care of themselves — and by April, you’ll be the student whose poster looks like it belongs there, because it does.
