Is It Better to Join an Established Lab or Start My Own Project?

It’s late at night, and you’ve got too many tabs open: lab websites, summer research programs, maybe a draft email to a PI you’re scared to send. You keep circling the same question:

Should you join an established lab or try to start your own research project?

Here’s the answer you’re looking for:
For almost all premeds and medical students, joining an established lab first is the better move. Starting your own project can be great, but it should usually come after you’ve built some foundation.

Let’s walk through how to decide, with specific scenarios, not generic “it depends” fluff.

(See also: Should I Prioritize Depth or Breadth in My Research Experiences? for more details.)

---

The Core Question: What Do Med Schools Actually Want?

Med schools don’t care if you were the “founder” of some fancy-sounding project that never went anywhere.

They care about:

• Evidence you can:
  • Commit to something over time
  • Work in a team
  • Think scientifically
  • See a project through to a result (poster, abstract, paper, presentation)
• Concrete outcomes:
  • Posters and abstracts
  • Publications (even middle author)
  • Presentations at local/regional/national meetings
  • Strong letters from researchers who can say, “I’ve seen this student think and work”

You can accomplish those much more reliably in an established lab than in a self-created project with no infrastructure.

So the main principle:

If your primary goal is to strengthen your application, join an established lab first.
If you have time, mentorship, and a clear niche opportunity, then consider starting a project.

---

When It’s Better to Join an Established Lab

You should default to joining an existing lab if any of this sounds like you:

• You’ve never done research or only done short, surface-level projects
• You don’t know basic research methods or stats
• You don’t have a PI willing to mentor a new project from scratch
• You’re early in college (freshman/sophomore) or early in med school (M1)
• You have limited time (heavy coursework, MCAT studying, clinical obligations)

Here’s why established labs are so powerful for premeds and med students.

1. You Get a System That Already Works

In a functioning lab, you step into a machine that’s already moving:

• IRB or IACUC approvals already in place
• Data collection pipelines set up
• Lab protocols written and tested
• Regular meetings where you can present and get feedback
• A PI who knows how to turn data into posters and papers

Example:
A cardiology lab at a big academic center might already have a database of thousands of PCI patients. You don’t have to build the dataset from scratch; you can help with:

• Data cleaning
• Running analyses
• Writing parts of an abstract
• Building figures for a conference poster

You’re learning the full research workflow without having to invent every step.

2. You Get Mentorship (Which You Don’t Realize You Need Yet)

New researchers underestimate how many micro-decisions go into a project:

• Is this question actually answerable with available data?
• How many subjects do we need?
• Which variables should we collect?
• Which statistical tests make sense?
• How do you write the methods clearly?

Working in an established lab gives you:

• A PI to set direction
• Postdocs/grad students/residents to show you the “how”
• Senior students who’ve already made the same mistakes you’re about to make

This mentorship is exactly what leads to good letters and tangible outputs.

3. Your Chance of Finishing Something Goes Way Up

Med schools like completed things, not big ideas.

In an established lab, the odds are much higher that you’ll:

• Get on a poster or abstract in 6–12 months
• Contribute meaningfully to a manuscript
• Present at a local/regional meeting
• Learn how to respond to reviewer comments and revise

Compare that to a solo project where you:

• Have to find data or patients
• Write an IRB from scratch
• Design the study, survey, or protocol
• Navigate approvals, logistics, and analysis mostly alone

All of that is doable, but not fast or easy. And with tight application timelines, “doable” isn’t good enough—you need “very likely to finish.”

---

When It Actually Makes Sense to Start Your Own Project

There are times when launching your own project is a smart move. But they’re narrower than most students think.

Starting your own project makes sense if:

1. You already have some research experience

   • You understand basic study designs (retrospective chart review vs prospective study vs survey)
   • You’ve seen at least one project go from start to finish

2. You have a committed mentor

   • A faculty member or resident who says:
     • “Yes, I’ll be your PI on this”
     • “Let’s meet every X weeks to keep things on track”
     • “I’ll help with IRB, design, and edits”

3. You’re plugging into existing infrastructure, like:

   • Clinic or hospital with data already being collected for QA
   • Medical school course where an educational project is encouraged
   • Residency program or department eager to publish program evaluation, quality improvement, or educational outcomes

4. You can realistically finish something before your key deadline:

   • For premeds: before AMCAS/AACOMAS submission or secondary season
   • For med students: before ERAS submission

Good “Start Your Own Project” Scenarios

• M2 in an underserved clinic who notices high no-show rates in diabetic patients

  • Faculty mentor agrees to help design a short survey + chart review
  • The clinic already tracks appointments and patient demographics
  • Project: identify predictors of no-shows, propose and test a small intervention
  • Outcome: QI poster at a regional conference, maybe a short paper

• Premed in a psych department volunteer role

  • You notice many patients are being referred to group CBT but few attend
  • A psych fellow agrees to help you design a brief phone follow-up study
  • Minimal IRB risk, existing patient lists available
  • Outcome: departmental presentation, senior thesis, maybe a publishable analysis

• M1 interested in medical education

  • Faculty director of a first-year course supports you formally evaluating a new team-based learning module
  • Built-in access to student evaluations and exam scores
  • Outcome: Med ed conference poster (e.g., AAMC, regional), possible publication

All of these involve: mentorship, infrastructure, and feasible scope.

---

How to Decide: A Simple Framework

Use this quick filter. If you answer no to any of these, you should probably start in an established lab.

Question 1: Do you have a committed mentor willing to be PI or co-PI?

• Yes → go to Question 2
• No → join an established lab

Question 2: Is there existing infrastructure or data you can tap into?

• Yes → go to Question 3
• No → your project will likely stall; join a lab first

Question 3: Can you get from idea to a realistic output in 6–18 months?

• For undergrads: poster/abstract, thesis, or small paper before med school apps
• For med students: something before ERAS
• Yes → own project is reasonable
• No → your timeline doesn’t match your ambitions; go established first

Question 4: Do you have at least some prior research exposure?

• Yes → you’re in safer territory
• No → learning research while trying to be PI is tough; join a lab

---

How to Make an Established Lab Feel Like “Your Project”

You might be worried you’ll just be a “data monkey” in someone else’s lab. That happens, but you can avoid it if you’re intentional.

Here’s how to carve out ownership inside an existing lab:

1. Ask for a defined sub-project early

   • Example: “Is there a discrete question or dataset I could be primarily responsible for under your guidance?”

2. Volunteer where there are gaps

   • “Do you have any projects that started but stalled that I could help revive?”

3. Propose a small, focused angle on an existing larger project

   • “I noticed we have a lot of data on X—could I look specifically at Y subgroup?”

4. Show up consistently and follow through

   • Reliability is what makes PIs trust you with more responsibility and authorship

Over time you can go from:

• Simple tasks → managing parts of a project → driving an entire sub-project

On your CV and in interviews, this is easy to frame as:
“I joined Dr. Smith’s lab, then developed and led a sub-project on X…”

That’s effectively “starting your own project,” but with support and credibility behind it.

---

Premed vs Med Student: Does the Answer Change?

Yes, slightly.

For Premeds

• You usually have:
  • Less time in any single institution
  • Less access to patients and clinical data
  • Fewer mentors who know your work well

So as a premed, joining an existing lab is almost always the best move.

Good targets:

• Labs aligned with your college’s biology, neuroscience, psychology, or public health departments
• Clinical research programs affiliated with your local hospital
• Summer research programs (SURF, REU, MSTP summer programs)

Solo projects (like surveys or local quality improvement) are harder to set up well as a premed without strong faculty backing.

For Med Students

You have:

• Access to patients, EMRs, and hospital-based projects
• More clinically relevant angles for QI and outcomes research
• Exposure to residents/fellows who actively want research partners

So for med students:

• M1/M2: start with an established lab
• After you’ve got your feet under you: consider starting a focused project if:
  • You find a mentor
  • There’s a clear clinical or educational problem
  • You see a way to do something small but meaningful

---

Red Flags: When “Starting Your Own Project” Is a Bad Idea

Be cautious if:

• The “mentor” says: “Just email me updates—I’m really busy,” and offers no concrete plan
• There’s no clear path to ethics/IRB approval
• The project relies on:
  • Cold-emailing random doctors for data
  • Getting thousands of survey responses from strangers with no recruitment pipeline
  • Building a dataset manually with no help and no protected time

Also a warning sign: you’re doing it primarily to put “PI” or “Founder” on your CV. Admissions committees can smell fluff.

---

How to Talk About Either Path on Your Application

Med schools care less about whether the project was “yours” and more about:

• What you actually did
• What you learned
• What came out of it

So in your AMCAS/AACOMAS or ERAS:

For an established lab project, emphasize:

• Your specific role (data analysis, literature review, experiment design, survey development)
• A clear outcome (poster at XYZ, manuscript in submission, abstract accepted to ABC conference)
• A brief insight (“I learned how to interpret conflicting results and revise hypotheses”)

For a self-started project, emphasize:

• How you identified the problem or gap
• How you designed and executed the project with your mentor
• Any measurable impact (improved workflow, patient outcomes, course evaluations, etc.)

Both can be powerful—if they’re real, specific, and complete.

---

FAQ (Exactly 5 Questions)

1. I’m a freshman premed with no experience. Should I even think about starting my own project?
No. Your priority should be learning how research works from the inside. Join an established lab, learn basic methods, understand how projects move from idea to publication, and build relationships. You can always develop more independent ideas after you’ve seen the process once or twice.

2. Will med schools be less impressed if I just joined an existing lab instead of starting my own thing?
Not if you actually contributed and have something to show for it. A middle-author paper, a poster, and a strong letter from a respected PI in an established lab will impress committees far more than a “self-started” project that never led to anything concrete. Real outcomes and strong letters beat titles every time.

3. How long do I need to stay in a lab for it to be meaningful on my application?
Ideally at least a year, longer if possible. Under 6 months often looks superficial unless it’s a defined summer program. A multi-year relationship with a lab can lead to multiple outputs and a powerful letter, even if you only worked a few hours per week during semesters.

4. What if the established lab I join just gives me boring, repetitive tasks?
That’s common at first. Do the simple tasks well and on time, then explicitly ask for more responsibility: “Once I finish this, I’d love to take on a small project or analysis I can help own. Are there any opportunities like that?” Many PIs are happy to give more meaningful work to students who prove they’re reliable.

5. I have a great idea but no mentor yet. Should I wait or just start on my own?
Wait—and use that time to find a mentor. You can do background reading and refine your question, but don’t launch a project that involves data, patients, or surveys without faculty support. The risk of stalling, making design errors, or running into approval problems is high. A mentor can turn your good idea into a realistic, publishable project.

---

Key takeaways:

1. For most premeds and early med students, joining an established lab first is the smartest, most efficient move.
2. Starting your own project works best when you already have some experience, a committed mentor, and existing infrastructure to plug into.
3. Med schools care about real contributions and finished work, not whether the project technically started with you.