Research projects fail in predictable patterns, yet most students encounter the same problems repeatedly. Understanding these common failure points can save you months of wasted effort and help you identify projects unlikely to succeed before you invest significant time.

This guide walks through the six points where research projects typically fail, from poorly defined research questions to journal rejections. You'll learn to recognize early warning signs, build prevention strategies into your planning, and salvage projects that are heading toward failure.

Whether you're starting a new project or trying to understand why your current research isn't progressing, this analysis of common failure patterns will help you avoid the mistakes that prevent most trainee research projects from succeeding.  

Failure Point 1: Research Question

Let’s examine each failure point in detail, starting with the foundation of any research project. This is actually where you want bad research projects to fail. You should eliminate as many poor research questions as possible at this stage, investing minimal time in questions that won't lead anywhere productive.

Good research questions are specific, answerable, and are interesting no matter the outcome.  If the analysis is interesting even if the study is null, then you’re working with something good. An example question might be “In low-income adults, is telehealth access associated with lower no-show rates?” If the answer is yes, then it’s interesting because telehealth boosts access to care for a vulnerable population. If not, then it’s still interesting because it suggests the presence of a structural or societal barrier that prevents this population from using telehealth, like digital health literacy or internet access.

If you can't clearly articulate what you're studying and why it matters, stop here. Better to spend a week refining your question than six months collecting data for something unpublishable.

Failure Point 2: Institutional Review Board (IRB) and Study Design 

Once you've developed a solid research question, the next hurdle involves navigating the approval process and finalizing your study design. Projects rarely fail during IRB approval itself, but two related issues can derail you here: time delays and poor study design during the planning phase. 

Part of the IRB application is to design a study and analysis plan. If you don't design a solid plan, you're setting yourself up for failure later. Statistical analysis can adjust for some things like confounding variables, but it can't fix bias introduced by poor study design. Think through your analysis strategy before you collect a single data point. The worst case scenario is to invest time into a project that was doomed to fail before the first data point was collected. Thankfully, this risk can be mitigated with good mentorship or training on how to design appropriate studies. 

Failure Point 3: Data Collection

With IRB approval secured and your methodology locked in, you'll face what often becomes the most labor-intensive phase of research. I can’t emphasize this enough. This is where most projects actually fail because it requires so much manual work before getting to the core research work. Data collection is low-level work that does not produce any meaningful research experience. It requires dedicated research time and focus. 

Chart reviews are the biggest causes of failed projects. Many students get assigned a project to build a database that requires hundreds of hours to complete. This is why we designed Lumono with pre-cleaned datasets - so you can focus on learning research methodology rather than data entry.

If there isn’t a clear research vision before the project starts, then find a new project. I have a more detailed article on how to evaluate whether you should join a research project with a mentor [Link here].

How to avoid this: Be realistic about data collection time upfront. My typical recommendation to mentees is to learn the research process using existing databases. Instead of spending time doing manual work, you’d learn more of how research is done. 

Failure Point 4: Statistical Analysis 

After investing weeks or months collecting data, many students hit an unexpected roadblock that can bring their momentum to a complete halt. Projects can sit for months waiting for statistical analysis if the team doesn't have access to a dedicated statistician or the skills to do their own analysis. 

I've experienced situations where hundreds of hours of data collection remained unused because no one was available to perform the statistical analysis. This felt particularly frustrating because I’d done a lot of work, but just needed someone to analyze the data. At the time, I wished I had some built-in tools to automate my statistical analysis and one of the reasons Lumono can run all the necessary statistical analyses for the user.

There are two main ways to get around this. You can learn how to do your own statistical analysis. Or you can trust that the mentor has the statistical support to keep the project moving forward. The latter is much easier and less time-consuming.

How to avoid this: Confirm statistical support before starting data collection. Know who will analyze your data and when they're available to do it.

Failure Point 5: Writing 

Successfully completing your statistical analysis should feel like crossing the finish line, but another challenge awaits that catches many researchers off guard. Projects less commonly fail during writing, but it happens—especially for projects that aren't particularly novel. I've seen multiple meta-analyses stop progressing because the writing phase dragged on too long.

Once you have results from a well-designed study with completed data collection, it feels like a huge loss not to write it up. If you’re working in a group, there is also social pressure to make sure you don’t let everyone down. No one expects the first draft to be the final draft. The key is maintaining momentum and not letting perfect be the enemy of good. Go through multiple cycles and drafts so that you can push this forward quickly. 

Failure Point 6: Journal Rejection 

Even after crafting a well-written manuscript, one final obstacle stands between you and publication. The last place projects fail is after journal rejection. It's normal for papers to get rejected from the first or second journal you submit to, but some authors give up instead of trying elsewhere. [For more information about the peer review process, check out my other blog here]

The bigger problem is poor journal-project fit. If you're studying rare lung diseases and there's only one journal that publishes that research, you're in trouble if they reject your paper. Always have multiple potential journal destinations in mind before you start your project.

The Bottom Line

Understanding these six failure points reveals a clear pattern that can guide your approach to any research project. Most research failures are preventable with better planning upfront. But even with perfect planning, students still face practical barriers: limited access to clean datasets, statistical software, and writing support.

This is exactly why we created Lumono - to remove the infrastructure barriers that cause projects to fail in Failure Points 1-4, while providing AI-assisted writing support for Point 5. Instead of spending months on chart reviews, you can focus on developing your research skills using pre-structured datasets and built-in analysis tools. Sign up for product updates and to be an early user today.