How to Troubleshoot Experiments that Just Aren’t Working

Thomas Edison, inventor and world-record holder of over 1,000 patents, said “I have not failed. I’ve just found 10,000 ways that won’t work.”

In research and development labs, failures can often lead to more progress than the experiment itself, if you know how to navigate them. Getting stuck and not making progress can be a challenge, especially mentally. When you get discouraged, it’s easier to give up and move on, chalking it up to a bad project, but more fruitful if you can figure out what’s going wrong, recalibrate, and try again.

Why Failure is Important to Scientific Progress

Arguably one of the most famous scientific innovations, modern antibiotics as we know them were discovered by Alexander Fleming as an accident in the 1920s, when he noticed a blob of mold on a petri dish killing bacteria. Failure of the bacteria experiment? Absolutely! Did it lead to one of the most impactful scientific discoveries? Also yes!

Scientific progress is driven by failure even though the majority of work that is shared, celebrated, written about, and published is about success. It takes a team mentality to lift people up after a failed experiment, make a new plan, and move forward.

8 Common Causes of Stalled Experiments

Experiments can fail for a number of reasons, a lot of which are avoidable. Here are a few of the most common root causes of failed or stalled experiments.

1. Not enough data
2. Faulty or incorrect equipment or supplies
3. Human error
4. Missed issues like a syncing problem or subjects respond in an unexpected way
5. No clearly defined protocol/plan for the research
6. Improper storage of materials and reagents
7. No documentation, causing errors or duplicate work
8. Taking shortcuts like not waiting the entire incubation period

Research requires a close attention to detail. Use checklists and lab management software whenever possible to avoid human error and simple mistakes that can cost an entire experiment, a lot of budget, and morale.

4 Steps to Troubleshoot Experiments that Aren’t Working

When you do hit a plateau in moving an experiment in the lab forward, here are actionable steps you can take to turn a “failure” into progress and learn from the experience, no matter what.

Step 1: Analyze all of the elements individually.

Carefully review everything used for the project. Are any of the agents or supplies expired? Were they all correct? Is your lab equipment calibrated and was it recently serviced? Re-trace all of the steps of the experiment with an eye out for errors.

Step 2: If budget allows, re-run the experiment with new supplies.

When you go to re-run the experiment, double check all steps with a colleague to ensure accuracy. Tests like Western Blots and antibody experiments in general can be tricky and may require an expert lab technician to get it done right.

Step 3: Ask questions and consult with colleagues or experts.

Ask why something failed and talk to your lab manager or PI about the resources you have to explore the “failure.”

Step 4: Take risks and advocate for additional research.

Once you do identify a potential source of the problem and if it might bring up new research, take risks. The highest research risks often put out the highest reward, but some labs are hesitant about riskier projects. Use research, data, and technology to advocate for your ideas and pursue grant funding for your project.

.  .  .

Experiments are optimized for success when they start with a solid plan, are organized, and collect data and leverage technology. Plan your experiments using software to help maximize research output and avoid unnecessary research mistakes.