How To Make Fewer Mistakes In The Lab

How often do you make errors in the lab that ruin a good experiment? Rather than flaws in experimental design, I mean errors like forgetting to add a reagent, pipetting the wrong amount or following a protocol step wrongly.

Especially early on in your career, errors like this can be a real drain on your productivity. As I talked about earlier, listening to music in the lab may or may not help you here, but here are 10 other ways in which you can reduce your error rate and get more results.

1. Use a checklist. I have always found that using checklists during experiments is a great way to focus the mind and stop me from forgetting to add something or doing things in the wrong order. But you don’t have to just believe me… a recent study in the New England Journal of Medicine showed that the use of simple checklists during surgery cut deaths and complications by one third. So perhaps by adopting checklists you can reduce the attrition rate for your experiments.

2. New protocols and SOPs: write out your own version. When you are performing a new protocol, the worst thing you can do is just jump in and get started without gaining a knowledge of the whole procedure first. That is just asking for things to go wrong. So one rule I always keep is to write out my own version of the protocol in Word before I start, using a standard format. This gives two benefits. Firstly, it forces you to read through the whole protocol and secondly it gives you a standardised copy of the protocol that you can keep, annotate and make checklists from.

3. Annotate. If you make a mistake in a protocol, annotate your copy so that you won’t make the same mistake again.

4. Repetitive pipetting: be consistant and use bookmarks. If you are doing a large experiment with a lot of repetition, just letting your mind wander for a few seconds can ruin your experiment if you don’t put in safeguards. One example is where you have a whole rack of tubes into which you are repetitively pipetting — it’s so easy to get lost and forget where you were. You can guard against this by being rigidly consistant with your pipetting order and by bookmarking your progress by closing the lid (or similar) after you have pipetted.

5. Don’t multitask too much. Multitasking is essential in the lab, but should be practiced with care as overdoing it can make you error-prone. More on this here.

6. Get set up before you start. This is obvious, but it’s often neglected… Make sure you have everything you need before you start. The moment you wander off to pick something up you are insertingĀ  an error-inviting distraction into your procedure.

7. Prepare in bulk. For experiments you perform often, it pays to put in a bit of time upfront to prepare for a lot of future experiments in advance. Make stock solutions in bulk (aliquoted if required), prefill tubes with reagents so you can just take them out of the fridge/freezer and go, and create experiment forms to save you from writing out the same stuff in your lab-book over and over again (for more on this click here). The more you can prepare upfront, the less room there is for error later.

8. Don’t spend so long in the lab. Ok so you need to put in the hours to make sure you generate results but there is a tradeoff because spending too long in the lab will dull your sharpness. Where the cut-off lies is a personal thing.

9. Get enough sleep. Just like too many hours in the lab will dull you, so will lack of hours in bed. Get some sleep!

10. Take responsibility. Especially if you are an early career scientist, problems can occur if you expect others to be responsible for your work, for ensuring that you have everything you need etc. Taking the viewpoint early on that the buck starts and stops with you will empower you to make sure that you create the situation where your experiments have the optimal chance of working.

Those are my thoughts on reducing your error rate in the lab. What approaches do you use?

8 Comments

  1. Nasir Ali on August 25, 2016 at 5:35 pm

    As a I am related to Molecular biology field, initially I forgot to which tube I have added reagents and which not. To clear this confusion I count the tips and equal to the tubes and then start pippeting it helped me alot. But I dont need this anymore.

  2. Roberto on November 25, 2009 at 1:24 pm

    dqlover: point #10 means, don’t expect other people to take care of your work for you: if your experiment doesn’t work, because “John made my solutions with the wrong pH”, then John is not the only one to blame šŸ˜‰

    I don’t use checklists, but rather move the tubes with reagents and samples (say, from one corner to another of the ice bucket) to remember that I’ve already pipetted them. It helps.

  3. Guillaume Carpentier on November 24, 2009 at 4:18 pm

    As you have to share the devices with others scientist, one way to prevent problems is also to indicate that you will use a particular device for a given period (centrifuge, imager…).
    These 10 steps are clearly a good answer for waste of time and will.

  4. Seena on November 24, 2009 at 5:40 am

    I believe planning your work one or two days beforehand helps a lot. You can prepare reagents and check all lab wares are autoclaved and ready for use and that all the instruments are working properly before you start.Make a note of every deviation you make from the protocol(improper time maintainence,new chemicals) so that it will be easier to find what went wrong.

  5. dqlover on November 24, 2009 at 3:06 am

    As a non-English reader, I can not understand point 10. Can you make it readily understood?

  6. dqlover on November 24, 2009 at 3:03 am

    As a non-English speaker, I can not understand point 10.

  7. Duygu on November 23, 2009 at 7:53 pm

    Nice list, will be useful for especially new students!

    One of the things that help me is I do not chit chat when I am working. This seems to be a very obvious thing but by experience and by also observing what others experience, you do not realize how much chatting distracts you. I just kindly tell people that I cannot talk to them at that point if someone comes and they want to talk to me while I am doing experiments.

    I definitely agree with number 8 and 9 too! I do a lot of microsurgeries on chick embryos and sometimes they do not grow fast and I realized I am going to end up being in the lab till midnight. In that case I go home in the middle of the day, relax, eat and go back to lab all refreshed.

  8. Christopher Dieni on November 23, 2009 at 3:50 pm

    #2 most definitely hits home for me! In my graduate lab we had a system of protocols which we called our “Technical Bulletin” system, and it was organized into two letter, two number systems. For example, EB-02 would be the second protocol in our electroblotting use, and PK-14 would be the fourteenth protocol in assaying or manipulating protein kinases, and so on. It gave the two benefits you mentioned (forcing one to read through the whole protocol and getting a standardized copy for your use) and additionally we used them to a third major benefit: the trick was the write these technical bulletins in such a way that a student in the lab who had virtually no experience with a given protocol, could pick up a “techbull” and read through it, and be able to go from beginning to end without any problems. They were put to the test too; if a new student did have problems with the techbull as it was written, then a re-write would be done to make it even more user-friendly for future students.
    I myself wrote or revised over twenty of these techbulls in the course of my grad studies.
    Additionally, I of course fully support and agree with #8 and #9 as well!

Leave a Comment