Something in my research failed or went badly wrong
Steady yourself, work out what can be saved, find the cause and bring your supervisor options rather than only bad news.
In short
Start here, today. Before you change anything, write down exactly what happened while it's fresh: what you did, when, with which materials, equipment or contacts, what you expected and what you saw. Include anything unusual, however small. This takes about 20 minutes.
After that, work through these steps. Choose one to jump to it.
Step by step
Tick each step as you finish it. Open a step for the how.
0 of 6 steps done
Pause before you react
Don't throw anything away, rerun the whole thing or send a panicked email in the first hour. Secure what's left: freeze what can be frozen, save logs, photograph the setup, back up files. The exception is anything affecting other people or safety, such as a freezer alarm, a spill or lost participant data, which you report straight away.
Separate what's lost from what's salvageable
Make two lists. More often survives than it first seems: some replicates were fine, data up to a certain date is intact, a cut-short archive visit still produced usable material, or another site is already covered by your approval. Knowing exactly what remains turns a disaster into a specific gap.
Troubleshoot one thing at a time
List every possible cause, from materials and equipment to protocol, timing, handling and data processing. Rank them by how likely and how cheap they are to check, and test them in that order, changing one variable at a time. Ask the lab manager, a technician or a senior student who has used the method, because they have often seen this exact failure. For a lost site or partner, the equivalent is finding out why it fell through and whether the same thing could happen elsewhere.
Tell your supervisor early, with options
Within a day or two, send a short note: what happened, what's salvageable, what you think caused it, and two or three ways forward with a rough sense of the time each would take. Supervisors have seen plenty of failures. What frustrates them is hearing about one late.
Write the failure up properly
Record it in your lab notebook, field notes or analysis log as if someone else will need to understand it. Failed approaches often become a paragraph in your methods chapter and sometimes a finding: a protocol that doesn't transfer to a new population, a source that isn't what the catalogue claims, a site where access works differently than the literature suggests. Log your decision about what comes next in the Decision Log.
Adjust the whole plan, not just next week
Once the cause is clearer, look at what the loss means for your timeline. Update your Milestone Map, decide whether the scope needs to change, and check whether any funding or program deadlines are affected.
Words you can use
Is this you? Signs to look for
- An experiment or batch didn't work, and you can't tell whether it was the method, the materials or you.
- Samples were lost, contaminated or mislabelled, or a freezer, server or instrument failed.
- A field site, archive visit or partner organisation fell through after you'd planned around it.
- You keep replaying what you should have done differently, and you haven't told your supervisor yet.
Why this happens
Most PhDs include at least one week like this. Protocols that work in one lab fail in another, reagents vary between batches, instruments break, partners withdraw, and archives turn out not to hold what the catalogue suggested. Knowing that doesn't stop it feeling like a personal failure, especially when months of work sit behind it.
The first hours matter. Panic pushes people towards one of two mistakes: rerunning everything straight away without understanding what went wrong, or hiding the problem until it's fixed. Both cost time. A slower, systematic response usually gets you back on track sooner, and the reasons something failed are often worth knowing in their own right.
Worth a look (2)
- Sage Research MethodsBooks, case studies, videos and practice data sets on research methods, from interviews to regression. Your library usually provides access.
- PRISMAThe reporting guideline for systematic reviews and meta-analyses, with the checklist, flow diagram templates and extensions for scoping and other reviews. Many journals expect you to follow it.
When to bring someone in
For technical failures, ask your lab manager, a technician, core facility staff or a senior student; troubleshooting help is usually given generously. If identifiable participant data has been lost, stolen or exposed, contact your research ethics office (and your institution's privacy office, if there is one) straight away, because most institutions have a set procedure. If the loss puts a funding or program deadline at risk, talk to your graduate coordinator. If it has shaken you more than you expected, your counselling service is a reasonable place to take that.
Find the right person, or talk to Ify