The most common mistakes when choosing a thesis topic and how to avoid them

You choose your topic once, but you may live with it for a whole year. A bad choice rarely shows straight away. It shows when a company won't hand over the data it promised, when you can't find any literature, or when your supervisor tells you three months in that this isn't really what your field deals with. The mistakes below come up across all fields, and you can catch each one with an evening or two of work.

1. Choosing a topic because of how it sounds

"Artificial intelligence in medicine" sounds great on a slide, but on its own it isn't enough. It's just as risky to pick a topic only because a friend recommended it or because it looks like the smallest amount of work. Interest you didn't have at the start won't suddenly appear after a few weeks of reading academic papers.

Try a simple question: would you read one article about this voluntarily, tonight? If not, keep looking. Other good signs are a course you got more out of than you needed for the exam, or a problem you keep running into at work.

2. A topic so broad it has no end

This may be the most common mistake of all. A broad topic feels safe because there are plenty of sources. But then there are so many that you can't cover them, and you end up with a shallow overview and nothing of your own.

You narrow a topic by adding a specific group, place, period or angle:

• Broad: Social media marketing
• Narrower: Instagram marketing by small cafés
• Narrow enough: How small cafés in Brno use Instagram to win regular customers

The last version tells you who to contact and what to find out. That's a sign you can actually start working with it.

3. Checking the literature when it's too late

The opposite problem affects topics that are so narrow or so new that hardly anyone has written about them. Before you commit, spend an hour on a rough search in Google Scholar and your school's library catalogue. If an hour later you don't have at least a few relevant academic sources, the topic probably needs rewording. There's a fuller method in How to find and verify academic sources.

Also look at theses.cz and your school's repository. Finding several similar theses isn't the end of the road, but you'll need to be clear about what you're adding. A different group of respondents or newer data is sometimes enough.

4. Counting on data you can't get

A thesis about the recruitment strategy of a particular bank sounds great until the bank refuses to share anything. Plans to interview the CEOs of large companies go the same way if you don't know any of them.

Before you get attached to a topic, check three things. Can you access the data, ideally with written consent? Can you collect and analyse it before the deadline? Do you know the method you'll need, or can you learn it in time?

As for the popular question of how many respondents you need, there is no universal number. It depends on the method, the field and what your department expects, so ask your supervisor directly.

5. Having a topic but no question

"Employee motivation" is an area, not a proposal. Until you can say exactly what your thesis should answer, you don't know what to read or what to measure. Try finishing this sentence: "I want to find out whether/how/why..." If you can't, the topic isn't ready yet.

Not every thesis needs a hypothesis, by the way. Hypotheses are common in quantitative research, while a qualitative study or a practical project can often work with research questions or a clear objective. How to write a thesis objective and research questions shows how to word them.

6. Talking to your supervisor and reading the rules too late

Some students spend a month on a topic and only then find out they have to pick from a list of approved topics, that their preferred supervisor has no places left, or that the deadline for submitting the proposal has passed. So read your faculty's rules and dates first.

Approach a supervisor when you have two or three reasonable options, rather than a final decision or nothing at all. Say what interests you, where you'd get data and what you've already found. And if they talk you out of something, ask why. They may see a problem you can't see yet. If you get to choose your supervisor, ask older students what working with them was like.

A quick check before you decide

Your topic is in good shape if you can say in one sentence what you're investigating and among whom, you know where the data will come from, you've found a few academic sources, and your supervisor has no fundamental objections. For the full process from the start, see How to choose a thesis topic.

If you're stuck at the very first step and don't know where to start, Thesios.ai can help. Based on your field and interests, it suggests 10 narrow topics for free. Treat them as inspiration: you still have to check that they're doable and agree on them with your supervisor.