Two written examples from the same fictional thesis defense: the complete opening of the presentation and a question-and-answer block from the discussion. The student, company, and numbers are fictional; the technique is directly transferable. Structure, flow, and slide logic are explained in thesis defense.
Example 1: The opening of the presentation
Situation: Undergraduate business thesis defense. Topic: “The influence of click and collect on purchase frequency in brick-and-mortar retail: a quantitative analysis using three home improvement stores.” Student Jacob Reed has 15 minutes to present; these are the first two minutes.
Professor Aydin, Dr Collins, thank you for the opportunity to present my thesis today.
In 2024, home improvement retail generated around eleven percent of its revenue online, and almost half of that came through click and collect: customers order online and pick up in store. The sector celebrates this as a bridge between online and physical retail. Whether that bridge actually brings customers into the store, or simply shifts orders that would have happened anyway, has barely been studied for this sector. That is where my thesis begins.
My research question is: what influence does the introduction of click and collect have on the purchase frequency of existing customers in physical home improvement stores? My answer upfront: a measurable positive influence, though much smaller than sector forecasts claim. Across the three Hagedorn Home stores examined, purchase frequency among click-and-collect users rose by an average of 14 percent during the study period; supplier marketing claims promise two to three times that.
I will show how I reached this result in three steps. First, my research design: loyalty-card data from 4,800 existing customers over eight weeks, analysed using a difference-in-differences approach. Then the two main findings. Finally, what the result means for investment decisions by mid-sized retailers, and where the limits of my data lie.
Why this speech works: The greeting is one sentence, then the opening moves straight to a sector number that proves relevance. There is no warm-up about motivation or outline. The research question appears word for word, and the answer follows in the next sentence with the strongest number from the thesis: 14 percent against “two to three times that.” Examiners know the work already, so manufactured suspense would waste time. The roadmap at the end has exactly three stops and announces limitations voluntarily. That removes the first attack point before the questions begin.
Example 2: Three critical examiner questions, answered calmly
Situation: Question round in the same defense, around minute 25. Second examiner Dr Collins asks the methodologically critical questions. The technique behind each answer: concede the valid core, place the limitation precisely, then return to a strength of the work.
Question 1: “Mr Reed, your sample comes from three locations in one chain. What do your findings say about retail overall?”
About retail overall: very little, and I acknowledge that directly. I name it in chapter 6 as a central limitation. My findings initially apply to home improvement stores with loyalty-card systems in mid-sized towns. However, I chose the three stores deliberately because they differ clearly in size and catchment area. The effect appears at all three in a similar range. That makes me confident that the finding holds within this sector. For other sectors, it would be a hypothesis for a follow-up study.
Question 2: “Why a purely quantitative design? Customer interviews would tell you why frequency rises.”
That is true: my data cannot answer the why, only whether and by how much. I still chose the quantitative design because the gap in the research was exactly at the whether. The sector forecasts I cite in chapter 2 are based on surveys about purchase intentions, and intentions systematically overstate real behaviour. Loyalty-card data measures what customers actually do. An interview study on the why would be the logical second step. I outline it in the outlook.
Question 3: “Your eight weeks are in spring, the peak season for home improvement stores. How do you know you are not measuring a seasonal effect?”
The season does lift all figures in my study period. That is exactly why a simple before-and-after comparison would have been worthless here. I chose the difference-in-differences approach for that reason: I compare click-and-collect users with a control group of existing customers from the same stores in the same period. The spring effect affects both groups and therefore cancels out. What I cannot rule out is that the effect would be smaller in winter. A repeat measurement in the fourth quarter would be the clean test.
Why this speech works: All three answers follow the same three-step pattern. Concede: the valid core of the criticism is admitted in the first sentence, without evasions. That lowers the tension and shows the student assesses the work realistically. Place: the limitation is then located precisely: what the results apply to, what they do not apply to, and why the decision was still sound. Return: each answer ends at a strength of the work or a concrete follow-up study, never in retreat. Notice what is missing too: no automatic flattery of the examiner, no defensive tone, no answer longer than 30 seconds.
What both examples have in common
The opening puts the research question and answer into the first two minutes and announces its own limits. The question round turns those same limits into evidence of methodical care. Anyone who prepares both parts together, the talk and the ten most likely examiner questions, enters the room with a rehearsed home advantage rather than a cross-examination. How to get there, slide by slide and question by question, is explained on the thesis defense page. There, eloqole turns your research question, findings, and time limit into the complete presentation with answer frameworks.