WEBVTT 1 00:00:00.400 --> 00:00:37.826 Hello, and welcome to Practise Out Loud, English job interview practice from EnglishJobReady. I am an AI practice interviewer. This is an AI-generated voice, not a real person and not a real recruiter. In this episode, we practise five interview questions for a data analyst job in Germany. After each question, you will hear a short tone. Then you have about half a minute to answer out loud. After the second tone, you will hear one example of a strong answer, and one tip. If you cannot speak right now, say your answer quietly in your head. Let's begin. 2 00:00:38.726 --> 00:00:48.784 Question one. How do you document your analysis so another person can repeat it? In simpler words: How do you write down your analysis so others can repeat it? 3 00:00:49.484 --> 00:00:54.790 Now answer out loud. You have about twenty-five seconds. Start after the tone. 4 00:00:55.540 --> 00:01:20.540 [Your turn: 25 seconds of quiet to answer out loud, between two short tones] 5 00:01:21.390 --> 00:01:42.712 Thank you. A strong answer could sound like this: In my last job, I wrote each step clearly and kept my code in a shared repository. I noted the data sources, the dates and the versions I used, and I wrote down my assumptions. I stored everything in one folder. As a result, a new analyst repeated my monthly report in two hours, without my help. 6 00:01:43.512 --> 00:01:55.963 One tip. What they really want to know is this: Is your work reliable and reproducible? Two things to cover: clear steps and code, and data sources and versions. 7 00:01:57.363 --> 00:02:09.177 Question two. A senior leader wants one number to track a new project. How do you choose it? In simpler words: A leader wants one number to track a project. How do you choose it? 8 00:02:09.877 --> 00:02:15.183 Now answer out loud. You have about twenty-five seconds. Start after the tone. 9 00:02:15.933 --> 00:02:40.933 [Your turn: 25 seconds of quiet to answer out loud, between two short tones] 10 00:02:41.783 --> 00:03:01.547 Thank you. A strong answer could sound like this: First, I would ask the leader what the project needs to achieve. Then I would suggest one clear measure, like weekly active users for a new app. I would explain what the number does not show. I would set up a simple weekly report for it. 11 00:03:02.347 --> 00:03:13.909 One tip. What they really want to know is this: Can you guide leadership to the right measure? Two things to cover: understand the goal, and suggest one clear measure. 12 00:03:15.309 --> 00:03:24.559 Question three. Tell me about an analysis you did that changed a decision. In simpler words: Tell me about your analysis that helped a decision. 13 00:03:25.259 --> 00:03:30.942 Now answer out loud. You have about thirty seconds. Start after the tone. 14 00:03:31.692 --> 00:04:01.692 [Your turn: 30 seconds of quiet to answer out loud, between two short tones] 15 00:04:02.542 --> 00:04:23.016 Thank you. A strong answer could sound like this: My manager asked why customer returns were rising. I joined the sales and returns data and found that most returns came from one product size. I showed the finding in one simple chart. The team changed the size guide on the website, and returns for that product went down. 16 00:04:23.816 --> 00:04:33.285 One tip. What they really want to know is this: Do you deliver insight, not only queries? Two things to cover: the business question, and the data and method. 17 00:04:34.685 --> 00:04:45.225 Question four. Walk me through how you would approach a take-home task with messy data. In simpler words: How would you work on a task with messy data? 18 00:04:45.925 --> 00:04:51.231 Now answer out loud. You have about twenty-five seconds. Start after the tone. 19 00:04:51.981 --> 00:05:16.981 [Your turn: 25 seconds of quiet to answer out loud, between two short tones] 20 00:05:17.831 --> 00:05:37.675 Thank you. A strong answer could sound like this: First, I would make sure I understand the business question. Then I would check the data for missing values, duplicates and odd numbers, and I would write down what I changed. I would start with a simple method. At the end, I would give a clear summary and explain the limits. 21 00:05:38.475 --> 00:05:51.063 One tip. What they really want to know is this: Can you structure a problem and explain your choices? Two things to cover: understanding the question, and cleaning and checking the data. 22 00:05:52.463 --> 00:06:02.377 Question five. How do you explain a technical finding to a manager who does not work with data? In simpler words: How do you explain data results to a manager? 23 00:06:03.077 --> 00:06:08.383 Now answer out loud. You have about twenty-five seconds. Start after the tone. 24 00:06:09.133 --> 00:06:34.133 [Your turn: 25 seconds of quiet to answer out loud, between two short tones] 25 00:06:34.983 --> 00:06:51.781 Thank you. A strong answer could sound like this: I start with the answer in one sentence. Then I show one clear chart. I explain what it means for their team and what they could do next. I also say what the data cannot tell us. 26 00:06:52.581 --> 00:07:03.695 One tip. What they really want to know is this: Can you make data useful for decision-makers? Two things to cover: start with the answer, and one clear chart. 27 00:07:05.095 --> 00:07:29.653 That was five questions for a data analyst job in Germany. Well done for practising out loud. If you want to know what to fix next, get your free English Interview Readiness Score at englishjobready dot com. You practise a short, realistic interview and see what to fix next. This was EnglishJobReady's AI practice interviewer. Thank you for listening, and see you in the next episode.