How the German Hören (listening) exam works from A1 to C1, why play counts matter, and the strategies that turn listening from your weakest section into a reliable one.
Listening feels unfair to a lot of candidates. The audio plays, and if you miss a phrase there is no going back and re-reading. But the Hören section is more predictable than it sounds. The task types repeat at every level, the recordings follow set patterns, and once you know how many times each part plays, you can build a plan that catches the answers instead of chasing them. This guide covers the Hören section from A1 to C1 and the strategies that actually move your score.
Exam formats last verified July 2026 against the Goethe-Institut Durchführungsbestimmungen (Stand 1 September 2025) and the current interactive Modellsätze.
What the listening section looks like at each level
Listening runs about 20 minutes at A1 and around 40 minutes from B1 upward. The number of parts and the difficulty of the audio grow with the level. Every part has 30 items in total from B1 up, so no single answer is worth much on its own. That is good news: you can miss a few and still pass, as long as you keep moving.
| Level | Time | Parts / items |
|---|---|---|
| A1 | ~20 min | 3 parts, 15 items (pictures, true/false, short conversations) |
| A2 | ~30 min | 4 parts, 20 items |
| B1 | ~40 min | 4 parts, 30 items |
| B2 | ~40 min | 4 parts, 30 items |
| C1 | ~40 min | 4 parts, 30 items, all multiple choice |
What each part actually contains
The parts are not random. Each Teil tests a different listening skill, and knowing the shape tells you what to listen for before the audio starts.
A1
Three parts, 15 items. Teil 1 gives short conversations with a three-picture choice for each item. Teil 2 uses public announcements with true/false statements. Teil 3 returns to short exchanges with a three-option choice. You are catching one fact per item: a price, a platform number, a time.
A2
Four parts, 20 items. Teil 1 and Teil 3 are multiple choice on short dialogues and messages. Teil 2 matches spoken content to pictures. Teil 4 shifts to Ja/Nein: you decide whether a statement matches an interview. The jump from A1 is the interview length, not the vocabulary.
B1
Four parts, 30 items. Teil 1 mixes true/false with multiple choice over short everyday recordings. Teil 2 is a single longer monologue with three-option questions. Teil 3 is true/false on a conversation. Teil 4 is speaker attribution: you assign statements to a discussion between two people. The B1 exam guide walks through the full module timing if you want the wider picture.
B2
Four parts, 30 items. Teil 1 blends true/false and multiple choice. Teil 2 is a three-option monologue. Teil 3 is speaker attribution on a discussion. Teil 4 is three-option multiple choice on an interview or lecture. The content is more abstract than B1, so opinions and reasons matter more than plain facts.
C1
Four parts, 30 items, all three-option multiple choice. The audio is longer, denser, and closer to real radio and lecture speech. There are no easy warm-up items here. The C1 exam guide covers how the module fits with the rest of the exam.
The detail that changes everything: play counts
Not every part is played twice. In the current formats, some parts play once and some play twice, and which is which is fixed per level. This is the single most useful thing to memorise, because it decides your strategy for each part.
| Level | Play pattern by Teil |
|---|---|
| A1 | Teil 1: 2× / Teil 2: 1× / Teil 3: 2× |
| A2 | Teil 1: 2× / Teil 2: 1× / Teil 3: 1× / Teil 4: 2× |
| B1 | Teil 1: 2× / Teil 2: 1× / Teil 3: 1× / Teil 4: 2× |
| B2 | Teil 1: 1× / Teil 2: 2× / Teil 3: 1× / Teil 4: 2× |
| C1 | Teil 1: 1× / Teil 2: 2× / Teil 3: 1× / Teil 4: 2× |
Notice the pattern flips as you go up. At A1, A2, and B1 the first part plays twice. At B2 and C1 the first part plays only once, so the exam hits you with a single-listen task while you are still settling in. Rehearse to your own level's pattern, not a generic one.
- Played once (often the longer monologue or the opening part at B2 and C1): your first and only listen must count. Read the questions in the pause beforehand so you know exactly what to catch, and commit to an answer as you hear it.
- Played twice: use the first listen for the general answer and the second to confirm the detail and fix anything you missed. Do not leave every item for the second play, because the recording moves at the same speed both times.
Turn every question into a listening target
Before each part you get seconds to read the questions. This is not filler time. Turn each question into a specific thing you are waiting to hear. Underline the key noun, and predict the kind of answer: a time, a reason, a price, an opinion.
Here is how that works on a real item. Suppose the question reads: "Warum kommt Herr Berger zu spät?" with options (a) der Zug hatte Verspätung, (b) er hat den Bus verpasst, (c) das Auto ist kaputt. Your target is now narrow. You are listening for a reason tied to Herr Berger and lateness, and you already expect it to involve transport. When the speaker says something about the train, you check it against option (a) and hold. If he then adds that he took the car instead because the train was cancelled, the answer is still about the train, but a careless listener grabs "Auto" and picks (c). Reading the options first turns a wall of speech into three things you can test.
Note, do not translate
The moment you start translating in your head, the audio has already moved on. Train yourself to note answers in shorthand: a number, a keyword, a tick. Then keep listening. Comprehension at exam speed is about staying with the flow, not decoding every word. A half-caught note you can act on beats a perfect translation that arrives two sentences late.
The traps built into listening
The recordings are written to catch people who match words instead of meaning. Four traps show up again and again.
- Distractors. Recordings often mention a wrong option first, then correct it. A speaker says the meeting is at three, then adds "nein, warte, doch erst um vier." Wait for the speaker to finish the thought before you commit. The first number you hear is often the bait.
- Negation. A single "nicht", "kein", or "nie" flips the meaning. "Ich fahre nicht mit dem Zug" is the opposite of what a word-matcher hears. Listen for the negation as hard as you listen for the keyword.
- Paraphrase. The right answer rarely uses the exact words of the recording. If the audio says "die Veranstaltung wurde verschoben" the correct option might read "der Termin findet später statt." Match meaning, not vocabulary. Options that repeat the recording word for word are frequently the trap.
- Speaker attribution. In the "who says what" parts at B1 and B2, two people hold different positions. Track each speaker separately, or the views blur together. Jot the speaker's initial next to each opinion as you hear it, so you are not reconstructing the whole discussion from memory at the end.
How to practise listening
Textbook audio is too slow and too clean. Train on authentic German at natural speed: news in slow-then-normal German, podcasts, interviews, and the official model-set recordings for your level. Listen actively, pausing to summarise what was said, rather than letting it wash over you.
Then sit full timed listening modules so the once-only parts stop rattling you. The skill that separates a pass from a near-miss is not vocabulary, it is holding your nerve through a single-play part while your pen keeps up. That only comes from reps under real conditions. Working through real exam-format listening sets, the kind you can practise in SagaDeutsch mock exams, builds the stamina and pattern recognition the section rewards. If you drill Hören alongside the reading section, the same target-setting habit carries over; the Lesen reading playbook covers that side.
Frequently asked questions
Yes, note on your question sheet as you listen. Keep it to shorthand so you do not fall behind the audio. Transfer answers to the answer sheet in the pauses the recording gives you, not while a new item is playing.
Let it go and lock onto the next question. Chasing a missed item usually costs you the next one too. With 30 items in the module, one blank answer is survivable; two in a row because you froze is the real damage.
Recordings use clear standard German, occasionally with mild regional colour. Exposure to varied authentic audio in practice removes the surprise, so a slightly different voice does not throw you on exam day.
You are not meant to understand every word. Focus on the specific thing each question asks, and infer the rest from context. A single unknown word almost never decides an answer; the meaning around it does.
Speaker attribution loads your memory more than other tasks, because you hold several positions at once and match them to statements. The fix is mechanical: label each opinion with the speaker as you hear it, and match afterwards. Do not try to remember who said what across the whole recording.
Answer as you go, especially on single-play parts. On parts played twice, pencil an answer during the first listen and confirm it on the second. Waiting until the end means reconstructing the recording from memory, which is exactly what the exam is testing you not to rely on.
