MET Speaking Task 1, describe a picture: a timed route through the scene

Task 1 is the first thing you say in MET Speaking, and it comes with a clock, a microphone and no examiner. The skill is not spotting things in a picture. It is keeping an ordered stream of description going for 60 seconds, from the first second to the last. This guide covers the official facts, then a route you can run on any scene, a worked example, the mistakes that waste time and some practice scenes.

What does MET Speaking Task 1 ask you to do?

You see a picture of a situation with plenty to talk about. An instruction appears on screen and is read aloud, then you hear "Begin speaking now" and describe what you can see in as much detail as you can. Michigan's own sample instruction in the Speaking Resource Pack is only a few words long ("Describe the family meal"), so the picture is the content and the instruction is just the prompt to start.

Task 1 opens Part One, which has three 60-second tasks on one general topic. Tasks 2 and 3 ask for a personal experience and an opinion about that same topic, so the vocabulary you use here can carry forward. See Task 2 and Task 3.

Michigan uses two names for this task: "Describe a picture" in the Test Taker Handbook and "Describe a visual scene (image)" in the Speaking Resource Pack. They are the same task. For the other four tasks and the timings in one place, see MET Speaking, task by task.

Checked 15 and 16 October 2026 against Michigan's Speaking Resource Pack, March 2026 and Test Taker Handbook. ScoreFluent is an independent practice platform and is not affiliated with Michigan Language Assessment.

Is there any preparation time before you speak?

Michigan's Speaking Resource Pack describes no separate planning period. Each task appears on screen and is read aloud, you hear "Begin speaking now", the recording starts automatically, and it runs for the full time allowed. Anything you say after the time is up is not recorded and does not count towards your score.

That makes the read-aloud your only planning time. It happens before the timer starts, so it does not cost you any of your 60 seconds. This is our reading of the process, not a Michigan statement, but it is a sensible way to use those seconds:

1. Scan the whole picture once and find the main action.
2. Choose where you will start and which way you will move.
3. Decide your first sentence, so you do not start with a pause.

What are raters looking for in a picture description?

Trained raters score MET Speaking on three criteria, each rated 0 to 4. The criteria names are Michigan's. How they show up in this task is our reading of the published scale.

• Task Completion. Michigan lists relevance, the quantity of language you produce, and your ability to elaborate with relevant supporting detail. Its scale separates "extensive supporting detail" at the top rating from "general details but not original or extensive" one rating below. In a picture task, relevant means the picture in front of you, not your own life.
• Language Resources. Range and accuracy of vocabulary and grammar, including complex sentences used with control. Describing a scene naturally calls for position words, present continuous verbs and linking words, so use them on purpose.
• Intelligibility / Delivery. Fluency, hesitation, rhythm and pronunciation. Smooth delivery with little hesitation, clear and easy to understand, is how the scale describes the top rating.

Michigan's own tips for this task are specific. Use vivid verbs and adjectives, prepositions to say where things are, and adverbs to say how things are being done. Add what is happening and suggest why. Join ideas with connectors instead of listing them. For how the ratings turn into a 0 to 80 section score, see how MET is scored.

How do you describe a picture for 60 seconds without running dry?

Michigan suggests imagining a camera moving around the scene, giving details and then moving on to the next point of interest. Here is one way to turn that into a route. The order and the timings are ScoreFluent's technique, not an exam rule, so adjust them until they feel natural.

1. Frame, about 8 seconds. Say where this is, who is there and what the main action is. One or two sentences.
2. Foreground, about 15 seconds. The main people or objects: what they look like, what they are wearing, what they are doing right now.
3. Middle ground, about 15 seconds. The next group of people or objects, placed in relation to the first.
4. Background and small details, about 12 seconds. Walls, shelves, signs, objects, weather. Small details are where extra seconds come from.
5. Inference, about 10 seconds. What might be happening, why, and how people might be feeling.

Keep a small language kit ready:

• Where things are: in the foreground, in the background, on the left, next to, behind, above, in front of.
• What is happening: "is holding", "are waiting", "is reaching for", all in the present continuous.
• Careful guesses: "seems to be", "might be", "probably", "it looks as if".
• Moving on: "next to that", "meanwhile", "further back".

Two warnings. First, stay in one tense. Michigan advises keeping your tenses consistent, and a mix of present simple and present continuous inside one description is easy to avoid. Second, do not start a long story about one person. The task is a description, and the details come from moving across the scene.

Worked example

This scene is from the ScoreFluent practice bank. The bank's own description of the picture is: a pet shop; a girl kneels in front of a glass pen with two puppies inside; a shop assistant in a green apron stands beside her and smiles; a shelf with bags of pet food stands behind them; a fish tank is on the wall.

The answer below was written by ScoreFluent to show the route. It is not a Michigan sample and it has not been rated.

This is a pet shop, and the main action is a girl looking at some puppies. In the foreground, the girl is kneeling in front of a glass pen. There are two puppies inside, and she is leaning close to them, so I think she is deciding whether to buy one. Next to her, a shop assistant is standing and smiling. She is wearing a green apron, and she seems to be helping the girl. Behind them there is a shelf with bags of pet food on it, and I can also see a fish tank on the wall, so the shop sells more than dogs. Overall, everyone looks relaxed and friendly. The girl might be asking about the price of the puppies, and the assistant is probably telling her how to look after them.

What each part does: the first sentence frames the scene, the middle sentences move foreground to background with position words, and the last two make careful guesses. The answer uses "I think", "seems to", "might" and "probably" for the guesses, and it never claims to know what is said.

What mistakes cost marks in Task 1?

• Listing nouns. "A girl, a dog, a shelf, a tank." Every item should get a place or an action, or you are naming objects rather than describing a situation.
• Stopping early. Michigan's advice is to say as much as you can. Unused seconds cannot be banked, so move to the next zone instead of stopping.
• Starting every sentence with "I can see". Vary the openings with position words and the names of what you are describing.
• Long silent searches for one word. Describe around it ("the glass box the puppies are in") and keep moving. Michigan also notes that correcting yourself shows control, so a quick fix is better than freezing.
• Inventing a backstory. Names, ages and long histories use up time and are not in the picture. Keep guesses short and hedge them.
• Switching tenses. If you begin with "is kneeling", do not drift to "knelt" half-way through.

What practice scenes can you try for Task 1?

Practice items from the ScoreFluent bank (not real exam questions). The pictures themselves appear in the app. Each scene comes with an instruction such as "Look at the picture carefully. Describe what you see in as much detail as possible. Talk about the people, the place, and what is happening." Here is what the four scenes show:

1. A girl looks at puppies in a pet shop.
2. Customers browse in a bookstore.
3. Friends celebrate a birthday with a cake.
4. People wait in a doctor's waiting room.

The bank has no model answers for these scenes, so the worked example above is ours. Our blog set MET Speaking practice questions uses a different scene, a street market, so you can run the route there too.

How do you practise Task 1 on your own?

Michigan's Speaking Resource Pack suggests this practice activity. Set a timer and describe a picture for 60 seconds, recording yourself. Hide the picture, listen back and try to draw the scene from your own description. Then compare your drawing with the original. Missing or wrong details show where your description was too general.

Add these ScoreFluent practice activities:

• One zone at a time. Spend 20 seconds on the foreground only, then 20 on the background only. Then join the zones into one 60-second run.
• Two passes. On the first pass, name what is there. On the second, say how things relate and what might be happening.
• Any picture, anywhere. A photo in a magazine, a station platform, a shop window. Describe it aloud for 60 seconds.

Reading a model route helps, but only a recording shows whether your own description held together on the clock. Start practising MET free and you get 40 free credits. Credits never expire and there is no subscription. A Speaking examiner report costs 8 credits. Results are practice estimates, not official MET scores. See pricing for the full details.

Next in Part One: Task 2, talk about a personal experience. Back to the Michigan English Test overview or the full list of MET task guides.