CELPIP Speaking Task 4: Make Predictions That Score CLB 9
Task 4 shows you two pictures and asks you to predict what will happen next. Most test-takers state one obvious guess and stop. Below: the speculative framework that fills 60 seconds with scored content, real sample answers at three levels, and the language patterns that separate CLB 7 from CLB 9.
What Is CELPIP Speaking Task 4? (Making Predictions)
In Task 4, you see a series of pictures that tell a story, and you must predict what will happen next. You have 30 seconds to prepare and 60 seconds to respond. The challenge is not having the right answer — there is no right answer. The challenge is building a well-reasoned, developed prediction.
Preparation time
30 seconds. Study the sequence of events in the pictures. Identify the cause-and-effect chain. Your prediction should logically follow from what you see — not be a random guess.
Response time
60 seconds. You need more than one prediction. Describe what you see, make your main prediction, then offer an alternative or explain consequences. One sentence is not enough.
What examiners score
Logical reasoning (does your prediction follow from the pictures?), developed ideas (not just a single guess), speculative language range, and natural fluency. Examiners do not care if your prediction is "correct."
How to Use Your 30 Seconds (Prep Strategy)
Your prep time should focus on understanding the story in the pictures and planning at least two prediction angles. The CLB 9 strategy guide covers how to apply logical reasoning across all speaking tasks.
Seconds 1–10: Read the picture sequence
Look at each picture in order. What is happening first? What changed? What is the problem or situation developing? Identify the trajectory before you predict where it goes.
Seconds 10–20: Choose your main prediction
What is the most logical outcome? "Based on the pictures, it looks like the man is going to miss his flight because he is stuck in traffic." Your prediction must connect to what you see — not come from nowhere.
Seconds 20–30: Plan an alternative or consequence
"On the other hand, he might call a taxi and make it just in time." Having two angles ensures you fill the full 60 seconds and demonstrates speculative language range — which is exactly what examiners score.
The Prediction Structure That Scores CLB 9
This four-part structure ensures your prediction is developed, logical, and fills the full response time. The same reasoning patterns apply in Task 5 (comparison) and Task 7 (opinion) too.
1. Summarize what you see (10–15 sec)
"In the first picture, a man is rushing out of his house. In the second picture, he is stuck in heavy traffic." This shows you understand the situation — which is required before your prediction is credible.
2. State your main prediction (15 sec)
"I think what will happen next is that he will miss his appointment because the traffic looks completely stopped." Use speculative language: "I think," "it seems likely," "based on what I see."
3. Develop or explain why (15 sec)
"The reason I think this is because in the second picture, he looks stressed and the road ahead appears completely packed. There does not seem to be any way for him to get through quickly." Reasoning is what separates CLB 7 predictions from CLB 9.
4. Offer an alternative or consequence (10–15 sec)
"However, it is also possible that he might find a shortcut or call someone to reschedule. Either way, this situation will probably make him plan better next time." Hedging and alternative scenarios demonstrate advanced speculative control.
CELPIP Speaking Task 4 Sample Answers at Three CLB Levels
Same picture sequence — notice how predictions become more developed and reasoned at higher levels.
Prompt: Picture 1 shows a man waking up late. Picture 2 shows him stuck in traffic. What do you think will happen next?
"I think he will be late for work. Because there is traffic. Maybe his boss will be angry. That is what I think."
"Looking at these pictures, I think the man is probably going to be late for work. In the first picture, he clearly woke up late because he looks panicked, and in the second picture he is stuck in traffic, which makes the situation worse. I believe he will probably miss an important meeting or arrive at the office much later than expected. His boss might be upset with him, and he will likely feel stressed for the rest of the day. In the future, he will probably set an extra alarm to make sure this does not happen again."
"Based on what I can see in these pictures, it looks like this man is having a rough morning. In the first image, he has clearly overslept — his expression suggests he just realized how late it is. By the second picture, things have gone from bad to worse because he is now sitting in bumper-to-bumper traffic with no way out. I think what is most likely to happen next is that he will arrive at work significantly late and have to apologize or explain himself to his colleagues or supervisor. However, there is also a chance he might decide to call ahead and let them know about the situation, which would at least show some responsibility on his part. What strikes me is that this seems like it could be a turning point for him — the kind of stressful morning that finally motivates someone to change their routine. I would guess he will start setting a backup alarm or preparing things the night before to avoid this happening again."
Examiner-Level Score Analysis: Why Each Response Gets Its Score
Prediction tasks are scored on reasoning quality, development, speculative language, and fluency.
CLB 4–5: Why a Single Guess Cannot Pass
Reasoning
"Because there is traffic" is the only reasoning offered. There is no connection between the pictures and the prediction. The speaker jumps to a conclusion without showing any analytical thinking.
Development
Three short sentences and then silence. The prediction is a single idea with no elaboration, no alternative, and no explanation. The response ends in about 15 seconds — well under half the allowed time.
Speculative language
"I think" and "maybe" are the only speculative markers. There is no range: no "it seems likely," "there is a chance," or "based on what I can see." Examiners need to hear varied speculative structures for higher scores.
Fluency
Very short, choppy sentences followed by an abrupt "That is what I think" closing. The speaker sounds like they had one idea and could not develop it further.
CLB 7–8: Developed — But Linear and Predictable
This is a solid response with good time management. The CLB 9 strategy guide shows how to add the depth that pushes predictions past the CLB 8 ceiling.
Reasoning
References both pictures and draws a logical connection. But the reasoning is straightforward and predictable — "woke up late, stuck in traffic, will be late for work." Nothing unexpected or insightful.
Development
Multiple predictions (late for work, boss upset, set extra alarm). Good time management. But every prediction follows the same obvious trajectory — there is no alternative perspective or nuance.
Speculative language
"Probably," "I believe," "might," "likely" — adequate range. But the structures are formulaic: "I think X," "He will probably Y." CLB 9 responses vary their speculative structures more naturally.
Fluency
Fills the time well with smooth delivery. But the response sounds like a linear prediction list rather than a natural, thoughtful analysis. The tone does not shift between ideas.
CLB 9–12: What Top Predictions Actually Sound Like
Reasoning
Reads the pictures like a story: "his expression suggests he just realized how late it is," "things have gone from bad to worse." The reasoning shows genuine engagement with visual details, not just surface-level observation.
Development
Multiple angles: immediate consequence (late for work), mitigation (calling ahead), long-term impact (changing routine). The prediction builds progressively rather than listing isolated guesses.
Speculative language
"Based on what I can see," "it looks like," "there is also a chance," "I would guess" — varied, natural, and contextually appropriate. The speculative markers are woven into natural sentences rather than bolted on.
Fluency
Fills the full 60 seconds with natural pacing. The speaker adds personal observations ("What strikes me is") and shifts between prediction angles smoothly. The delivery sounds like genuine thinking aloud.
Common CELPIP Task 4 Mistakes That Kill Your Score
Task 4 mistakes are almost always about underdevelopment and lack of speculative language — not wrong predictions.
Making one prediction and stopping
"He will be late for work." That is one sentence — not a 60-second response. You must develop your prediction with reasoning, consequences, and alternatives. One prediction cannot score above CLB 5.
Stating facts instead of predicting
"In the first picture, a man is sleeping. In the second picture, there is traffic." Describing what you see is setup — not the task. Examiners want to hear what happens NEXT, not a summary of what already happened.
No speculative language
"He will be late. His boss will fire him." These are stated as facts, not predictions. Using "will" without hedging sounds like you are reporting, not speculating. Add "I think," "it seems likely," "there is a good chance" to every prediction.
Illogical jumps
"He is stuck in traffic, so he will probably move to a new city." Predictions that do not logically follow from the pictures cost task fulfillment points. Your prediction must be reasonable given what you see, even if it is speculative. If you struggle with connecting observations to predictions, a tutoring session can build this habit quickly.
Only using "I think" and "maybe"
Speculative language range matters. If every sentence starts with "I think" or "maybe," your vocabulary score hits a ceiling. Practice using: "it seems likely," "there is a good chance," "based on this," "I would imagine," "it is possible that."
How to Move from CLB 7 to CLB 9 in Task 4
If you can make reasonable predictions but your scores plateau at CLB 7–8, the problem is development and speculative language variety.
Stop doing this
Making one prediction and stopping
Describing the pictures without actually predicting
Using only "I think" and "maybe" as speculative markers
Making predictions that do not connect to what is in the pictures
Start doing this
Briefly summarize the pictures, then pivot to prediction
Make a main prediction AND offer an alternative
Explain WHY you think something will happen — reasoning is where CLB 9 points live
Use varied speculative language: "it seems likely," "there is a chance," "based on what I see"
The key insight: CLB 9 predictions are not more creative — they are more developed. The same prediction, with reasoning and alternatives added, scores two levels higher.
Continue Your CELPIP Speaking Preparation
← Task 3: Describing a Scene
Master organized picture descriptions with spatial structure.
Task 5: Comparing and Persuading →
Learn the comparison framework that convinces examiners.
Full CLB 9 Strategy Guide
The complete system for reaching CLB 9 across all CELPIP sections.
All Speaking Tasks
Return to the speaking hub for all 8 task pages.
CELPIP Overview
See the full preparation system with all skill hubs and resources.
Private Speaking Coaching
Get one-on-one feedback on your prediction skills and speculative language.
Still Struggling with Task 4? Get Targeted Help
If your predictions keep scoring CLB 7, the problem is usually development and speculative language range — not your ability to guess correctly.
Private Speaking Coaching
Your tutor gives real-time feedback on your prediction development, reasoning patterns, and speculative language variety.
Book a Speaking SessionWriting Correction
Speculative and analytical writing skills directly improve your speaking predictions. Building reasoning in writing transfers to spoken tasks.
Submit Your WritingWeekly CELPIP Webinar
Practice prediction tasks with other students and hear examiner-perspective feedback on what developed predictions sound like.
Join This Week