CELPIP Speaking Task 5 Template (Comparing and Persuading)

Three minutes in total, split across two parts: 60 seconds to choose, 60 seconds to prepare, 60 seconds to speak. Task 5 is the only CELPIP Speaking task built this way, and it is the longest single task on the test.

Those figures come from CELPIP's 2026 Speaking Pro study pack. Here is where Task 5 sits against the rest:

The practical consequence: you get twice the planning time of most tasks, and almost nobody uses it well. Sixty seconds to choose feels generous until you are sitting in front of it with a clock running and no microphone to fill.

Sixty seconds of speaking is roughly 130 to 150 words at a comfortable pace. That is enough for a greeting, three comparison points and a closing sentence. Not more.

In Part 1 you choose one of two options. In Part 2 you persuade someone that your choice is better than the option they are proposing. CELPIP's own description of the task makes one point explicit: there is no speaking at all in Part 1.

Hold on to that. Part 1 is silent. It is the single most useful fact about this task and most prep material skips it, including the AI summaries that appear above these search results.

CELPIP's overview material makes the same point in different words: the first screen is only a choice, and the second screen gives you a new preparation time and then the speaking time. Two screens, one voice recording.

Three things follow from the wording:

You choose, and then somebody disagrees with you. Part 2 is not a fresh start. Someone has come back with the other option, or a new one, and you are responding to a person who has already made a case.

There is no wrong choice. CELPIP does not say one option scores better than the other. Pick the one you can support with specific details, not the one you would actually buy.

"Persuade someone" means a person, not an essay. Greet them. Acknowledge their idea. Then disagree with reasons. A response that reads like a product review misses the task.

The task is also called Comparing and Persuading.

The real screen shows a short situation, then two options side by side, each with a picture, a few details and a price. Here is one in that format, written by us.

Part 1 (60 seconds, no speaking). Your sister is starting university in September and wants a laptop. Choose the one you would recommend.

• Option A: 14-inch ultrabook. 1.2 kg. Battery about 12 hours. $1,150.
• Option B: 15.6-inch laptop with a large screen and a number pad. 2.1 kg. Battery about 6 hours. $780.

Part 2 (60 seconds to prepare, 60 to speak). Your sister says she prefers Option B. It is cheaper, and she says she will use it at her desk anyway. Persuade her that your option is better.

Notice what Part 2 adds: she has a reason. A good answer answers her reason, not just the price tag. Here is a 60-second answer that argues for the more expensive option, using the template on this page (140 words):

Hi Mina. I saw that you prefer the bigger laptop, and I understand why, because seven eighty is a lot less than eleven fifty. However, I think the ultrabook is the better choice for your first year.

First, about the price, yes, it costs three hundred and seventy dollars more. But you said you will work at your desk, and from what I remember of first year, you will not. You will be in the library, in lectures and on the bus, and that is where the twelve-hour battery and the one-point-two kilos matter.

Also, the bigger laptop weighs almost a kilo more. Carrying that across campus every day gets old by October.

Finally, even though the number pad is handy, a ten-dollar external one solves that.

So I really think the ultrabook is the one. What do you think?

What makes this answer work is what CELPIP's own rated samples for this task reward, in our reading of them: three reasons that build on each other instead of three separate complaints, specific details that go past what the screen says (library, bus, October), varied sentences, a warm tone, and a finish inside the minute. A weaker answer reads the two prices back, says the ultrabook is "better" twice and runs out of time on the third point.

Choose fast, then start building your case with the time left over. CELPIP publishes six key strategies for this task, three for each part. Here they are in our words, with the place in this template where each one happens:

Source: CELPIP Speaking Pro Target 9 Study Pack (2026).

Note the word quickly. The choosing is not the work. Once you have picked, the remaining seconds are free planning time that most candidates waste sitting still because the screen has not asked them for anything yet.

Use them for three things: the two prices, said out loud in your head the way you will say them in Part 2, one contextual reason your choice fits the person in the scenario, and one weakness in the other option you can name politely.

The Part 2 strategies are the ones that separate scores, and the last of them most of all.

Going beyond the picture is the move. The prompt gives you a price and two or three features. A weak answer reads them back. A strong one adds what they mean for the person in the scenario.

What CELPIP actually rates on Task 5

CELPIP Speaking is rated on four dimensions: Content/Coherence, Vocabulary, Listenability and Task Fulfillment. Those four names come from CELPIP's Speaking Performance Standards, and they apply to all eight tasks.

Three things about this list, because most CELPIP prep gets them wrong:

Each response is rated separately on all four dimensions, and one hard task does not decide your level. CELPIP's 2026 study pack explains that your Speaking level comes from all eight tasks together, so there are many routes to the same level. No points formula is published, and no per-dimension weighting is published either. Task 5 has a reputation as the hardest task. It is still one of eight.

There is no separate Grammar score. Look at the Listenability row. Grammar is not one of the four dimensions, it is one item inside Listenability. That does not make the comparative rules below optional, because "more cheaper" is exactly the kind of error that pulls a listener out of your argument. But nobody scores your comparatives on their own line.

"Coherence and Cohesion", "Lexical Resource" and "Fluency and Pronunciation" are IELTS terms. If a Task 5 template promises to fix your Lexical Resource, it was written for a different test.

For how the four dimensions turn into a level, see the CELPIP Speaking score chart.

What strong Task 5 answers do well

These are not the official rating criteria. They are the three habits that tend to earn what the criteria reward on a persuasion task.

Try this on a real Task 5 question, free

Sign up free and answer the first Speaking Task 5 question on the real clock. You get up to three AI-scored tries, with feedback on what to fix before test day.

Answer a free Speaking Task 5 question

Each runs the same three points and lands inside the 60-second window. Notice that two of the three argue for the more expensive option, because that is the version most people never practise.

Almost certainly not, and chasing one specific scenario is the worst way to prepare for this task.

A handful of Task 5 scenarios circulate widely in study groups, Reddit threads and coaching classes, and people search for them by name hoping to find the "right" answer before test day. It is an understandable instinct. It does not work, for three reasons.

Nobody can confirm any of them. CELPIP does not publish a public bank of Task 5 prompts. What it does publish free is two full CELPIP-General practice tests, which between them contain two Task 5 items, plus the practice items inside the paid Speaking Pro study packs. Beyond those, any scenario described online as "the real one" is somebody's recollection. There is no way to verify it and no reason to build your preparation on it.

Start with those two free practice tests. They are the only Task 5 items anyone can point to and say where they came from.

A memorised answer is easy to hear. Task Fulfillment includes tone, and Vocabulary rewards precision fitted to the situation. An answer assembled for a different scenario and bent to fit this one produces exactly the generic adjectives the Context Rule above warns you about. Raters listen to eight of these responses in a row.

The structure transfers and the scenario does not. Every Task 5 prompt has the same skeleton: two options, at least one price, a person with a reason to prefer the other one. Greeting, soft rejection, price, contextual value, detail, close. That shape handles a hotel, a fridge, a dog breeder, a childcare centre or a car without any adjustment.

What to do instead. Practice the shape across many different scenarios until the opening comes out automatically, and force yourself to argue for the more expensive option half the time, because that is the version that catches people out. Our Task 5 question bank has tens of them, and it scores what you record, so you find out whether your third point actually landed or you ran out of things to say at second forty.

Learn the shape of these lines, not the exact words. A sentence repeated word for word sounds rehearsed, and raters hear eight of these in a row. For the laptop question above, "I understand your point, but looking at the bigger picture, I think..." becomes "I get why the price matters, but think about where you will actually use it." Same move, your words.

Task 5 is the task where reading the template is furthest from being able to do it. Three points in sixty seconds, with two prices said aloud naturally and an adjective chosen to fit a stranger in a scenario you have never seen, is a lot to assemble live. It comes apart in the gap between point two and point three.

Record against a real clock at least fifteen times before you book, more than you would for any other task. Our Task 5 question bank holds more than 80 comparing and persuading scenarios and scores what you record.

Two more things worth doing. Read the Task 4 template for Making Predictions and the Task 6 template for Dealing with a Difficult Situation, because Tasks 5 and 6 both give you a full 60 seconds of preparation and they are the two most people find hardest. Then check how the CELPIP Speaking test is organised so that the silent screen in Part 1 is not a surprise.

The Task 5 vocabulary and comparison phrases list goes deeper than the adjective bank above, and the full set of eight speaking templates is there when you are ready to work through the rest.