Templates
Describe Image frames for PTE Core
Twenty-five seconds to look, forty to speak. The scorer wants the image's content, which means its labels, numbers and shape, delivered fluently. A memorised paragraph about "this informative image" fills the forty seconds with nothing. This frame fills them with the chart.
Pearson scores content 0 for a response made of memorised material that does not describe the image, and content 0 zeroes the item. The engine checks whether the words you say relate to the image. "This image provides a great deal of information about an important topic" relates to no image. Say the labels and the numbers.
The four-part shape
Forty seconds is about 90 to 110 words at a natural pace. Divide them into four parts and give each one a frame. The frames are short so that almost every word after them is read off the image.
| Part | Seconds | What you say |
|---|---|---|
| 1. What it shows | 0–8 | Type of image, subject, units, time period or categories |
| 2. The extremes | 8–22 | Highest and lowest, with their numbers and labels |
| 3. The pattern | 22–32 | The trend, the comparison or the odd one out, with one more number |
| 4. One conclusion | 32–40 | A single sentence that follows from the numbers |
Frames
Part 1: what it shows
- "The [bar chart / line graph / pie chart / table / map] shows ___ in ___ [between ___ and ___]."
- "This is a ___ comparing ___ across ___."
Part 2: the extremes
- "The highest figure is ___, at ___, while the lowest is ___, at ___."
- "___ accounts for the largest share, about ___ percent, and ___ the smallest, at ___."
Part 3: the pattern
- Change over time: "Overall, ___ [rose / fell / stayed flat] from ___ to ___, with a [peak / dip] in ___."
- Comparison: "___ is roughly [twice / half / three times] ___, and the remaining categories sit between ___ and ___."
- Map or picture: "Most of the ___ is in the ___, while the ___ is concentrated in the ___."
Part 4: one conclusion
- "In short, ___ is clearly ___ than ___."
- "This suggests that ___."
A filled example
A bar chart titled "Weekly hours spent commuting, by city, 2025". Five bars: Toronto 6.5, Vancouver 5.8, Calgary 4.1, Montreal 5.2, Halifax 3.3. Vertical axis in hours per week.
The bar chart shows the number of hours people spent commuting each week in five Canadian cities in 2025. The highest figure is Toronto, at six and a half hours a week, while the lowest is Halifax, at just over three hours. Vancouver and Montreal sit close behind Toronto, at roughly five point eight and five point two hours, whereas Calgary is noticeably lower at around four. Overall, the two largest cities take almost twice as long as the smallest one. In short, commuting time rises clearly with the size of the city.
Every sentence after the first carries at least one number or label from the chart: five city names, five figures and a comparison. The frames are twenty words of the ninety-six. That ratio is what the content trait is looking for, and the complete sentences, each with a subject and a verb, are what the fluency trait is looking for.
Using the twenty-five seconds
- Seconds 1–5: read the title and the axes. Say the subject and the unit in your head. This fills Part 1.
- Seconds 5–15: find the highest and lowest. Note the label and the number for each on the notepad. This fills Part 2.
- Seconds 15–22: find the shape. Rising, falling, one big and the rest small, two groups. Pick one more number to mention. This fills Part 3.
- Seconds 22–25: decide the conclusion. One clause. Then wait for the tone and start speaking on it.
When the image is not a chart
PTE Core also uses photographs, diagrams, maps and process pictures. The same four parts work with different content. For a photograph: what the scene is, the two most prominent things in it, what is happening or how they relate, and what the picture seems to be for. For a map: what area, the largest and smallest regions or the key features, where things are concentrated, one conclusion. For a process: what it produces, the first and last steps, how many steps and where they branch, what the process is for. The vocabulary changes; the frame does not.
Saying numbers
Numbers are the cheapest content in the task, and the scorer gives credit for them, so say them clearly rather than rounding everything to "a lot". A few habits help the pronunciation trait: "six point five" rather than "six five"; "about four thousand" rather than attempting "three thousand nine hundred and eighty-seven" at speed; "two thousand and twenty-five" or "twenty twenty-five" for years, chosen in advance so you do not hesitate; and percentages with the word "percent" every time. If a number is too awkward to say, approximate it with "roughly" or "just under" and keep going. A pause to decode a figure costs more fluency than an approximation costs content.
The mistakes that cost the most
- Opening with a template paragraph. Ten seconds of generic praise for the image is a quarter of the recording spent on zero content.
- Stopping early. The scorer has nothing to rate during silence. If you run out of pattern, say another number.
- Restarting a sentence. A false start costs fluency; a small grammar slip costs almost nothing. Keep moving.
- Describing everything. Five bars is fine; a table with twenty cells is not. Highest, lowest, the shape, one conclusion, then stop.
- Trailing off before the tone. Finish the conclusion sentence with falling intonation at around thirty-six seconds, then stay quiet. A clear ending is read as fluency.
Practise on the Describe Image page with the timer running. The scorer there counts the labels and numbers you said against the image, which is the fastest way to see whether the frame has become a habit.