The Task Isn't Working: Change the Decision, Not Just the Timing
Learn to choose ELT tasks by their learning purpose. Examples of task difficulty, interaction patterns, monitoring and changing tasks reactively.
Matching, discussions and gap-fills aren't lesson aims. They're tools. What matters is what you ask learners to do with them.
You've prepared a lovely matching activity. The pictures are clear, the slides look professional, and learners have finished all ten items.
The lesson is running smoothly.
But what did the activity achieve?
Ask that question before you tell me how engaging it was.
I've seen teachers spend a lot of energy deciding whether an activity should be a quiz, a card game or a pairwork worksheet. Those choices can matter. But they come after a more important decision: what do learners need to think about, practise or show?
An attractive activity with an unclear purpose is still an unclear activity. And when a task fails, the problem isn't always that learners need two more minutes.
Before you read on: a task isn't working. What is your first reaction usually?
Most of us either give more time or explain again in the same way. Both change the timing, not the decision. This article asks you to diagnose first: is the problem the procedure, the language, the content, or too many demands at once? Then change one thing.
Activity, exercise, task: don't get trapped by labels
In everyday ELT talk, a task is anything learners do: match items, discuss a question, do a gap-fill, write an email.
In Task-Based Language Teaching (TBLT), task has a more technical meaning. It usually includes a primary focus on meaning and a communicative outcome. You'll find this distinction in the work of Rod Ellis (2003) and others.
So a grammar gap-fill is a task in the classroom sense, but may not be a task under a TBLT definition. That's fine. (More key terms in ELT Terminology, Part 1.)
Is matching a good activity? What is this matching activity meant to do for these learners?
One format, three purposes
Three teachers all use a matching exercise. Tap each one to see what changes.
Teacher A · Before listening
Task: match workplace problems to pictures.
Aim: activate knowledge before a conversation about workplace problems.
Monitor for: are they engaged and ready?
Feedback: light. A long answer check is probably unnecessary.
Teacher B · Listening for detail
Task: hear four conversations, match each speaker to a problem.
Aim: extract specific information accurately.
Monitor for: wrong answers and why they chose them.
Feedback: confirm answers clearly and deal with misunderstandings.
Teacher C · Before speaking
Task: match responses to complaints, discuss which are more appropriate, then negotiate solutions.
Aim: prepare for freer speaking.
Monitor for: the language they use in the speaking that follows.
Feedback: the real evidence comes later, not from the matching.
The first question: what is the learner doing in their head?
A task can be busy without being demanding, or demanding in an unhelpful way.
| Instruction | Mental demand | Fits when… |
|---|---|---|
| "Underline all the adjectives." | Identify | Learners are starting to recognise adjectives |
| "Which adjectives make this review sound unusually negative? Explain two." | Interpret and justify | The aim is critical reading and evaluative language |
Same material. The second asks for more. That doesn't make it better. If learners are still learning basic adjectives, the first could be exactly right.
And don't confuse a long instruction with a demanding task. A complicated instruction can make a simple task harder to understand without adding any learning.
Try it. Aim: learners will be better able to infer a writer's attitude from word choice in short reviews. Which task fits best? A. Underline every adjective. B. Find two expressions showing approval or criticism and explain what each implies. C. Copy the review into your notebook.
B is the strongest match: it connects language choices with the writer's attitude. A could be a preparation step if learners need help finding adjectives. C may serve another aim (transcription, close reading) but doesn't test interpretation.
When a task is too easy
Learners finish much earlier than you expected. Before you add extra questions, ask why.
- Was the task genuinely easy?
- Did they already know the language?
- Were the answers obvious from the layout?
- Did partners just copy each other?
- Was your time estimate too generous?
If they achieved the aim, move on. If the task gave you no useful evidence, redesign the next one. If the aim was fluency, retrieval or confidence, an easy task may still have been valuable.
Don't make learners repeat work just because there are six minutes left in your stage plan.
When a task is too difficult
Now the opposite. You've explained, and learners still haven't started. Find the obstacle first:
- Do they understand what to do?No → demonstrate one example; don't just repeat the instruction.Yes → next question.
- Do they have the key language?No → give the missing phrase, or add a short language-support step.Yes → next question.
- Do they know enough about the topic?No → add input: a text, a picture, a quick brainstorm.Yes → next question.
- Are there too many demands at once?Yes → reduce one demand, but keep the aim.No → the task may be wrong for this class: replace it.
This links to the Zone of Proximal Development, but it isn't a precise measurement of it. You can't place everyone in a three-colour zone with one look. Watch what learners do alone and how they respond to support.
Does every task need "alone → pairs → class feedback"?
No.
That routine works well for many accuracy-focused reading, listening or language-analysis tasks. Learners think alone, compare, then resolve problems together. It reduces copying and creates peer explanation.
But it isn't a universal task cycle:
- A role-play needs pair interaction from the start.
- Collaborative writing may be designed so that ideas grow through co-construction.
- A timed fluency activity can be killed by an unnecessary "work alone first" stage.
- A quick prediction task may not need formal whole-class feedback.
Also, task cycle has a specific meaning in task-based frameworks. In Jane Willis's (1996) framework, the task cycle is:
- TaskLearners do the communicative task in pairs or groups.
- PlanningThey prepare to tell the class what they did or found.
- ReportThey present; the teacher responds to content and language.
So don't present "set → individual → peer check → open-class feedback" as something every TBLT lesson must follow. The better question: what interaction pattern serves this aim and this task?
Process or product: what are you looking for?
In a brainstorm, the value may be in participation, language use and ideas. There may be no single right answer. In a reading task for specific information, the accuracy of the answer matters.
Some teacher-training guides call these process and product tasks. It's a helpful lens, but real tasks often mix both: how learners reach an answer can matter as much as the answer.
This changes your monitoring and your feedback. Don't spend five minutes correcting every idea from a lead-in. And don't move on from a key language-analysis task while misunderstandings remain.
Looks successful, teaches very little
Illustrative scenario: B1 listening. The aim is listening for specific information about travel changes. The teacher shows the transcript before the recording, highlights all five answers, then plays the audio. Everyone gets 5/5.
Think first: what did learners actually practise?
Maybe following a transcript. Maybe recognising pronunciation. But if the aim was independent listening for specific information, the procedure removed the challenge the aim needed.
A better design: learners listen with a focused purpose, then compare. If it's still too hard, give selective support and replay the key section. If you use the transcript, be clear what it helps them notice, then give a later chance to listen with less support.
5/5 means the task worked. 5/5 means something worked. Check it was the skill in the aim.
Too much at once
Illustrative scenario: A2 speaking. Learners must negotiate a holiday plan on a budget, persuade partners, compare hotel reviews, calculate costs and use the second conditional.
The slides look fantastic. The cognitive load does not.
What's the main aim? Negotiation? Functional language? Comparing options? The conditional? All are legitimate, but together they may be too much for the same learners at the same time.
Keep the communicative goal and reduce the load: fewer options, a ready-made price table, the key language on the board. Or separate the language clarification from the problem-solving task.
Change one variable, not everything
When a task stalls, teachers often drop it or explain it again in exactly the same way. Try a diagnostic response instead.
| What you see | Possible change | What you protect |
|---|---|---|
| Learners don't understand roles | Demonstrate one exchange | Communicative purpose |
| Learners lack one key phrase | Give that phrase | The decision-making task |
| Every answer is obvious | Add a "why?" or a transfer challenge | Main learning goal |
| Too many options | Reduce or group the information | Relevant challenge |
| One learner dominates | Give each partner different information or roles | Both learners contribute |
These are options, not guaranteed fixes. Check the result after you change something.
A short planning test
Before any major task, finish these four sentences:
- PurposeLearners are doing this because…
- EvidenceI'll know it served the aim when…
- DiagnosisIf they struggle, I'll first check whether…
- FeedbackIn feedback, we'll need to establish…
Our lesson-planning guide shows how to keep aims, stages, tasks and feedback aligned.
Your turn: learners did a matching exercise on polite requests perfectly. In the freer task, everyone still says "Give me…" where a tactful request is needed. How would you redesign?
The matching checked recognition, not the ability to choose an expression in context.
Add a short comparison of directness and relationship (Give me the report. Could you send me the report when you get a chance?), then supported spoken rehearsal, then a new situation with no answer options visible. Watch whether learners transfer the choice.
CELTA, DELTA and principled choice
CELTA trainees often need help connecting aims, stages, tasks and feedback. A lesson can be beautifully staged and still fail to develop the skill if every activity removes the relevant challenge.
DELTA candidates must justify more deeply: why this procedure, what learning assumptions support it, and how the evidence informs evaluation. The vocabulary is more advanced, but the core question is the same: did the task contribute to the learning goal?
This is where principled eclecticism becomes more than a slogan. Matching, discussion, guided discovery or explicit teaching: what matters is a defensible reason rooted in the learners and the aim.
Try it tomorrow
Choose one activity you already plan to use. Keep the material, but write two different instructions that serve two different aims. Decide which fits your learners and what evidence you'll need from their performance.
You may find the activity wasn't the problem. The decision behind it was.
Plan for everything. Teach what they need.
Quick FAQs
Is a gap-fill a task?
In everyday classroom language, yes. Under some TBLT definitions, a form-focused gap-fill may not count as a pedagogic task.
Must learners always work alone before pairs?
No. It's useful for some tasks, especially accuracy-focused ones, but it isn't a universal sequence.
Is a harder task always better?
No. The challenge should serve the aim and be reachable with appropriate support.
Can a task be engaging but ineffective?
Yes, if it doesn't produce the practice or evidence the learning goal needs.
Keep learning with Lumina
- Lesson planning: keep aims, stages and tasks pulling in the same direction.
- Scaffolding or spoon-feeding?: support that keeps the challenge alive.
- Monitoring in ELT: collect the evidence your task was designed to produce.
- Error correction: now or later?: how the task's aim decides your feedback.
- Principled eclecticism: choosing methods for reasons, not habits.
References
- Ellis, R. (2003). Task-based Language Learning and Teaching. Oxford University Press.
- Willis, J. (1996). A Framework for Task-Based Learning. Longman.
- Skehan, P. (1998). A Cognitive Approach to Language Learning. Oxford University Press.
- Black, P., & Wiliam, D. (2009). Developing the theory of formative assessment. Educational Assessment, Evaluation and Accountability, 21, 5–31. https://doi.org/10.1007/s11092-008-9068-5
- Cambridge English. CELTA syllabus and assessment guidelines. cambridgeenglish.org
Check your understanding
6 short questions. No account needed, no score, and nothing you choose or write is saved or sent anywhere.
Show the answer (no automatic checking)
Answer: C. The learner must connect textual evidence with interpretation.
By Lumina Trainers
Show the answer (no automatic checking)
Answer: C. Task success is not necessarily transferable competence.
By Lumina Trainers
Show the answer (no automatic checking)
Answer: B. Interaction patterns depend on the purpose and design.
By Lumina Trainers
Show the answer (no automatic checking)
Answer: B. Frameworks employ different definitions and criteria.
By Lumina Trainers
Show the answer (no automatic checking)
Answer: 1–B, 2–A, 3–C. Compare the learning need and the purpose of the response; there is no benefit in using one routine for every situation.
By Lumina Trainers
6 Redesign an A2 task that requires comparing five hotels, negotiating prices, debating sustainability and using a new conditional form simultaneously.
Show a suggested answer
Identify one priority aim, reduce simultaneous demands, provide only essential background/functional language, use a manageable decision with fewer options, and retain a meaningful communication goal. Check whether the adapted task still demands the intended skill.
Suggested answer by Lumina Trainers
Ahmed Bin NasserCELTA Tutor & Teacher Trainer | Lumina


