Task-Based Language Teaching has the strongest research base of any approach to second language speaking development. More so than Communicative Language Teaching, more so than grammar-translation, more so than content-based instruction. The studies are consistent: students who learn language through completing genuine communicative tasks develop both fluency and accuracy faster than students who study language in isolation.
And yet most teachers who describe themselves as using TBLT are not. They're using CLT with tasks sprinkled in. Understanding what genuinely constitutes a task - and what distinguishes TBLT from activities that merely resemble it - is the most practical thing a speaking teacher can take from the research.
YapYapGo is a classroom speaking practice tool for ESL and EFL teachers built around the principle that structured communicative tasks produce more acquisition than open-ended discussion. Here's what TBLT actually means for your speaking classroom.What counts as a task
In TBLT terms, a task has four essential features. Miss any one and it becomes an exercise rather than a task.
1. A primary focus on meaning. Students are focused on what they're communicating, not how they're communicating it. Grammar and vocabulary are tools, not goals. If the primary focus shifts to form - even briefly, even for correction - the activity has stopped being a task. 2. A gap to bridge. Something is unknown that needs to be discovered. An information gap (one student has information the other needs), an opinion gap (students genuinely hold different views), or a reasoning gap (students must reach a conclusion through logical argument). Without a gap, there is nothing to communicate. 3. Real-world relevance. The task should mirror something students might genuinely do in English outside the classroom. Planning a route, resolving a conflict, making a decision, giving advice. Not describing pictures or filling in blanks. 4. An outcome that can be evaluated. Did they solve the problem? Did they reach an agreement? Did they complete the map? A task has a measurable endpoint. A discussion question has none - which is why discussion questions are not tasks.By this definition, "discuss your opinion of social media" is not a task. "Decide which of these three social media policies your company should adopt, and be ready to present your decision with reasons" is a task.
The three-stage TBLT framework
Willis (1996) outlined the most practically useful TBLT framework for classroom teachers: pre-task, task cycle, and language focus.
Pre-task (5 minutes)
Introduce the topic and the task. Crucially, this is not a vocabulary lesson or a grammar review. It's contextualisation. What is the situation? Who are the people involved? What do they need to decide or discover?
You can include a brief example of the task - model the output, not the language. "Here's what a completed version might look like" rather than "here are the phrases you'll need."
Brief preparation time (two to three minutes) significantly improves task performance, especially for lower-level students. Students who can gather their thoughts before starting produce more complex language during the task.
Task cycle (15-20 minutes)
Students complete the task in pairs or small groups. This is the core of the lesson and should take the most time.
During the task: You circulate and observe. You do not interrupt. You note language that arises - both successful communication and gaps. You are collecting data for the language focus stage. After the task: Students report their outcome. Not a long presentation - just the result. "We decided..." / "We found that..." / "Our advice is..." This reporting phase produces a second round of language production.Tool tip: YapYapGo provides the structured pair interaction that is the engine of the task cycle. Pair students automatically, deliver levelled questions or prompts, and manage timing - so the task runs without logistical intervention from you. A visible activity timer marks the task phase clearly without requiring you to signal the end.
Language focus (10 minutes)
This is where TBLT differs most from traditional CLT. The language focus comes after the task, not before. Students have now encountered the language they needed (and the gaps in what they had). The language focus addresses those specific gaps.
Ask students: "What language did you find difficult to express?" Note what you observed while circulating. Then teach to those specific gaps.
This is reactive teaching: you address what actually came up during authentic communication rather than pre-teaching what you predict students might need. The research shows this produces significantly better retention - students who encounter a language gap during a meaningful task and then receive focused instruction on it retain the structure far better than students who are pre-taught the same structure.
Practical task types for speaking
Decision-making tasks: Students must reach a consensus. "You are a city council with £2 million to spend on one of five projects. Decide which one and explain why." Everyone has a stake in the outcome. Problem-solving tasks: A genuine problem with a solution to find. "You need to relocate your office to a new city. You have three options with different trade-offs. Which do you choose?" Information asymmetry can be built in: give each student different facts about each option. Opinion-exchange tasks: More structured than open discussion. "Interview your partner about three topics. Find out their opinion, the reasons behind it, and one experience that shaped it. Be ready to report to the class." Narrative tasks: Students reconstruct or create a narrative. "You both have three pictures from the same event but in different orders. By describing what you can see, decide together on the correct sequence without showing each other your pictures." Comparison tasks: Students compare options and reach a judgment. "Here are five definitions of success. Rank them from most to least important. Compare your rankings with your partner and agree on a final list of three."How TBLT connects to YapYapGo's modes
YapYapGo's speaking modes are all task-based rather than exercise-based:
- Free Conversation: Opinion-gap tasks (genuine exchange of personal views)
- Timed Talk: Performance tasks (sustaining a one-minute talk requires planning and execution)
- Topic Discussion: Opinion-exchange tasks (structured conversation with a specific topic)
- Debate: Decision-making and argumentation tasks (reach and defend a position)
- IELTS Mode: Exam simulation tasks (real-world task with an evaluable outcome)
- AI Discussion: Problem-posing tasks (respond to AI-generated questions on unpredictable topics)
For the research context behind why these modes work, see our posts on communicative language teaching in 2026 and fluency vs accuracy: when to correct.
A random student picker and classroom group maker are useful in the reporting stage of the task cycle - selecting pairs to share their outcome creates genuine accountability without the social pressure of whole-class volunteering.
Sources:
- Willis, J. (1996). A Framework for Task-Based Learning. Longman. - The pre-task/task cycle/language focus framework.
- Ellis, R. (2003). Task-Based Language Learning and Teaching. Oxford University Press. - The research base for TBLT over other approaches.
- Skehan, P. (1998). A Cognitive Approach to Language Learning. Oxford University Press. - Fluency, complexity, and accuracy as competing demands in task completion.
- Long, M. (2015). Second Language Acquisition and Task-Based Language Teaching. Wiley-Blackwell. - The theoretical and empirical case for TBLT.