Meet LARA
LARA is an evidence-informed AI tool that enables teachers to provide fast, high-quality, individual formative feedback to every student in a class, enabling them to edit and improve their written responses during a lesson.
What is LARA for?
What is LARA for?
What is LARA for?
Research consistently shows that feedback is most effective when it is specific, tied to clear criteria, and usable during learning. LARA is built to support that moment, inside a classroom cycle, so students can revise straight away rather than waiting for later comments.
Who is LARA for?
What is LARA for?
What is LARA for?
LARA is designed for classroom-based learning with students typically aged 12 and over, using short-cycle tasks where timely, improvement-focused feedback supports better drafts,
reasoning, or explanations.
How LARA Works
Step 1 Teacher sets the task
Step 3 LARA generates draft feedback aligned to criteria
Step 1 Teacher sets the task
Teachers create a written task and add success criteria if they want. If criteria are not provided, LARA can draw on Universal Learning Expectations as a research-aligned baseline.
Step 2 Students draft response
Step 3 LARA generates draft feedback aligned to criteria
Step 1 Teacher sets the task
Students join with a temporary class code and submit a response.
Step 3 LARA generates draft feedback aligned to criteria
Step 3 LARA generates draft feedback aligned to criteria
Step 4 Teacher reviews feedback before anything is shared
Draft feedback is structured to connect learning goals, current progress, and practical next steps.
Step 4 Teacher reviews feedback before anything is shared
Step 4 Teacher reviews feedback before anything is shared
Step 4 Teacher reviews feedback before anything is shared
Nothing is released until the teacher checks it. Teachers can approve, edit, or reject each draft.
Step 5 Students choose a focus and revise their response
Step 4 Teacher reviews feedback before anything is shared
Step 5 Students choose a focus and revise their response
Students use the feedback to select an improvement focus and revise their draft. LARA guides next steps without rewriting student work.
Step 6 Teacher reviews class insights
Step 4 Teacher reviews feedback before anything is shared
Step 5 Students choose a focus and revise their response
Teachers can review class-level insights, including common strengths, areas for development, and suggested next steps, to inform their teaching
Fast, High-Quality Feedback
LARA feedback is designed to be usable straight away. It follows a clear three-part structure aligned to strong formative practice: the learning goal, current progress, and next steps a student can apply during revision.
When you provide success criteria, LARA aligns feedback to what you have asked students to do. When you do not supply criteria, LARA can fall back to Universal Learning Expectations, which provide a stable, student-facing baseline focused on clarity, evidence, reasoning, organisation, and appropriate language.
LARA Feedback Principles
Research-Aligned Feedback Support with Teacher Judgement Built in
LARA is an AI-powered feedback engine designed for classroom use. It generates draft feedback aligned to your task criteria and your success criteria, then routes it through a teacher check before anything is shared with students. Teachers can approve, edit, or reject each draft, so professional judgement remains the key final step.
A Complete Feedback Routine
Each response connects to learning goals, describes current progress against criteria, and suggests practical next steps students can apply in their next attempt.
The three questions students need answered
- Where am I going? Clear links to learning goals and success criteria
- How am I doing? Descriptive feedback on what is working and what needs attention
- What’s next? Two or three high-impact actions for revision
Designed for student progress
- Feedback focuses on the work and the strategy, using language that supports effort and improvement.
- Reflection prompts help students interpret feedback, self-assess, and plan their revision.
Key Features
Precision: Feedback is tied to your criteria with concrete examples and revision steps.
Timeliness: Short drafts can be produced during the learning window, with limited focus points to avoid overload.
Teacher control: Nothing is released until the teacher approves it.
Privacy-first: Students use session-only codes. No names. No accounts. Codes are deleted after 16 hours.
The Research Behind LARA
LARA Feedback Principles: Evidence-Base Summary
1. Formative by design, not just comments
Feedback embedded in a continuous learning loop (clarifying goals, eliciting evidence, and guiding next steps) is consistently shown to improve learning more than isolated comments (Black & Wiliam; Hattie; Shute).
2. Always answer the three questions
Hattie & Timperley's model underpins this principle: effective feedback addresses goals, current progress, and specific next steps. Adopted across QCAA, MIT TLL, and NSW Education.
3. Focus on task, process, and self-regulation
Research (Hattie & Timperley; Wisniewski et al.) shows feedback is most effective when focused on the work, strategies, and self-regulation - not personal judgments.
4. Be specific, descriptive, and improvement focused
Shute and others emphasise that concrete, descriptive feedback paired with strategies for improvement significantly outperforms vague praise.
5. Timely, concise, and usable now
Feedback given during or soon after learning is far more impactful. NSW and QCAA recommend short, focused comments targeting a small number of high leverage improvements.
6. Build feedback literacy and student agency
Carless & Boud highlight that feedback only works when students can interpret and act on it. Prompts for reflection and choice build agency and uptake.
7. Emotionally safe and motivational
Students act on feedback when it is respectful, balanced, and growth focused. Research warns that harsh or overwhelming comments reduce engagement (Brandmo & Gamlem; Shute).
8. Tightly aligned to intentions and criteria
High impact feedback clearly links to learning intentions and success criteria, making visible what is met and what needs work (Wiliam; AITSL; QCAA).
9. Support dialogue and multiple feedback roles
Feedback is most powerful when part of dialogue, including teacher, peer, and self-feedback. Wiliam's model and QCAA frameworks reinforce this.
10. High impact, low workload
Meta analyses show targeted, concise feedback improves learning more than volume of comments. NSW advocates focusing on high leverage elements to keep workload sustainable.
Perfect for Modern Classrooms
Empowering Student Agency
Empowering Student Agency
Empowering Student Agency
Students choose their own revision focus from the feedback provided, building independence and ownership in the learning process.
Smart Workload Management
Empowering Student Agency
Empowering Student Agency
High-impact feedback that improves learning without overwhelming anyone. LARA focuses on what research shows matters most.
Seamless Integration
Empowering Student Agency
Seamless Integration
Teachers input their questions and can add their own success criteria. LARA also applies universal learning criteria built by teachers for teachers, reflecting current research. Enhance what you're already doing without starting over.
Ready to Transform your Student Feedback?
LARA delivers what every teacher wants:
- Evidence-informed feedback design grounded in formative assessment research
- Feedback students can use to improve their written responses during the lesson
- Class-scale support for individual feedback, with learning quality kept central
- Safety and privacy by design, with minimal student data collection
Register now for a special launch price discount when LARA goes live In April