Each submission is scored on the same four pillars the real examiners use: grammatical correctness (varied tenses, accurate verb endings, pronoun usage), vocabulary range (precision, idiomatic expressions, lexical variety), coherence and cohesion (logical connectors, paragraph flow), and task completion (did you answer the prompt and hit the word count). You get a numerical score per pillar, sentence-level corrections, and an NCLC band estimate.

Most students need 8–15 timed writing sessions before they're comfortable hitting the word count and structure on demand. The Success Suite (10 writing tests) covers a 2–3 week sprint; the Ultimate Proficiency Pass (20 writing tests) covers a full month of intensive prep or a retake campaign.

TEF Canada has 2 tasks (short message at ~80 words, then a longer 200-word opinion piece). TCF Canada has 3 tasks (short message, formal letter, then a 200-word argumentative essay). Both grade you on the same four criteria; the TCF format gives more chances to demonstrate range. You can switch between them in the dashboard.

CLB 7 writers attempt complex structures (subjunctive, conditional, concession) and use varied connectors (cependant, par conséquent, étant donné que) even with small errors. CLB 5 writers stay safe with simple sentences and avoid risk. NCLC examiners reward attempted complexity over flat correctness — a varied response with a small error scores higher than ten perfectly simple sentences.

Every submission is saved to your dashboard with the original prompt, your answer, the AI breakdown, and the corrected version. You can review past submissions to track which mistakes you're repeating. Bulk export is on the roadmap.

Occasionally — especially on edge cases like uncommon idioms or culturally specific phrasing. The score is a strong proxy for what a TEF / TCF examiner would give, but it's a model, not a human. We weight grammatical accuracy and task completion most heavily because those are the easiest for AI to verify reliably.