An app says you’re “B1” after three weeks. Sounds nice, but is it true? CEFR labels can be as reliable as a nutrition label, or as vague as “healthy-ish,” depending on the product.
If you need to verify CEFR alignment fast, you don’t need a full research project. You need two things: a quick diagnostic (what the app claims and shows), and a real task (what you can actually do).
This 15-minute audit works for independent learners, tutors, and L&D buyers who want a clear go or no-go signal.
What CEFR alignment really means (and what it doesn’t)
CEFR is a framework for describing language ability across levels A1 to C2. It’s built around can-do descriptors across skills, not around how many units you finished. The official starting point is the Council of Europe’s CEFR hub: Common European Framework of Reference for Languages.
To verify CEFR alignment, you’re checking whether an app’s level labels match the kind of performance CEFR describes across:
- Reception (listening, reading)
- Production (spoken, written)
- Interaction (spoken, online, collaborative)
- Mediation (relaying meaning, summarizing, explaining for someone else)
The CEFR Companion Volume expands and updates descriptors (including mediation and online interaction). If an app claims “CEFR-aligned” but ignores these areas, that’s a clue. A key reference is the CEFR Companion Volume with new descriptors.
One more boundary: CEFR doesn’t prescribe a method. Flashcards can be part of an aligned program, but only if they build toward real-world ability that matches descriptors.
The 15-minute CEFR alignment audit (one screen, one timer)
Use this simple schedule. It’s designed for a quick decision, not a perfect verdict.
| Time | What you do | What you’re looking for |
|---|---|---|
| Minutes 0 to 7 | Quick diagnostic inside the app and its level claims | Clear mapping to CEFR descriptors, balanced skills, consistent difficulty |
| Minutes 8 to 15 | One real task at the app’s claimed level | Performance that fits CEFR in range, accuracy, fluency, coherence, and interaction |
Do it twice if you can, once at the level you’re in and once one level above. Weak alignment often shows up at level boundaries (A2 to B1, B1 to B2).
Step 1 (7 minutes): Quick diagnostic that catches most “CEFR-washing”
Open the app’s level page, placement test info, or curriculum outline. You’re not hunting for pretty badges. You’re hunting for traceable claims.
1) Find the app’s CEFR statement and check its wording
Good signs:
- It names levels and ties them to can-do outcomes (not just “grammar topics”).
- It states which skills are covered (reception, production, interaction, mediation).
- It explains what “B1” means in the app (not only what you studied).
Weak signs:
- “CEFR-based” with no explanation.
- Level labels without descriptors, samples, or benchmarks.
2) Look for a descriptor map (or anything close)
In strong products, you can spot a mapping, even if it’s not called that. Examples include “At B1 you can…” lists, checklists by skill, or progress reports tied to functions.
If the app never links tasks to CEFR descriptors, you can’t really verify CEFR alignment, you can only trust marketing.
3) Scan the content mix in 60 seconds
At the claimed level, count what you see:
- Mostly multiple-choice and gap fills?
- Or do you see speaking prompts, writing prompts, messages, role-plays, or summarizing tasks?
A CEFR level is hard to defend without productive skills (speaking, writing) and interaction. If the app avoids them, it can still be useful, but its CEFR label is shaky.
4) Spot “difficulty drift”
Open three lessons at the same level and sample one item each. Red flags:
- Random jumps from basic forms to dense texts with no support.
- “B2” content that feels like A2 sentence building.
- Recycled beginner prompts with harder vocabulary sprinkled in.
Alignment needs consistency, not surprises.
Step 2 (8 minutes): One real task that a CEFR level should support
Now you test the claim with a single task. Think of it like a test drive. A label isn’t meaningful until you steer it around a corner.
Choose a task type (match the app’s stated level)
Pick one based on what the app says you are. Keep it short, timed, and a bit uncomfortable.
- A2 real task (interaction): Send a short message to arrange a plan (time, place, price). Add one constraint (“I might be late”).
- B1 real task (production): Speak for 60 to 90 seconds summarizing a simple article or a short video you just watched.
- B2 real task (mediation): Read a short text, then explain the key points to a friend in your own words, with one recommendation.
- C1 real task (interaction + coherence): Give a 2-minute opinion with one counter-argument, then respond to one follow-up question.
If the app includes speaking or writing tools, do the task inside the app. If it doesn’t, do it outside and judge whether the app prepared you for it.
Score your performance with a fast CEFR-style lens
Don’t overthink it. Use four criteria that show up across CEFR thinking:
- Range: Could you vary words and structures, or did you repeat the same patterns?
- Accuracy: Did errors block meaning, or were they minor?
- Fluency: Did you keep going with natural pauses, or stall every sentence?
- Coherence: Did it hang together with basic linking, examples, and clear order?
Add one skill-based check:
- Interaction: Could you respond and adapt, not just recite?
A simple rule: if meaning breaks often, the level is probably overstated. If meaning holds and you can manage the task with some strain, the level claim may be reasonable.
Interpreting the result: Pass, unclear, or fail
After 15 minutes, you want a practical decision.
Pass (likely aligned enough): The app’s claimed level matches the kind of tasks you can do, and its curriculum shows descriptor-like outcomes across more than one skill.
Unclear (needs evidence): The app seems helpful, but its CEFR label can’t be traced to descriptors or benchmarks. You need proof from the provider.
Fail (label is probably inflated): The app’s “B1/B2” is mostly drills, there’s little interaction or production, difficulty is inconsistent, and your real-task performance collapses.
Quick red flags that should change your buying decision
Watch for these patterns:
- Only grammar drills and vocabulary taps, with no speaking or writing.
- “CEFR” badges with no descriptor mapping or sample outcomes.
- Inconsistent difficulty inside one level.
- Placement tests that only check recognition (multiple-choice) and never production.
- No mention of validation, benchmarking, or how levels were set.
What credible evidence looks like (and what to ask for)
If you’re buying for a team, ask for documents, not slogans. Strong providers can usually share at least a summary.
Look for these evidence types:
Alignment study: A documented process showing how tasks map to CEFR descriptors and how decisions were checked.
Standard setting: A formal method for setting cut scores or level boundaries, often using trained judges.
External benchmarks: Comparisons to known frameworks or tests, or learner outcomes mapped against external measures.
A practical reference for what “alignment work” involves is the EALTA CEFR alignment handbook (PDF). If an app claims alignment but can’t describe anything like the steps in that handbook, treat the CEFR label as unproven.
For additional benchmarking context, you can also consult ALTE’s resources, including can-do oriented materials: ALTE resources and reference materials.
Conclusion: Trust tasks more than labels
You can verify CEFR alignment quickly by checking two things: whether the app ties learning to CEFR-style can-do outcomes, and whether you can complete a real task that fits the claimed level.
If the app passes, keep using it, but re-check at the next level boundary. If it fails, the app may still help with basics, but its level badge isn’t a safe guide for goals, hiring, or training plans. The most reliable signal is still performance: what you can understand, produce, and do with other people in real time.
