A CEFR test is an English assessment whose results are reported on the Common European Framework of Reference for Languages (CEFR) scale, from A1 to C2. For companies, it gives L&D one scale for comparing employees' English across countries.

A CEFR level works like a shoe size. It narrows the options, but it can't tell you whether the shoe suits the job.

Illustrative scenario: A company with sales teams in four countries gives every employee a CEFR-aligned test. Most results land in the middle of the scale, so the report looks reassuring. It doesn't show who can handle a live customer call.

This guide covers CEFR tests that companies use for employees' English, and what L&D can and can't read from the result.

What is a CEFR test, and who runs it?

The Council of Europe publishes the CEFR framework, while independent providers write the tests and report results against it.

The Council of Europe's CEFR levels page describes six levels, A1 to C2. Each level is defined by "can-do" descriptors.

The Council of Europe does not run tests. Its page on relating examinations to the CEFR says it does not verify or validate the quality of the links between examinations and CEFR levels. Providers are responsible for using the CEFR coherently, so a "CEFR-aligned" label is the provider's own statement.

What do the six CEFR levels cover?

The six levels run from A1 to C2 and group into Basic, Independent and Proficient users.

User groupLevels
Basic userA1, A2
Independent userB1, B2
Proficient userC1, C2

The descriptors were created without reference to any specific language. A level therefore reports what a person can do, not which exam they passed.

How does a CEFR test for employees work?

Employees take an assessment, the provider converts the scores into CEFR levels, and L&D compares those levels with what each role needs.

  1. Decide which skills matter: speaking, listening, reading, writing, or all four.
  2. Employees sit the assessment. Formats differ by provider.
  3. The provider reports results as A1 to C2. Ask whether that is one overall level or one level per skill.
  4. L&D compares each result with the level its role calls for.
  5. L&D plans practice, training or a retest for the gaps.

How do the CEFR, a CEFR test and a speaking measure differ?

The CEFR is a scale, a CEFR test is one provider's assessment reported on that scale, and a speaking measure scores spoken performance.

TermWhat it isWho publishes itWhat it reports
CEFRA framework of six proficiency levels and descriptorsCouncil of EuropeLevel definitions, not scores
CEFR-aligned testAn assessment linked to the framework by its providerThe test providerA level from A1 to C2, overall or by skill
Speaking-performance measureA score based on how a person speaksThe measure's ownerA score on the owner's own scale

ELSA Business uses the ELSA Proficiency Score (EPS), which measures spoken-English proficiency based on speaking performance. EPS is ELSA's own score, so it is not a CEFR certificate.

Where does a CEFR test fall short?

One CEFR level can hide uneven skills, and a level on its own is not a job requirement.

One level can hide a weak skill

In the EF English Proficiency Index 2025, published on 19 November 2025, speaking was the weakest English skill in over half of the countries measured. The index draws on about 2.2 million adults who took EF's own online test, and EF notes that these test takers are self-selected, so the results may not represent each country's population.

A single level is a blunt job requirement

A Norwegian research project, DISCEFRN, funded by the EU and run at Western Norway University of Applied Sciences, reviewed CEFR-based language requirements in tens of thousands of Norwegian job ads. Experts judged about half of those requirements adequate, and the rest were more often set too high than too low.

For a considerable share of ads, the experts recommended different levels for speaking, writing, listening and reading instead of one flat level. The study looked at Norwegian job ads, not English requirements at multinational companies, but its lesson travels: a single level is a blunt target.

A result describes one moment

A CEFR result does not show whether an employee is practising between assessments.

What should L&D ask a CEFR test provider?

Ask which skills are reported separately, how the link to the CEFR was made, and how speaking is assessed.

  1. Does the report give one overall level or one level per skill?
  2. How did the provider link its test to the CEFR, and is that evidence available to read?
  3. How is speaking assessed: recorded speech or typed answers, and scored by people or software?
  4. Can the same employees retake the assessment, and how often?
  5. Can the results be grouped by team, role and country?

How should L&D use CEFR results?

Use the result as a baseline for each skill, then set targets by role rather than one level for everyone.

The practical first step is to name the two or three roles where spoken English carries the most weight, such as customer calls or international meetings. ELSA's page for HR leaders covers the program view.

Aviation sets targets the same way: ELSA's post on ICAO English proficiency and Level 4 explains a standard built for specific roles, not for everyone. Once the baseline is set, ELSA's guide to comparing corporate English training options covers how to weigh the formats that can close the gaps.

Frequently asked questions

Is there an official CEFR certificate for companies?

The Council of Europe publishes the framework and does not verify how examinations link to it. A "CEFR-aligned" claim therefore comes from the test provider. Ask for the evidence behind the link before you rely on a result for hiring or promotion.

What does a level such as B2 mean?

B2 sits in the Independent user group with B1. Each level is described with "can-do" statements, and the descriptors were created without reference to any specific language. A level reports what a person can do, so ask the provider which skills the result covers.

Can one level describe how well an employee speaks?

Not reliably when the report gives a single overall level. Skills can differ, and speaking was the weakest skill in over half of the countries in the EF index. Ask the provider for a separate speaking result for each employee.

How is a speaking-performance measure different from a CEFR test?

A CEFR-aligned test reports a level on a six-level scale. A speaking-performance measure such as EPS scores how a person speaks, on the measure owner's own scale. They answer different questions, so a program can use both. Ask a vendor which of the two it is describing.

Which CEFR level should employees reach?

It depends on the role and the skill. In the Norwegian study, experts judged about half of the set requirements adequate, and the rest were more often too high than too low. Set targets per skill for each role, starting from the skills each role depends on most.

Next step

  • A CEFR test gives L&D a common scale, not a verdict.
  • "CEFR-aligned" is the provider's own statement, so ask how the link was made.
  • Ask for results by skill and set targets by role.
  • Add speaking practice and speaking measurement where spoken English drives the work.

To see how ELSA Business could fit alongside your current training, book a demo and talk through your teams' speaking needs.