How Many Skills Should You Test in One Hiring Round? Reading CEFR Results by Skill

The problem with one blended score

Most hiring teams still ask a deceptively simple question: what is the candidate's English level? They want one letter and one number they can compare across a shortlist. The trouble is that language ability is not one thing. A person can read technical documents comfortably yet freeze on a live call, or speak fluently while producing written English that no one would put in front of a client.

When you collapse all of that into a single blended CEFR score, you lose the very information that predicts on-the-job performance. This guide is about reading CEFR results by skill: how a per-skill breakdown works, how to set pass marks that reflect what the role actually needs, and how to decide how many skills are worth testing in one hiring round.

The core argument is straightforward. A blended B2 can hide a B1 in the one skill that matters most for the job. If you screen on the average, you can hire someone who looks qualified on paper and struggles exactly where the role is most demanding.

What a per-skill CEFR breakdown actually tells you

A per-skill result reports a separate CEFR level for each of the four skills: listening, reading, speaking, and writing. Instead of one label, you get four:

• Listening — C1
• Reading — C1
• Speaking — B1
• Writing — B2

That profile would blend to something around B2 overall. But the story is not "B2." The story is that this candidate understands and reads English at a high level and writes competently, yet speaks at a noticeably lower level. For a report-heavy analyst role, that might be perfectly acceptable. For a phone-based support role, it is a serious gap that a blended score would have concealed.

The reason skills can diverge so widely is that each one is defined by its own set of descriptors. The CEFR does not treat "B2" as a single monolithic ability; it spells out separate can-do statements for what a speaker, listener, reader, and writer can each do at every level. You can see these per-skill can-do descriptors laid out by the Council of Europe. That is precisely why a candidate can sit at different levels across the four skills, and why a per-skill breakdown is more honest than an average.

The International English Test assesses all four skills and maps each to a CEFR level from A1 to C2, giving you that per-skill breakdown rather than a single blended figure. Scoring is automated and AI-assisted, so results come back in minutes with a level attached to every skill. If you want the mechanics of how each skill is levelled, we set that out in how we score.

Set pass marks per skill, not one overall bar

Once you can read results by skill, the natural next step is to stop screening against one overall threshold. A single bar forces every role through the same filter, which means either rejecting good candidates for weaknesses that do not matter, or accepting weak ones because a strong skill dragged the average up.

Instead, set a per-skill pass mark weighted to the role:

1. List the skills the job actually uses. Be honest about frequency and stakes. A warehouse team lead may listen and speak all day but rarely write more than a short message.
2. Assign each relevant skill a minimum CEFR level. This is the floor a candidate must clear for that specific skill.
3. Treat each floor as a separate gate. A candidate must meet every required threshold. A brilliant reading score does not buy back a speaking level that sits below the line for a customer-facing role.
4. Decide which skills are optional or informational. You might record writing for a phone role without gating on it, so you have the data if the job later evolves.

This approach turns a vague "we need good English" into a defensible, role-specific standard. It also makes rejections easier to justify and easier to defend, because you are measuring against a requirement you defined in advance rather than a gut feeling. For a fuller framework on choosing those thresholds, see CEFR levels for hiring.

Three roles, three very different profiles

The same shortlist of candidates can be right or wrong depending on the role. Here is how per-skill pass marks might look across three common jobs.

The call-centre handler lives on the phone. Listening and speaking are the make-or-break skills, so they carry the high bars; the candidate needs to understand accented, fast speech and respond clearly under time pressure. Writing rarely goes beyond short notes, so a B1 floor is plenty. A blended-B2 candidate who is actually B1 in speaking would be a poor hire here, and only a per-skill view catches it.

The report writer is the mirror image. The output is written English that others rely on, so writing sits at C1 and reading at B2. Spoken fluency is pleasant but not essential, so speaking can pass at B1. Screening this role on a blended score risks rejecting an excellent writer whose speaking average pulled them down, or hiring a confident talker whose writing is not client-ready.

The front-desk role needs warm, clear spoken interaction and solid listening, but almost no formal writing, so writing can pass as low as A2. Setting a high writing bar here would simply shrink your candidate pool for no operational reason.

The point of the table is not the exact levels, which you should tune to your own context, but the shape: the same four skills, weighted completely differently, producing different pass marks for each job.

How many skills should you actually test?

If a per-skill breakdown is so useful, why not always test all four? Often you should. But there is a genuine trade-off between signal on one side and cost and candidate time on the other.

Testing all four skills gives you the complete profile. You can reuse one result across several similar roles, you future-proof against a job that evolves, and you never get caught out by a weakness in a skill you chose not to measure. This is the safest default, especially for roles where English is central.

Testing a focused subset can make sense when a role clearly leans on one or two skills and you are running high volumes where every extra minute per candidate adds up. A screening round for phone agents might prioritise listening and speaking, with the other two skills either skipped or recorded without a pass gate.

A few principles help you decide:

• The cost of missing a skill is asymmetric. Skipping a skill you assumed was fine is how bad hires slip through. When in doubt, test it; you can always choose not to gate on it.
• A full four-skill result is reusable. One assessment can serve multiple roles and future openings, which often makes "test everything once" cheaper than repeated narrow tests.
• Candidate experience matters, but a complete test is short. A four-skill assessment is not a day-long ordeal, and a fair, thorough process signals a serious employer.

Because the International English Test runs on a credit model at roughly £8.99–£11.99 per test depending on volume, with no contracts, the marginal cost of a fuller picture is modest and predictable. That changes the calculation: the four-skill result is usually worth having even when only two skills gate the decision. It is worth weighing this against the real cost of over-screening a role, where setting bars too high across too many skills quietly shrinks your pool and slows hiring.

Putting it together

Reading CEFR results by skill is the difference between a label and a decision. A blended score answers "roughly how good is their English?" A per-skill breakdown answers the question you actually care about: "can they do the parts of this job that happen in English?"

Do three things. First, define which skills the role uses and how much. Second, set a pass mark per skill, weighted to the role, and treat each as its own gate rather than averaging them away. Third, lean towards testing all four skills unless volume and a clearly single-skill role justify narrowing the scope; the fuller profile is cheap insurance against the weakness you did not think to measure. Backed by its status as an ALTE Associate Member, the International English Test gives you a CEFR level for every skill in minutes, so you can screen on the skill that matters instead of an average that hides it.

English assessment tests for companies