BPO and contact centres

Screen a hundred agents without booking one interview

Send a link, get a CEFR level back in minutes. 7 minutes an agent for a voice queue, about 20 if you need writing too. 12 languages, no scheduling and no live proctor, from $0.60 a test.

Start free with 10 assessmentsSee pricing
100agents assessed
0interviews scheduled
7 mineach, for a voice queue

You are hiring forty agents this month. Every one of them ends up on a call with someone already annoyed.

You cannot interview your way through that. Your recruiters are not language examiners, your team leads are on the floor, and a sit-down test with a live examiner turns a two week ramp into a six week one.

So most people guess. Listen for a minute on the screening call, make a judgement, move on. It works often enough that nobody questions it, and it fails expensively: the agent who reads beautifully and freezes live, the one whose email follow-up gets rewritten by a team lead every single time.

Testing it takes 7 minutes an agent for voice, about 20 if you need writing too, and costs $0.60. That is the entire pitch.

Three places it earns its keep

  1. Before you hire

    Screen the whole shortlist

    Every candidate sits the same test with the same timings, so you compare people rather than interviewers.

  2. After you hire

    Check the floor on a cycle

    Run it monthly or quarterly across your existing agents. Same test, so month six sits next to week one.

  3. Before you promote

    Move people up with evidence

    An escalation desk needs more language than a scripted queue. A level tells you who is ready for it.

What actually changes

Screening by interview
  • A bilingual interviewer has to be free
  • Every candidate needs a slot in someone’s calendar
  • Two interviewers reach two different verdicts
  • The verdict is an impression, not a record
  • Forty candidates is forty conversations
Screening by assessment
  • Nobody has to be available for anything
  • Candidates take it between calls or after a shift
  • Same structure, same timings, same rubric
  • A CEFR level with ten scores behind it
  • Forty candidates is one bulk send
Your client asked forB2

Now you can show them B2

CEFR is the international standard, so an agent level is a fact your client can check rather than your opinion of your own staff. Every result carries the ten dimension scores and the reasoning behind the level, so it holds up when somebody asks how you know.

Same scale in all 12 languages, so a B2 in Manila means what a B2 in Bucharest means. That is what makes a group-wide standard possible at all.

A1A2B1B2C1C2

One scale across every site, every language and every queue.

$0.60

an agent

No setup fee, no per-seat licence, no proctoring surcharge. Screening a hundred candidates costs a hundred tests and nothing else. Start free with ten and pay by card when you are ready.

See full pricingRead the API docs

Questions people ask

Can we assess agents we have already hired?
Yes, and it is one of the commonest reasons people use it. Send the same assessment to your existing floor on whatever cycle suits you, monthly or quarterly, and you get a CEFR level per agent you can track over time. It is the same test either way, so a pre-hire score and a six-month score are directly comparable.
How many agents can we run at once?
As many as you like. Nothing is scheduled and nobody has to be available, so a hundred assessments take the same wall-clock time as one. Each agent gets their own link and takes it when their queue allows.
Can we show a client what level our agents are?
Yes. Every result is a CEFR level from A1 to C2, which is the international standard, so "our agents are B2" means something specific to a client rather than being your opinion of your own staff. Each result carries ten dimension scores and written reasoning behind the level, so it holds up when someone asks how you know.
Do agents need to book a time or be watched?
Neither. There is no live proctor and nothing in anyone's calendar. An agent opens their link between calls or at the end of a shift and takes about 7 minutes over a speaking assessment. Integrity is handled inside the test: randomised questions, no re-recording, no playback, and automatic checks for odd scoring patterns.
Does it work for chat and email agents, not just voice?
Yes. Speaking and writing are assessed separately, each on five dimensions, so you can run speaking only for a voice queue, writing only for a chat or email queue, or both. A written result covers task response, organisation, vocabulary, grammar and conventions, which is closer to what a chat agent is actually judged on.
Which languages can you assess?
12: English, Spanish, French, German, Portuguese, Italian, Chinese (Mandarin), Japanese, Korean, Russian, Dutch, Swedish. Every language is scored against the same CEFR scale, so a B2 means the same thing across every site you run.
What does it cost per agent?
From $0.60 per assessment. No setup fee, no per-seat licence and no proctoring surcharge, so the cost of screening a hundred candidates is the cost of a hundred tests and nothing else.
Can it connect to our ATS or workforce system?
Yes. Everything is available over a documented REST API: create assessments in bulk, poll status, pull back per-answer scores. Webhooks fire the moment a result is ready, and there is a test mode so your developers can build the integration without spending credit or waiting on real agents.

Try it on ten of your own agents

Ten free assessments, no card. Run them on agents your team leads already have an opinion about. If the levels match what you already know, you have your answer.

Get started free

No credit card required.