The same interview, at the depth each process asks for.

Here the bottleneck is not your team’s judgement: it is the calendar. There are more candidates than slots in the week, so the first cut ends up being made from résumés — the worst instrument available for finding out whether someone can actually do something.

The interview is configured short and focused: the three or four competencies that genuinely separate people, one item per competency, and a follow-up when the answer does not reach anything assessable. Each person gets their link and answers when they can — at night, on their day off, between shifts — and since there is no calendar to fill, there is no waiting list either.

What your team gets is not a ranking with an opaque number: it is a report per person, with the assessment of each competency and the sentence behind it, and a view that puts them side by side because everyone answered exactly the same script.

With no calendar to align, the constraint stops being how many interviews fit in the week and becomes how many reports your team wants to read.

A technical profile shows itself in the follow-up or it does not show itself at all. The difference between someone who did the thing and someone who was nearby while it got done appears when you ask for the detail: what they decided, against which alternative, what went wrong afterwards.

For that, the script is configured with more items per competency and more time per answer. The interview is instructed to ask for the how when it gets the what, and to close when a second follow-up produces nothing either: pushing a third time does not produce better evidence, it produces an uncomfortable person.

The report is written for someone who was not in the conversation and who does know the subject. The engineer who will work with whoever joins reads the verbatim answer and judges it themselves, without depending on Recruiting having correctly understood a term that is not theirs.

This is where the verbatim quote pays off most: whoever assesses the technical part is almost never whoever ran the interview.

Nobody is screening here. What is being built is a file that a committee, a board or a partner will read, and that has to survive a hostile reading. The competencies are about leading and about conflict, each answer gets far more room than in a volume round, and what matters is not only what the person answers but what they show while answering.

Depth here does not come from stretching one conversation: it comes from what gets asked and from how many rounds the process has. Fewer and harder competencies are declared, each answer gets more room, and the questions are about decisions that cost something: who they let go and how, which budget they defended and lost, where they were wrong and what they did the following quarter.

When someone on the committee asks where the idea comes from that this person has judgement for that, the answer is not an impression from whoever interviewed: it is the verbatim fragment, with its turn and its minute, and the recording behind it.

The interview with their future boss does not get replaced. What changes is that the conversation starts with the file already read.

A serious process has stages, and you need to follow one person through all of them without losing what they already said. One role carries as many rounds as your process needs, in the order you define, each with its own name and with the interviewer profile it calls for.

The person does not turn into three different candidates: their candidacy is one, and it accumulates whatever they answered in each round. The round’s name is frozen into the interview already answered, so renaming “First round” to “Technical” in September does not rewrite the file of whoever answered it in March.

Who moves to the next round is decided by someone on your team, or by your ATS through the API. At that moment the individual link for that round is issued, and your team or your ATS gets it to the candidate. The judgement to advance stays with whoever answers for the hire, and it is recorded with their name and the time.

The whole path stays on one candidacy, so the final conversation starts with everything the person already answered in view.

Criterio Talent separates two things almost everyone conflates: the account that groups, and the company that isolates. A recruiting agency, a group with several brands, or a corporation with subsidiaries all operate from a single account, and each client company lives inside its own space.

What gets separated is not cosmetic. Each company carries its own name and logo, its colour, its privacy notice, its retention period, and even the name the candidate hears when they start talking. One company’s legal notice shown to another company’s candidate is not a branding slip: it is an incident.

And belonging to the account does not grant access to the companies. Who gets into which is decided company by company, with distinct roles for administration, recruiting, the hiring manager, leadership and read-only access — because reading a report is not the same as downloading a recording.

For an in-house team with a single company, all of this is invisible and stays out of the way: the account and the company are the same thing.