Every hiring system scores people against something. Write yours down.
Define what a job actually requires — the competency groups, the skills, the weight of each one, and five observable levels per skill. Then every CV, every assessment, and every performance review measures against the same written standard. Unlimited frameworks, unlimited roles, no expiry.
- No credit card
- No sales call
- Start from a library, not a blank page
- Tanca
- Skillify
- Glinks
- GCafe
- GHR
- GWork
“Not quite the right fit.” Against what, exactly?
Every hiring decision is a comparison against a standard. In most companies that standard lives in one person’s head, changes between interviews, and has never been written down.
Which means two candidates get measured differently, nobody can explain a decision six months later, and the AI score you’re now relying on has nothing solid to compare against.
A framework is just the standard, written before anyone is measured. It takes an afternoon to build and it’s the difference between a hiring process you can defend and one you can only describe.
Unlimited frameworks. One neighbouring feature has an allowance.
We’d rather draw the line clearly than let you find it later.
Unlimited, permanently
- Frameworks
- Job titles covered
- Competency groups per framework
- Skills per group
- Custom skills
- The skill library
- Five-level ladders
- Weighting
- Framework versions
- Assessments against a framework
- Gap reports
- Team and individual views
- Export
Draws on your AI allowance
Building frameworks, running assessments and reading the gap reports never costs anything. Only the AI pass that reads a CV against a framework consumes GPU time.
Your frameworks are yours. Export them any time, in full, including every level and criterion you wrote.
Groups, skills, weights, levels. That’s the whole model.
Deliberately small. A framework nobody can hold in their head is a framework nobody uses.
The builder
Name the framework, tie it to a job title, then add competency groups and the skills inside them.
Weights must reach 100% per group. The form won’t save otherwise.
The level ladder
Five levels per skill, each with a criterion someone could actually observe. Pick the one this role requires.
The skill library
Start from skills that already exist, or write your own. Either way they’re reusable across every framework.
“Good communicator” is not a level. This is.
Most competency models describe levels with adjectives — basic, proficient, advanced, expert. Two people reading the same adjective score the same person differently, which defeats the purpose of having a scale.
Every level in Career GPS carries a criterion someone could verify: a response rate, a compliance threshold, a satisfaction score, a risk level. When two managers assess the same person, they’re reading the same sentence.
You can edit every one of them. The library gives you a starting point, not a rulebook.
- Observable criteria, not adjectives
- Same wording for hiring and review
- Fully editable
This score compares the CV against the framework “Admissions Advisor (Telesales)”, version 3, published 12/06/2026. Every score records which framework version produced it.
A number is only as good as what it was compared to.
Career GPS doesn’t score a candidate against a general model of good employees. It scores them against the framework you wrote for that job title, using the weights you set.
That means the score decomposes. You can see that someone scored 82 because they’re at level 4 on client relationships, level 3 on consultative selling, and level 2 on compliance awareness — and that compliance awareness is weighted at 25% for this role because you decided it should be.
Which is what you need when a candidate asks why, when your legal team asks how, or when you’re explaining a decision to the person who has to work with the hire.
- Scored against your written criteria
- Every score decomposes to skills
- Weights are yours, and visible
Score them at application. Score them again at six months. Same framework.
Everywhere else, hiring assessment and performance review are two unrelated systems with two unrelated scales. Here they’re the same object — so the question “did the people we scored highly actually do well?” finally has an answer.
The gap you hired with
AI Score compares the CV against the framework and returns a level per skill. You see exactly which requirements this person doesn’t meet yet, before you make the offer.
The plan you set
Their starting levels become their development plan. Skillify builds a path for the specific skills below target — not a course catalogue, a route.
The gap you closed
The same skills, the same five levels, assessed again. The chart is directly comparable to day 0 because it’s the same measuring stick.
And then the question that nobody in this market can currently answer.
Once you have day-0 scores and day-180 assessments for enough people, you can measure whether the score predicted anything. Career GPS reports that correlation back to you for your own hires. If it turns out our scoring doesn’t predict your outcomes, you’ll be the first to know — that’s the point of measuring it.
Where someone is, against where the role needs them.
One chart per person, one per team, on the same scale.
The same chart is available for a whole team, a location, or a job title across the company.
Written before anyone was measured.
The strongest position in any dispute over a hiring or promotion decision is a job-related standard that existed in writing beforehand and was applied the same way to everyone.
Job-related on its face.
Every criterion describes something the role does. Nothing in the model asks about background, school, age, or personal characteristics.
Applied identically.
The same framework version scores every candidate for that role. Not a different standard per interviewer, per day, per mood.
Versioned and timestamped.
Every score records which framework version produced it, so a decision from March stays explainable after the framework changes in June.
Human decision, always.
The framework produces a measurement. It never rejects anyone, never hides anyone, and never makes the call.
The framework is also what a candidate gets back. When someone isn’t selected, the gap they see is against these written criteria — not against other applicants, and not against anyone’s opinion.
How candidates see their results →This describes how the product works and is not legal advice.
Your first framework takes an afternoon.
- 01
Pick the job title.
Start with the role you hire most, not the one that’s most complicated.
- 02
Take groups from the library.
Ready-made competency groups and skills for retail, hospitality, logistics, manufacturing, contact centres and sales.
- 03
Set the weights.
Decide what actually matters for this role. The weights are the part only you can write.
- 04
Choose the level per skill.
Read the five criteria, pick the one this role needs. Done — every future CV measures against it.
Already have frameworks in a spreadsheet? Import them, including your own level descriptions.
One definition, used by everything downstream.
Competency Framework
You are here
AI Score
Reads every CV against it
Assessments
Measures people against it
Tanca HRM
Performance reviews on the same scale
Skillify
Learning paths for the gaps it finds
Write the definition once, and hiring, assessment, review and training all stop arguing about what the job requires.
Measured inside the product.
Food retail chain · 22 stores
Share of assessors who agreed on a candidate’s level after reading the criteria.
Logistics company · 3 southern warehouses
Time to build a shortlist on a warehouse repeat opening.
Admissions centre · telesales
Share of new hires meeting target level at their six-month review.
You’ll see the same numbers for your own frameworks from the first assessment round.
Questions people ask about frameworks.
How long does one framework take to build?
An afternoon for your first, starting from the library. Fifteen minutes for the next one in the same job family, because the groups and skills are reusable.
Do we have to use five levels?
Yes. One scale across the whole product is what makes hiring scores and review scores comparable, and comparability is the entire value.
Can we write our own skills and criteria?
Yes, all of them. The library is a starting point you can overwrite completely.
What happens when we change a framework?
It creates a new version. Existing scores keep pointing at the version that produced them, so old decisions stay explainable.
Does this need Tanca HRM?
No. Frameworks, assessments and gap reports all work in Career GPS alone. Tanca HRM is what lets you re-assess the same person at six months on the same scale.
Is any of this used to rank employees against each other?
No. Every chart shows one person against the role’s requirements. There is no leaderboard and no forced distribution.
What do candidates see?
Their own levels against the role’s written criteria, and which skills they’d need to develop. Never a comparison with other applicants.
Can we export a framework?
Yes, in full, including every level and criterion, at any time.
Write down what the job requires.
Start from the library, set your weights, pick your levels. It costs nothing and it takes an afternoon.
Start free- No credit card
- No sales call
- Export any time