Skip to content

Understand how AI changes work · without invented certainty.

Kasp is an AI system and can make mistakes. Check important details.

KASP / APP-1Work analysisKasp methodologyValues only after approval
APP-1 · Kasp AnalystRole, task or workflowKasp methodology · Values require approval

Kasp is an AI system and can make mistakes. Check important details.

Access is checked when you startSaved to: Personal history · Dossiers
KASP METHODOLOGY · IMPACT DIMENSIONSExactly eight impact dimensions
No universal overall score
  1. O01Technical capabilityWithheld
  2. O02Workflow distributionWithheld
  3. O03Reduction in human timeWithheld
  4. O04Review burdenWithheld
  5. O05Quality changeWithheld
  6. O06Augmentation impactWithheld
  7. O07Task substitution potentialWithheld
  8. O08Residual riskWithheld
APP-1Kasp methodology
Work analysisEight impact dimensions. No substitute values.
dimensions
8
Overall score
none
Publication
not approved
Analysis request

Describe a role, task or workflow.

Impact dimensions previewValues not approved
O01Technical capabilityApproval pending
O02Workflow distributionApproval pending
O03Reduction in human timeApproval pending

Public values appear only after approval of the relevant impact dimension.

ESCOO*NETISCOILOOECDIFRWEFEU AI Act
CONTEXT INTAKE · KASP METHODOLOGYRecord the analysis subject, workflow and AI deployment separately
Missing required context → no value
K01Occupation · Taxonomy identity and scope
Context
K02Specific role · Actual responsibilities and operating context
Context
K03Task · Distinct unit of work
Context
K04Workflow · Versioned process with a baseline
Context
K05Workflow step · Atomic step with an input and output
Context
K06Planned AI deployment · Specific system and control configuration
Context

Kasp treats user-supplied information as context, not independent evidence.

APP-1 · Kasp AnalystRole, task or workflowKasp methodology · Values require approval

Kasp is an AI system and can make mistakes. Check important details.

Access is checked when you startSaved to: Personal history · Dossiers
APP-1Kasp methodology
1.0 · IntakeA question becomes verifiable context.
6 context elementsWorkflow & baselinePlanned AI deployment
01
Occupation
record
02
Specific role
record
03
Task
record

01 Intake

A role becomes a task profile

Describe a role, task or workflow. Kasp records the analysis subject, baseline, planned AI deployment and missing required context separately.

1.0 Intake
EVIDENCE CONTRACT · KASP METHODOLOGYFour visible evidence classes, not silently mixed
per impact dimensionbound to statementsno substitute values
E01Evidence-backedQualified source bound to a statement and impact dimensionverifiableE02User contextSupplied context, not independent evidencedeclaredE03Kasp assessmentDerived assessment, explicitly labelled as an inferencelabelledE04Not approvedMissing evidence remains a visible data gapwithheld

A source does not automatically qualify every value of an impact dimension. User context and Kasp assessments are not presented as independent evidence; missing approvals remain visible.

APP-1Kasp methodology
2.0 · EvidenceProvenance stays visible.
4 Evidence classesbound to statementssafely withheld
E01
Evidence-backedQualified source bound to a statement and impact dimension
verifiableE02
User contextSupplied context, not independent evidence
declaredE03
Kasp assessmentDerived assessment, explicitly labelled as an inference
labelledE04
Not approvedMissing evidence remains a visible data gap
withheld

02 Evidence

Gather evidence, review it and keep it current

Sources are recorded with their origin, date, scope and binding to specific impact dimensions, separately from user context and Kasp assessments.

2.0 Evidence
KASP METHODOLOGY · IMPACT DIMENSIONSEight separate impact dimensions · no values until approval
No universal overall score.Public values have not yet been approved. Missing context or evidence does not produce a substitute value.
O01Technical capabilityCapability within the defined workflow and planned AI deployment.Separate evidence and approval requirednot approvedno public value
O02Workflow distributionDistribution between unchanged, assisted, partly automated and fully automated work.Separate evidence and approval requirednot approvedno public value
O03Reduction in human timeTotal human time, including review, correction and recovery.Separate evidence and approval requirednot approvedno public value
O04Review burdenHuman effort for review, correction, escalation and approval.Separate evidence and approval requirednot approvedno public value
O05Quality changeChange in a versioned, task-specific quality metric relative to the baseline.Separate evidence and approval requirednot approvedno public value
O06Augmentation impactAdditional benefit while the human role remains relevant.Separate evidence and approval requirednot approvedno public value
O07Task substitution potentialRealizable substitution subject to time, quality, legal and control criteria.Separate evidence and approval requirednot approvedno public value
O08Residual riskResidual risk after evidenced controls and their effectiveness.Separate evidence and approval requirednot approvedno public value

Shown: Task · exactly eight impact dimensions · no replacement of missing values and no aggregation into an overall result.

APP-1Kasp methodology
3.0 · AnalysisEight impact dimensions. No overall score.
8 dimensionsreviewed per dimensionApproval pending
ANALYSIS SUBJECTTask
O01Technical capabilitynot approved
O02Workflow distributionnot approved
O03Reduction in human timenot approved

03 Analysis

Eight impact dimensions instead of false precision

Work has eight separate impact dimensions: technical capability, workflow distribution, time, review burden, quality, augmentation, substitution and residual risk. Other analysis areas have their own result structures; there is no universal overall score.

3.0 Analysis

06 Methodology

Technical possibility is not automatic approval.

For work, the Kasp methodology separates exactly eight impact dimensions. Other analysis areas use their own result structures. Every value needs its own context, evidence, calibration, review and publication approval. If a prerequisite is missing, it remains withheld.

Approvals and scopes

Kasp methodology · eight impact dimensions

No universal overall score. Each impact dimension needs its own evidence and is approved separately.

  1. O01Technical capabilityCapability within the defined workflow and planned AI deployment.Value withheld until approval
  2. O02Workflow distributionDistribution between unchanged, assisted, partly automated and fully automated work.Value withheld until approval
  3. O03Reduction in human timeTotal human time, including review, correction and recovery.Value withheld until approval
  4. O04Review burdenHuman effort for review, correction, escalation and approval.Value withheld until approval
  5. O05Quality changeChange in a versioned, task-specific quality metric relative to the baseline.Value withheld until approval
  6. O06Augmentation impactAdditional benefit while the human role remains relevant.Value withheld until approval
  7. O07Task substitution potentialRealizable substitution subject to time, quality, legal and control criteria.Value withheld until approval
  8. O08Residual riskResidual risk after evidenced controls and their effectiveness.Value withheld until approval

07 Use

Two starting points. Clear boundaries of responsibility.

Personal analyses and Orbit workspaces serve different purposes. Kasp separates the two contexts and makes the permitted uses of results visible.

Personal analysis · B2C

My role and my tasks

Assess specific tasks, review sources and inform your learning or work decisions. Kasp predicts neither your individual career path nor whether a particular job will continue to exist.

Analyze your own role

Orbit · B2B

Role portfolios and transformation

Compare roles and tasks at an aggregate level, prioritize assistance and automation initiatives, and document actions. Evaluating identified employees or making automated hiring, performance, promotion or dismissal decisions is not an intended use.

Open Orbit workspaces

Examine the foundations of trust

The foundation sits beside the result.

Sources, data currency, coverage, confidence and limits should remain verifiable. Where evidence is missing, a gap must not appear as certainty.

The public catalog is a read-only reference view of released occupations, tasks and sources. Analysis functions, document processing and file uploads are technically and organizationally separate. No run starts without confirmed review; processing steps that have not been approved are not executed.

Kasp is an analysis and decision-support system. Results are not certainties, do not replace qualified review and must not be used as the sole basis for decisions about individuals.

Frequently asked questions

  • Does Kasp assess my occupation?

    Kasp assesses tasks, not job titles. An occupation is broken down into the tasks it actually consists of, and each task is assessed individually. Only then does a picture of the role emerge.

  • Where does the data come from?

    From official catalogs and cited sources: ESCO for occupations, O*NET for tasks and ISCO-08 as an international hub. Every statement carries its source, date and scope · available alongside the result.

  • What is behind an assessment?

    Exactly eight separate impact dimensions: technical capability, workflow distribution, change in human time, review burden, quality change, augmentation impact, task substitution potential and residual risk. Kasp does not combine them into a universal overall score.

  • What do coverage, confidence and evidence grade mean?

    Coverage indicates how much of the review chain for this task is actually supported. Evidence grade describes the strength of the evidence; confidence describes the certainty of the assessment. All three appear alongside the result, not in the small print.

  • Does Kasp predict whether a role will disappear?

    No. The impact dimensions refer to a specifically described workflow and planned AI deployment. Where required context, evidence or approval is missing, the value remains withheld · rather than appearing as a falsely precise number.

  • How do I get started?

    Describe your role in the chat. Kasp derives the tasks, maps them to the catalogs and shows the assessment with its sources. Choose your analysis depth in the input field: Free, Pro or Expert.

Task-based assessment. Starting today.

Start an analysisView pricing