Get instant results for objective questions
Multiple choice, matching, true/false and other objective questions are scored instantly based on predefined rules. No waiting, no manual work.
Fine-grained scoring rules for every scenario
Each answer option can be assigned a percentage value between 100% and +100%, supporting both positive and negative scoring. Partial credit can be enabled so test-takers receive proportional credit for partially correct selections.
- Assign percentage values between 100% and +100% per answer option
- Apply negative multipliers to penalize incorrect answers and discourage guessing
- Enable partial credit for multi-choice questions
- Set minimum score thresholds so low-quality partial answers receive 0 points
- Restrict the number of answer choices a test-taker can select
Standardize open-ended evaluation with rubrics
Open-ended responses are evaluated against rubrics that define clear scoring standards. Each criterion carries a weight that determines its contribution to the final score.
Multiple rubric scoring approaches in one editor
Weighted Criteria
Each criterion row can carry a different weight, determining its contribution to the final score.
100%
75%
50%
0%
Scoring Approaches
Assign fixed percentage values per level: Exemplary = 100%, Good = 75%. Or define custom intervals and allow manual input during evaluation.
Evaluator Feedback
Rubric structure also supports structured evaluator feedback. Evaluators can provide detailed written notes while scoring each response.
Generate structured scoring rubrics with AI
Generate evaluation rubrics instantly from assessment questions. AI proposes structured scoring criteria and performance levels that you can review, refine, and use.
Generate rubrics from assessment questions
AI analyzes the assessment question and proposes a structured rubric with evaluation criteria. Optionally, you can provide additional instructions to guide the generated rubric.
Support flexible scoring models
Generated rubrics can be used with both Level Selection and Score Entry scoring methods, supporting different evaluation workflows.
Preserve existing rubric settings
When regenerating an existing rubric, AI refreshes the evaluation criteria while preserving your current scoring configuration and rubric settings.
Review and customize before saving
Review and edit generated criteria, performance levels, descriptions, and scores before saving to ensure the rubric matches your evaluation standards.
Scale open-ended evaluation with AI
Essays, long answers, spoken responses, video interviews, and coding tasks are evaluated using AI systems powered by large language models guided by your own scoring instructions.
Author instructions guide the AI, not just the response
While creating a question, authors provide evaluation instructions and scoring criteria. The AI then combines the participant's response with the question and those guidelines to generate a score. Different LLMs can be used depending on the question type.
- Written responses
- Audio responses (automatically transcribed and evaluated)
- Video responses (automatically transcribed and evaluated)
- Coding submissions
- Flexibility to integrate with different AI models depending on assessment needs
Automate grading for open-form answers without AI
Short answers and numeric inputs can be graded automatically using text matching rules or custom JavaScript functions, without requiring AI evaluation.
Rule-based grading
Responses are evaluated using predefined text rules: exact string matching, regular expressions, or semantic equivalents. Multiple acceptable answers and paraphrased variations can be defined so the system recognizes different ways of expressing the same idea.
- Exact string matching
- Regular expression matching
- Semantic equivalents and paraphrased variations
- Custom feedback triggered automatically when rules are not met
Function-based grading
Responses can be evaluated using custom JavaScript functions written in a built-in code editor. Each function processes the test-taker's response and returns a score based on the defined logic.
- Full credit, no credit, or negative scoring
- Exact matching, partial matching, regex checks, character-level analysis
- Timing-based adjustments and penalties for errors
- Different evaluation functions for different question types or scenarios
Apply human judgment where it matters most
Essays, written explanations, spoken answers, and video submissions can be reviewed and graded manually. Evaluators provide written feedback and evaluation notes while scoring. Manual grading is useful when subjective judgment, detailed interpretation, or nuanced feedback is required during the evaluation process.
Evaluator feedback
Evaluators can write detailed reviews to share with the test-taker, providing clear feedback on performance, strengths, and areas for improvement.
Multiple evaluators
Multiple evaluators can review the same response when needed, helping maintain fairness and reliability in subjective assessments.
A straightforward scoring methodology that stays flexible
The test score is calculated by dividing the sum of points earned by the total points of the test. Each question awards the test-taker a percentage score, which is then multiplied by the question's point value.
Questions marked as ineffective or ignored are excluded from both the total points of the test and the points earned.
Turn grading results into decision-ready reports
Once responses are scored, results can be presented through structured scorecards, dimension-level breakdowns, and branded PDF reports. This helps teams review outcomes, compare performance, and share consistent results with candidates, employees, or stakeholders.
-
Overall, section, question, and dimension-level scores
-
Branded PDF reports
-
Result visibility controls for admins and test-takers
Built for any scenario where evaluation must be accurate, consistent, and scalable
From hiring to certification, apply the right grading method for each assessment, automate where possible, standardize where needed, and rely on human judgment when it matters.
Evaluate employees and candidates with consistent scoring
Use AI, automated scoring, rubrics, and human review to evaluate candidates, employees, and trainees with the same level of consistency. Support hiring decisions, training validation, and skill-gap analysis with reliable results.
Ensure defensible and standardized scoring
Apply strict evaluation criteria to maintain fairness and consistency in high-stakes exams. Combine automated scoring with human review where needed, and ensure every result can be justified and audited.
Evaluate at scale without losing grading quality
Grade essays, projects, and exams across large student groups using a mix of automated and manual methods. Maintain consistency between evaluators while reducing grading time.
Evaluate language skills with structured and scalable scoring
Assess writing, speaking, listening, and reading using a combination of AI evaluation, predefined scoring criteria, and human review where needed. Ensure consistent scoring across candidates while scaling language assessments efficiently.
Evaluation you can trust at scale
Apply consistent scoring across every candidate and scenario.