Prepare & analyse
Data checks & documented preparationFree browser release
Data Quality & Preparation Lab
What needs attention before this dataset is used?
Inspect missingness, duplicates, declared types, ranges and labels. Preview explicit cleaning and unit conversions while preserving the original data and a record of each change.
- Bring
- A small tabular CSV, its source and purpose, and declared column meanings, units and validation rules.
- Get
- Before/after profiles, diagnostic flags, a prepared CSV and a reproducible change and exclusion record.
Prepare & analyse
Study design · Statistical evidenceFree browser release
Research Statistics & Study Design Lab
What does the evidence estimate, and how uncertain is it?
Run a declared Python/SciPy analysis for descriptive estimates, independent or paired mean differences, Pearson correlation or precision sample-size planning. Keep the study design, exclusions and interpretation beside the result.
- Bring
- A research question, design and suitable numeric observations; or planning assumptions and a desired precision.
- Get
- Estimates and supported confidence intervals, p-values where applicable, source-row exclusions and a reproducible method record.
Prepare & analyse
Business indicators · Performance evidenceFree browser release
Business Intelligence & Performance Lab
Which performance measures support the next business decision?
Define KPIs from a documented fact table, filter the scope, inspect segments and time summaries, and compare the filtered total with current or alternative targets.
- Bring
- A bounded CSV with a clear row grain, numeric measures, units and optional dates, categories and targets.
- Get
- Scoped KPI totals, segment drilldowns, trend tables and charts, target gaps, coverage counts and an auditable dashboard report.
Model & evaluate
Python · Predictive evidenceFree browser release
Prediction & Decision Lab
What can the data predict, and how does that affect a decision?
Run Python on this device to train and compare classification or regression models. Keep cross-validation evidence separate from the final holdout test, then move a declared decision matrix into the connected labs.
- Bring
- A small tabular CSV, a target and selected input features.
- Get
- Model comparisons, new-data predictions, a reproducible Python notebook and optional weighting, ranking or sensitivity handoff.
Model & evaluate
Security evidence & decisionsFree browser release
Cybersecurity Decision Lab
Which security improvements deserve attention and investment?
Assess supporting evidence, prioritise remediation and select a control portfolio within a declared budget. Keep unknown evidence visible and examine how changed assumptions affect the decision.
- Bring
- Security findings, asset context, evidence and proposed controls with costs, dependencies or mandatory requirements.
- Get
- An evidence assessment, relative remediation priorities and a feasible control portfolio with sensitivity results.
Model & evaluate
Software quality & delivery decisionsFree browser release
Software Engineering Decision Lab
Which release, tests and improvements deserve attention?
Compare software quality using declared reference ranges and mandatory quality gates. Prioritise tests and select a feasible test set within the available time, then plan defect fixes, technical debt and improvements within a delivery capacity.
- Bring
- Comparable release metrics, quality requirements, component risks, test coverage and durations, plus backlog effort and dependencies.
- Get
- Eligible release rankings, a selected test set and coverage-aware execution order, and a feasible improvement portfolio with sensitivity findings.
Model & evaluate
Business fit, processes & user adoptionFree browser release
Information Systems Decision Lab
Which system fits the organization, and what evidence supports improvement?
Compare ERP, CRM, LMS or other system options against declared requirements and lifecycle costs. Examine process event logs and assess user adoption through actual responses and usage records, with independent workflows.
- Bring
- System requirements, cost evidence and alternatives; case/activity/timestamp logs; or a declared questionnaire, genuine responses and observed usage.
- Get
- An eligible system comparison and cost scenarios, process diagnostics, or construct-level adoption profiles with assumptions and evidence gaps.
Decide & explain
Criterion importance7-day trial / paid plan
MCDM Weighting Lab
How important is each criterion?
Work with expert judgements, performance data or declared hybrid choices. Inspect how each formulation produces criterion weights.
- Bring
- Defined criteria, expert judgements or a numerical performance matrix.
- Get
- Criterion weights, intermediate calculations and a report for the selected method.
Decide & explain
Alternative comparison7-day trial / paid plan
MCDM Ranking Lab
How do the alternatives compare?
Rank a decision matrix using declared formulations. Compare the ordering and examine the assumptions behind the result.
- Bring
- A decision matrix, criterion weights and benefit or cost directions.
- Get
- Scores, ranks and calculation reports from 13 declared ranking formulations.
Decide & explain
Sensitivity & uncertainty7-day trial / paid plan
MCDM Sensitivity Lab
What could change the decision?
Challenge a ranking by varying weights, criterion groups, observations and modelling choices under a declared experiment.
- Bring
- A decision matrix or compatible saved ranking, plus scenarios to examine.
- Get
- Rank changes and comparison reports for the scenarios you test.
Decide & explain
Explanations & evidence7-day trial / paid plan
CECE Research Lab
What supports the ordering?
Explore archived benchmark explanations and scoped certificate checks, or rank alternatives using your own decision matrix.
- Bring
- A published benchmark to explore, or your own decision matrix.
- Get
- Benchmark explanations and evidence, or the own-data calculator’s ranking report.
Decide & explain
AI system assessment7-day trial / paid plan
Trustworthy AI Lab
What supports confidence in this AI system?
Examine a named system in its intended setting. Record the evidence, technical findings and gaps across the declared assessment components.
- Bring
- The AI system’s intended use and evidence; optional binary-prediction outputs.
- Get
- Component findings, technical measures where supported and a record of evidence gaps.
Decide & explain
Decision assurance7-day trial / paid plan
Trustworthy MCDM Lab
What supports the whole decision?
Connect the purpose and constraints to the data, weights, ranking and uncertainty, then document the evidence needed for the intended use.
- Bring
- Alternatives, criteria, weights, constraints and supporting decision evidence.
- Get
- A decision record covering feasibility, ranking, sensitivity and evidence.