Review the business assessment
Synthetic development workspace. Candidate acceptance records your judgment about the disclosed proposal. It does not authorize implementation or verify a forecast.
Your next step
Sign in, then load the saved assessment.
Go to the next stepSign in to load the saved assessment.
Your session
Prepare from Markdown documents
Upload up to ten synthetic documents. Supported formats: four-column fact tables or Revenue: 120000 [USD/month; observation]. Other text stays visible for manual review. This is bounded extraction, not general language understanding.
For named-column diagnostic tables, start the packet with # BPB diagnostic tables 1.1. Supply Key (or Field/Metric), Value, Unit and Classification columns in any order; extra columns are retained. Field labels may use spaces and capitalization. Classification and units remain explicit; assumptions and unknowns are not observed values. For formatted numbers use # BPB diagnostic tables 1.2 and a Numeric Format column containing decimal-dot, en-US or de-DE. Grouping, decimal marks and currency/percentage marks follow that explicit declaration. Ambiguous or malformed values stay held; inspect the original token and conversion in source reconciliation. Parser # BPB diagnostic tables 1.3 also supports a finite set of documented field labels, including Direct Costs, COGS, On-Time Orders and Invoice Issuance Lag. Original labels and mappings remain visible. Ambiguous terms such as Turnover or Payment Delay remain unmapped. Parser # BPB diagnostic tables 1.4 requires Definition, Population, Reporting Basis, Period Start and Period End columns for numeric rows. Missing context, differing definitions for the same field, and incompatible populations or bases hold affected findings. Row dates must match the assessment period. Matching declarations are not verified source truth. Prose is retained separately in the business narrative inventory. Exact section headings Objectives, Operating Practices, Capabilities and Constraints organize statements; optional [Observation], [Interpretation] or [Assumption] labels preserve the source’s declared classification. Unlabeled prose remains unclassified and no narrative claim changes computed findings. A separate same-scope Markdown workpaper headed # BPB explanatory links 1.0 can link a finding and narrative statement IDs to a proposed explanation, an alternative and a validation question. Obtain statement IDs from source inspection; keep the original source bytes unchanged. Workpapers begin as unreviewed hypotheses; missing or stale links are held. After saving, select a hypothesis in the existing review controls. Acceptance is for investigation only; only current authenticated acceptances add validation questions to the research agenda. A same-scope source headed # BPB research evidence 1.0 can bind publication metadata, an inspected exact passage, claim scope, evidence design, causality, access, applicability and limitations to a hypothesis. Save the packet, accept the investigation hypothesis, then review the research claim. Acceptance confirms passage support for appraisal only. For retrieved evidence use # BPB research evidence 1.1 and append Capture Digest, Evidence Strength and Contrary Evidence columns. Curated research leads show retained capture metadata. The complete quoted excerpt and citation must match that capture; a text match does not establish claim support. Publication dates may retain year/month precision or not_stated when that is the captured value. Preview the supported business-specific tables, inspect sources and limits, then save a diagnostic assessment in an empty store. Intervention research remains pending.
Unmapped material that requires review
Revise this assessment’s source packet
Supply the complete replacement packet for the same business and reporting period. Earlier saved versions remain available.
For diagnostics, run the replacement business diagnostic preview first. Correction archives prior reviews, selections, plans and outcomes against their original sources; fresh review is required.
Missing fields are not invented. Conflicting values require an explicit choice and reason. Initial preparation cannot overwrite an existing workspace. Use the revision preview to change an existing assessment.
Saved assessment
No saved snapshot loaded.
Create is available to preparers and cannot overwrite an existing saved workspace.
Assess research applicability
Saving the first appraisal activates claim-level inclusion requirements and makes prior selections and reviews stale. Each included action then needs a current applicable-for-test judgment for every linked source. Method support does not prove a business effect.
Prerequisite evidence needed
Select the candidate portfolio
Inspect screening reasons before choosing each action. Inclusion proposes a bounded test; it does not authorize implementation. Deferred actions need a start trigger. Saving a selection makes earlier reviews stale.
Reloading replaces these fields with the latest saved selection. Recheck choices after a conflict.
Selected ranking, schedule and history
Explore costs, timing and priorities
Optional multi-criteria priorities require explicit weights, scoring rubrics, score ranges and reasons in the Markdown packet. Inspect contributions and sensitivity in the details. Missing judgments remain unranked; priority does not override the feasible schedule.
Upload explicit synthetic planning assumptions in Markdown. Inspect the preview, then explicitly save a conditional scenario into the Blueprint. Unknown effects remain unknown; hypothetical ROI is not a forecast.
Inspect saved assumptions, calculations and history
Measure observed outcomes
Record a plan for a currently included action, then upload synthetic observation evidence. A difference from baseline does not prove the action caused it. This screen does not authorize implementation.