0. Decision and Data Foundations

Reproducible Analytics Workflow

Preserve the evidence chain from source snapshot to monitored decision

Reproducible Analytics Workflow

Reproducibility means more than rerunning a notebook. Another analyst must be able to reconstruct the data available then, execute the transformation and explain why the resulting number supports the decision.

The analysis manifest

ItemHarborMart example
question versiondelivery-promise-v3
decision cutoffcheckout confirmation timestamp
source snapshotwarehouse tables as of 2026-06-30 23:59 UTC
query or extract hashimmutable identifier for extraction logic
data dictionaryfield, unit, allowed values, owner
code environmentPython/package versions or browser runtime
random seedfixed where randomness is used
split definitionweeks 1–18 train, 19–21 validation, 22–24 test
metric definitionsformulas, denominator and exclusions
model/policy versionartefact plus threshold and constraints
reviewerperson independent of original calculation

Saving only the final CSV loses the query, cutoff and exclusions. Saving only code loses the historical source state.

Separate work by purpose

Rendering diagram…

The decision record should contain both the model output and the action actually taken. Otherwise overrides disappear and later evaluation confuses model performance with policy performance.

Tests at four layers

LayerExample test
schemaorder_id non-null and unique at order grain
semanticpromised end is after promised start
statisticaldaily order count within investigated operating range
decisionno feature has availability time after checkout

A passing schema does not establish a sensible business value. A statistical anomaly is not automatically an error; it may be the event the business needs to understand.

Reproduce, then verify

These are different:

  • reproduction: rerun the same code and obtain the same result;
  • verification: independently check whether the result answers the intended question;
  • validation: test performance on appropriate unseen or future cases;
  • audit: inspect controls, evidence, responsibility and change history.

Two analysts can reproduce the same leaked model. Reproducibility is necessary, not sufficient.

Responsible use of generative AI

Record where AI assisted with code, text or classification. For each material output:

  1. preserve the input context without exposing confidential data;
  2. verify functions and claims against authoritative documentation or source data;
  3. test edge cases, units and temporal logic;
  4. identify the human accountable for acceptance;
  5. keep a non-AI route for sensitive or prohibited data.

An AI-generated query that executes successfully can still multiply rows, use a future field or silently change a denominator.

Reproducibility handoff

A compact handoff contains:

  • README: question, action, owner and run order;
  • data contract and source snapshot identifier;
  • executable transformations and tests;
  • frozen evaluation set definition;
  • result table with uncertainty;
  • policy logic, override and rollback rules;
  • model/data/decision monitoring owners.

Quick check

The model file and notebook are versioned, but the classification threshold is edited manually in a dashboard. Is the decision reproducible?

Answer
No. The deployed decision is the model plus threshold and surrounding rules. Version the threshold, approval, effective time and affected population with the model.

Production perspective

Hidden Technical Debt in Machine Learning Systems explains why data dependencies, feedback and undeclared consumers create system risk. Google’s ML Test Score turns this concern into concrete pre-production and monitoring tests.

Next: Descriptive Analytics

Copyright © 2026