← All posts
resourceJune 15, 2026·2 min read

The 20-point data platform readiness checklist

A practical self-assessment we run with every new client: twenty yes/no questions across ingestion, modeling, quality, governance, and cost that reveal where your platform actually stands.

Before any engagement, we run clients through the same twenty questions. Score one point per honest "yes." The goal isn't a grade; it's to make the gaps visible and fundable. Use it freely.

Ingestion & pipelines

  1. Can you add a new data source to the platform in under a week?
  2. Do pipeline failures page a human before a stakeholder notices?
  3. Can you replay or backfill any pipeline without hand-editing scripts?
  4. Is every pipeline's code in version control with review before deploy?

Modeling & transformation

  1. Is there a documented transformation layer (e.g. dbt) between raw data and consumption?
  2. Could a new analyst find the canonical "orders" table without asking anyone?
  3. Are core business entities (customer, order, product) modeled once, not per-team?
  4. Can you trace any dashboard number back to its source columns?

Quality & trust

  1. Do your critical tables have automated tests that run on every update?
  2. Is data freshness visible to consumers (not just engineers)?
  3. When a metric looks wrong, is there a defined owner to ask?
  4. Have you deprecated or deleted a data asset in the last quarter?

Governance & access

  1. Can you say, today, who has access to your most sensitive dataset?
  2. Is PII identified and handled under a written policy?
  3. Do metric definitions live somewhere versioned, not in tribal memory?
  4. Could you answer a regulator's data-lineage question within a week?

Cost & operations

  1. Do you know last month's total platform spend within 10%?
  2. Can you attribute that spend to teams or use cases?
  3. Has someone reviewed and killed unused compute/storage in the last quarter?
  4. Could the platform survive its most knowledgeable engineer leaving?

Reading your score

  • 16–20: You have a platform. Your leverage now comes from analytics products and AI on top of it.
  • 10–15: Solid bones, real gaps. Prioritize the "no" answers in quality and governance first; they erode trust fastest.
  • 5–9: You have pipelines, not a platform. Resist new use cases until the foundation stops shifting.
  • 0–4: Good news: you get to skip a decade of legacy decisions and build it right the first time.

If you'd like a second pair of eyes on your answers, we do this assessment as a structured two-week engagement, with a prioritized roadmap as the output.

#Data Engineering#Checklist