Certification/Recommendation Tag similar in functionality and visibility to TIER

[DovileKr]

Context:
The current Tier system is excellent for identifying the importance of data assets based on downstream usage (right side of the data lineage). However, there’s a need for an additional tag to represent trust and readiness of assets created on the left side of the lineage, such as curated, governed, and documented tables or views. These assets may not yet have downstream usage but are trusted and recommended for use.

Proposal:
Introduce a Certification/Recommendation Tag alongside the existing Tier tag. The two tags would serve distinct purposes:

TIER: Highlights the importance of an asset to the organization (e.g., Tier 1, Tier 2).
Certification/Recommendation: Highlights the trust and quality of an asset, regardless of its usage.
The Certification/Recommendation Tag could have classification levels such as:

Grade 1: Certified and ready for use.
Grade 2: Trusted but still in evaluation.
Grade 3: Under development.
Grade 4: Not-for-use.
Grade 5: Uncertified or unknown origin.
Benefits:
Discoverability: Users can find important assets (via Tier) and trusted assets (via Certification) separately or in combination.
Governance: Promotes use of high-quality, governed data assets over raw or ad-hoc alternatives.
Flexibility: Enables clear differentiation between importance (Tier) and quality/trust (Certification).

[DovileKr]

In a data catalog, the concepts of TIER and GRADE serve distinct purposes for classifying and assessing data assets. Here’s how they differ:

---

TIER

Definition:
Tiers are typically used to categorize data assets based on criticality, business impact, or operational importance. It focuses on how the data asset is used in the organization and its overall priority.

Key Characteristics:

• Purpose: Prioritization and resource allocation.
• Common Criteria:
  • Business-critical vs. non-critical
  • Data availability requirements (e.g., high availability for Tier 1 data)
  • Support levels (e.g., Tier 1 might require 24/7 support)
  • Regulatory importance
• Examples:
  • Tier 1: Mission-critical assets like financial reports or operational dashboards.
  • Tier 2: Important but not business-critical assets.
  • Tier 3: Informational or exploratory data with minimal impact if unavailable.

---

GRADE

Definition:
Grades evaluate the quality or trustworthiness of the data itself, based on factors like accuracy, completeness, timeliness, and reliability.

Key Characteristics:

• Purpose: Establish confidence and usability in the data for decision-making.
• Common Criteria:
  • Data quality metrics (e.g., error rates, null values)
  • Data lineage and governance adherence
  • User feedback or ratings
  • Compliance with data standards
• Examples:
  • Grade A: High-quality, fully governed, and highly trusted data assets.
  • Grade B: Reasonably good quality with minor issues.
  • Grade C: Lower quality, incomplete, or unreliable data requiring caution.

---

Key Differences

Aspect	TIER	GRADE
Focus	Importance to the business	Quality and trustworthiness
Purpose	Operational prioritization	Usability and confidence
Based On	Business impact, availability	Accuracy, completeness, governance
Typical Use Case	Resource allocation, risk planning	Decision-making, quality assurance
Assessment Level	Macro (organization-level)	Micro (data quality-level)

---

By using both TIER and GRADE, an organization can effectively manage its data assets by prioritizing critical resources for the most important data (TIER) while ensuring that decisions are made based on reliable and high-quality information (GRADE).