⚔️ CDMP Confusion Pair Finder
Use this page when two answers both sound reasonable. The goal is to isolate the deciding clue before you return to questions.
🔎 Master Topic Finder · ⚔️ Cross-Domain Confusion Book · 🧭 What Should I Study?
How to use a confusion pair
- Read the one-line deciding clue below.
- Open the linked Compare page.
- Say the difference aloud without looking.
- Give one example of each side.
- Then work a Scenario or Question Bank item.
Cross-domain priority pairs
| Confusion |
Deciding clue |
Best practice |
| Data Governance vs Data Management |
Governance decides/authorizes/sets decision rights; Data Management executes the broader management work. |
Cross-domain |
| Data Architecture vs Data Modeling & Design |
Architecture is enterprise blueprint/alignment; modeling creates precise conceptual/logical/physical representations. |
Cross-domain |
| Data Model vs Metamodel |
A data model represents domain data; a metamodel defines the structure/rules for models/Metadata. |
Cross-domain |
| Business Glossary vs Data Dictionary vs Catalog |
Glossary defines business meaning; dictionary details data structures; catalog helps discover/find assets and Metadata. |
Metadata Battle Cards |
| Reference Data vs Master Data |
Reference = controlled classification/value sets; Master = persistent real-world entities reused across processes. |
RMD Battle Cards |
| Profiling vs Initial DQ Assessment vs Monitoring |
Profiling describes actual patterns; assessment compares baseline to requirements; monitoring repeats operational checks over time. |
DQ Battle Cards |
| Prevention vs Correction vs Root-Cause Remediation |
Correction fixes bad data; root-cause remediation fixes the cause; prevention stops defects from being created. |
DQ Battle Cards |
| Data Owner vs Data Steward vs DQ Analyst |
Owner carries business accountability/decision authority; Steward operationalizes stewardship; DQ Analyst analyzes/measures quality. |
Cross-domain |
| Staging vs ODS vs Data Warehouse vs Data Mart |
Staging is transient processing; ODS supports current operational integration; DW is governed historical analytics; Mart is subject/consumer-focused analytics. |
DW/BI Battle Cards |
| Fact vs Dimension vs Conformed Dimension |
Fact records measures/events; Dimension provides descriptive context; Conformed Dimension is consistently reused across marts/processes. |
DW/BI Battle Cards |
| Mapping vs Remediation vs Transformation |
Mapping states source-to-target correspondence; remediation fixes defects; transformation changes representation/structure. |
DW/BI Battle Cards |
Data Management — Chapter 1
| Confusion |
Deciding clue |
Compare |
| Data Lifecycle vs Data Lineage |
Lifecycle = stages data passes through; lineage = path/origin/movement/transformation. |
Battle Cards |
| Data Lifecycle vs SDLC |
Data lifecycle follows the data; SDLC follows system/software development. |
Battle Cards |
| Data Strategy vs Data Management Program Strategy |
Data Strategy connects business need to data direction; program strategy organizes how Data Management capabilities will be built/executed. |
Battle Cards |
| Charter vs Scope Statement vs Roadmap |
Charter authorizes; Scope Statement bounds; Roadmap sequences implementation. |
Battle Cards |
Ethics — Chapter 2
| Confusion |
Deciding clue |
Compare |
| Legal vs Ethical |
Legal asks what rules permit/require; ethical reasoning also asks what responsible handling should be. |
Battle Cards |
| Bias vs misleading visualization/definition |
Bias skews selection/analysis; misleading presentation or definitions can distort interpretation even with technically accurate calculations. |
Battle Cards |
Governance — Chapter 3
| Confusion |
Deciding clue |
Compare |
| Policy vs Standard |
Policy states governing intent/rule; standard specifies required conventions/criteria. |
Battle Cards |
| Governance body vs Stewardship role |
Governance body makes/ratifies decisions; stewardship carries domain/data responsibilities into operation. |
Battle Cards |
| Decision right vs task responsibility |
Decision right = who may decide/approve; task responsibility = who performs the work. |
Battle Cards |
Architecture — Chapter 4
| Confusion |
Deciding clue |
Compare |
| Enterprise Data Model vs Data Flow Design |
EDM explains enterprise meaning/relationships; flow design explains movement, storage, transformation and context. |
Battle Cards |
| Target State vs Roadmap |
Target State is destination; Roadmap is sequenced path and transition states. |
Battle Cards |
| Emerging vs Preferred technology/pattern |
Emerging can be pilot/evaluation; Preferred is approved/recommended for standard use. |
Battle Cards |
| Zachman vs methodology |
Zachman classifies architecture descriptions; it does not prescribe an implementation sequence. |
Battle Cards |
Modeling & Design — Chapter 5
| Confusion |
Deciding clue |
Compare |
| Conceptual vs Logical vs Physical Model |
Conceptual = business concepts; Logical = detailed technology-independent structure; Physical = implementation-specific design. |
Battle Cards |
| Entity vs Instance |
Entity = class/type; Instance = one occurrence. |
Battle Cards |
| Cardinality vs Arity |
Cardinality describes relationship participation/count; arity is number of entity types participating in a relationship. |
Battle Cards |
| Candidate vs Primary vs Alternate Key |
Candidate can uniquely identify; Primary is selected candidate; Alternate is unselected candidate. |
Battle Cards |
| Business/Natural vs Surrogate Key |
Business key carries domain meaning; surrogate is artificial/system-assigned. |
Battle Cards |
| Normalization vs Denormalization |
Normalization reduces redundancy/anomalies; denormalization deliberately adds redundancy/combines structures for performance/usability. |
Battle Cards |
| Fact vs Dimension |
Fact captures measures/events; Dimension gives descriptive analysis context. |
Battle Cards |
| View vs Materialized View |
View stores query definition; materialized view persists the result. |
Battle Cards |
| Forward vs Reverse Engineering |
Forward moves model/design toward implementation; reverse derives model/design from existing implementation. |
Battle Cards |
Storage & Operations — Chapter 6
| Confusion |
Deciding clue |
Compare |
| Database Technology Support vs Database Operations Support |
Technology Support manages DBMS/platform technology; Operations Support runs stored-data operational processes/services. |
Battle Cards |
| Backup vs Archive/Retention |
Backup supports recovery; archive/retention manages lifecycle preservation and policy obligations. |
Battle Cards |
| Replication vs Backup |
Replication maintains another usable copy/state for availability/distribution; backup is a recovery point. |
Battle Cards |
Security — Chapter 7
| Confusion |
Deciding clue |
Compare |
| Threat vs Vulnerability vs Risk |
Threat can cause harm; vulnerability is exploitable weakness; risk combines likelihood/impact/exposure to loss. |
Battle Cards |
| Role vs Entitlement |
Role groups responsibilities/access logic; entitlement is a specific granted permission/right. |
Battle Cards |
| Monitoring vs Independent Audit |
Monitoring is ongoing control evidence; independent audit provides separate assurance/review. |
Battle Cards |
Integration — Chapter 8
| Confusion |
Deciding clue |
Compare |
| ETL vs ELT |
ETL transforms before target load; ELT loads then transforms in target environment. |
Battle Cards |
| Mapping vs Orchestration |
Mapping defines correspondence/transformation; orchestration coordinates sequence/dependencies/execution. |
Battle Cards |
| CDC vs Full Comparison |
CDC captures changes incrementally; full comparison compares full states to infer changes. |
Battle Cards |
| Integration vs Interoperability |
Integration combines/moves/coordinates data; interoperability emphasizes systems' ability to exchange/use data meaningfully. |
Battle Cards |
Document & Content — Chapter 9
| Confusion |
Deciding clue |
Compare |
| Content vs Document vs Record |
Content = information; Document = container/object; Record = evidence with stronger lifecycle controls. |
Battle Cards |
| Content Management vs Records Management |
Content Management optimizes creation/use/delivery; Records Management governs evidentiary retention/disposition. |
Battle Cards |
| Taxonomy vs Thesaurus vs Ontology |
Taxonomy classifies hierarchically; thesaurus adds controlled semantic relationships/synonyms; ontology models richer concepts/relationships. |
Battle Cards |
| Retention vs Disposition vs Legal Hold |
Retention says how long; disposition is authorized end action; Legal Hold suspends normal disposition. |
Battle Cards |
| Precision vs Recall |
Precision = relevance of returned items; Recall = proportion of relevant items successfully found. |
Battle Cards |
| Push vs Pull vs Interactive delivery |
Push sends proactively; Pull is user-requested retrieval; Interactive supports ongoing two-way interaction. |
Battle Cards |
Reference & Master Data — Chapter 10
| Confusion |
Deciding clue |
Compare |
| Reference vs Master vs Transaction vs Metadata |
Reference classifies; Master identifies persistent entities; Transaction records events; Metadata describes data/meaning/context. |
Battle Cards |
| System of Record vs System of Reference |
SoR is authoritative recording system for a domain/process; SoRef is authoritative source for shared Reference Data. |
Battle Cards |
| Trusted Source vs Golden Record |
Trusted Source is selected reliable source; Golden Record is reconciled authoritative representation assembled from sources. |
Battle Cards |
| Deterministic vs Probabilistic matching |
Deterministic relies on exact/rule conditions; probabilistic uses likelihood/scoring across evidence. |
Battle Cards |
| False Positive vs False Negative match |
False positive merges/links things that are not the same; false negative fails to link things that are the same. |
Battle Cards |
| Duplicate Identification vs Match-Link vs Match-Merge |
Identify flags candidates; link preserves records with relationship; merge consolidates representation. |
Battle Cards |
| Registry vs Transaction Hub vs Consolidated MDM |
Registry indexes/links; Transaction Hub actively manages master transactions; Consolidated collects/reconciles a central master view. |
Battle Cards |
| Standardization vs Enrichment vs Identity Resolution |
Standardization normalizes form; enrichment adds context/attributes; identity resolution determines which records represent the same entity. |
Battle Cards |
DW/BI — Chapter 11
| Confusion |
Deciding clue |
Compare |
| BI vs Data Warehouse vs Data Warehousing |
BI is analytical use/delivery; Data Warehouse is analytical data store; Data Warehousing is the broader discipline/process. |
Battle Cards |
| Inmon/CIF vs Kimball |
Inmon/CIF emphasizes integrated enterprise DW architecture; Kimball emphasizes dimensional delivery with conformed dimensions. |
Battle Cards |
| Staging vs ODS vs DW vs Mart vs Cube |
Use purpose, persistence/history, operational vs analytical use, and audience/subject scope. |
Battle Cards |
| Trickle vs Messaging vs Streaming |
Trickle = source-side incremental feeding; Messaging = bus/event transport; Streaming = continuous target/processing pattern. |
Battle Cards |
| Operational/Tactical vs Strategic BI |
Operational/tactical supports immediate/near-term actions; strategic supports longer-horizon direction/decisions. |
Battle Cards |
| Usage vs Coverage vs Performance vs Satisfaction |
Each metric answers a different adoption/service question; none substitutes for the others. |
Battle Cards |
| Confusion |
Deciding clue |
Compare |
| Business vs Technical vs Operational Metadata |
Business = meaning/governance; Technical = implementation/movement; Operational = runtime/process evidence. |
Battle Cards |
| Centralized vs Distributed vs Hybrid vs Bi-Directional |
Ask what persists centrally, what is queried at source, what is selectively copied, and whether updates can flow back. |
Battle Cards |
| Integrate vs Deliver Metadata |
Integrate creates a coherent Metadata body; Deliver makes it usable/available to consumers. |
Battle Cards |
| As Designed vs As Implemented |
Designed = intended model/flow; Implemented = actual code/configuration/runtime implementation. |
Battle Cards |
| Lineage vs Impact Analysis |
Lineage traces where data came from/went; Impact predicts what could be affected by change. |
Battle Cards |
| Metadata Administration vs Governance |
Administration operates/manages repositories/processes; Governance sets policy, accountability, standards and decision rights. |
Battle Cards |
| Completeness vs Usage vs Coverage vs Quality |
Completeness asks whether expected Metadata is present; usage asks whether people use it; coverage asks scope; quality asks correctness/fitness. |
Battle Cards |
Data Quality — Chapter 13
| Confusion |
Deciding clue |
Compare |
| Validity vs Accuracy |
Validity = conforms to allowed rule/domain; Accuracy = correctly represents reality. |
Battle Cards |
| Completeness vs Integrity |
Completeness = required values/records present; Integrity = relationships/coherence preserved. |
Battle Cards |
| Timeliness vs Currency |
Timeliness = when data becomes available; Currency = how current/stale the fact is. |
Battle Cards |
| Consistency vs Uniqueness |
Consistency = representations agree; Uniqueness = one representation per real-world entity where required. |
Battle Cards |
| Reasonableness vs Validity |
Reasonableness = plausible; Validity = formally allowed by rule/domain. |
Battle Cards |
| Dimension vs Rule vs Metric vs Threshold |
Dimension = quality characteristic; Rule = test; Metric = measured result; Threshold = acceptance cutoff. |
Battle Cards |
| Plan vs Do vs Check vs Act |
Plan defines/assesses; Do implements; Check measures; Act responds/improves. |
Battle Cards |
| Manage Rules vs Monitor Quality |
Manage Rules governs rule lifecycle; Monitor executes/observes quality results. |
Battle Cards |
| DQ SLA vs Scorecard vs Issue Report |
SLA is commitment; Scorecard summarizes performance; Issue Report tracks specific problems/action. |
Battle Cards |
| Parsing vs Standardization vs Enrichment |
Parsing decomposes; Standardization normalizes representation; Enrichment adds information/context. |
Battle Cards |
| Fully Automated vs Manually-Directed vs Manual Correction |
Automated needs no human case review; manually-directed uses automation plus human judgment for uncertainty; manual is human-executed. |
Battle Cards |
| Control Limit vs Acceptance Threshold |
Control limit is statistical process behavior boundary; threshold is business acceptance criterion. |
Battle Cards |
Big Data & Data Science — Chapter 14
| Confusion |
Deciding clue |
Compare |
| Big Data vs Data Science |
Big Data concerns scale/variety/velocity and architecture; Data Science concerns analytical discovery/modeling from data. |
Battle Cards |
| Descriptive vs Predictive vs Prescriptive |
Descriptive explains what happened; predictive estimates what may happen; prescriptive recommends what to do. |
Battle Cards |
| ETL vs ELT |
Transform before load vs load before transform. |
Battle Cards |
| Lake vs Swamp |
Lake is governed/usable raw-diverse store; swamp is poorly governed/unusable accumulation. |
Battle Cards |
| Supervised vs Unsupervised vs Reinforcement |
Labeled outcomes vs pattern discovery without labels vs learning through reward/feedback. |
Battle Cards |
| Training vs Validation vs Test |
Training fits; validation tunes/selects; test estimates final generalization. |
Battle Cards |
Maturity — Chapter 15
| Confusion |
Deciding clue |
Compare |
| Maturity Level vs Within-Level Score |
Level represents organizational capability stage; a criterion score evaluates an item within an assessment and does not equal the maturity level. |
Battle Cards |
| Assessment vs Improvement |
Assessment establishes/interprets current state; improvement changes capability after assessment. |
Battle Cards |
| Opinion vs Evidence |
Participant opinion is input; maturity conclusion should be reconciled against evidence. |
Battle Cards |
| Current State vs Target State |
Current = evidenced capability now; Target = capability required by business need. |
Battle Cards |
| Lowest Score vs Highest Priority |
Priority comes from business-relevant gap/risk/value, not score alone. |
Battle Cards |
Organization & Roles — Chapter 16
| Confusion |
Deciding clue |
Compare |
| Operating Model vs Org Chart |
Operating model defines how decision/accountability/work flows; org chart shows reporting structure. |
Battle Cards |
| Decentralized vs Network |
Decentralized has local autonomy without strong coordinating network; Network adds coordinated relationships/shared responsibility across decentralized units. |
Battle Cards |
| Centralized vs Hybrid |
Centralized concentrates authority/execution; Hybrid retains meaningful local accountability with central coordination. |
Battle Cards |
| Hybrid vs Federated |
Federated adds formal divisional/regional layers within coordinated enterprise governance; Hybrid is broader central/local combination. |
Battle Cards |
| Organizational Function vs Individual Role |
Function/service describes organizational capability; individual role describes a person's responsibility. |
Battle Cards |
Organizational Change — Chapter 17
| Confusion |
Deciding clue |
Compare |
| Bridges vs Kotter |
Bridges explains psychological transition; Kotter diagnoses/builds organizational transformation. |
Battle Cards |
| Kotter Errors vs Kotter Stages |
Errors diagnose failure; stages provide ordered transformation actions. Similar themes do not make them the same list. |
Battle Cards |
| Gleicher vs Rogers |
Gleicher asks whether change energy can overcome resistance; Rogers explains how adoption diffuses and individuals move toward adoption. |
Battle Cards |
| Rogers vs Communication Planning |
Rogers explains adoption/diffusion; Communication Planning designs message/channel/audience/feedback system. |
Battle Cards |
| Urgency vs Vision vs Coalition |
Urgency motivates action; Vision describes desired direction; Coalition provides leadership power/support to drive change. |
Battle Cards |
When the pair crosses chapters
Use the dedicated interleaving system after you understand each side individually:
- Confusion Book
- Battle Rounds
- Mixed Scenarios
- Cross-Domain Case Studies
← Master Topic Finder