Information Storage and Retrieval: Data Marts, Analytical Pipelines and Dashboards

Lesson overview

Explain how analytical data is prepared and delivered through pipelines, data marts, dashboards, KPIs, quality checks and refresh schedules.

CourseInformation Storage and Retrieval

Topic strandAnalytics delivery

Assessment styleWorked scenario, applied task and digital exam practice

EvidenceDefinitions, representation, method, result and interpretation

Starter

Write down the user need, the data being stored or searched, and the decision the system has to support. Then predict which representation, index, query method or governance control will matter most in this lesson.

Learning objectives

Define data mart and explain its relationship to a warehouse.
Describe an analytical pipeline from source to dashboard.
Evaluate dashboard KPIs, refresh timing and data quality checks.

Learning outcomes

Students can define data mart and explain its relationship to a warehouse.
Students can describe an analytical pipeline from source to dashboard.
Students can evaluate dashboard KPIs, refresh timing and data quality checks.

Key vocabulary, acronyms and terminology

data mart: A subject-specific analytical store for a team or business area.
pipeline: A repeatable sequence that moves and transforms data.
dashboard: A visual display of selected metrics and status indicators.
KPI: Key Performance Indicator used to monitor progress or performance.
data quality: Fitness of data for its intended analytical use.
refresh schedule: How often analytical data is updated.

Detailed teaching notes

Core concept

Analytics is not complete when data is stored. Pipelines must transform, validate and deliver data to users in forms that support decisions.

Representation choice

A data mart might contain only admissions metrics for the recruitment team, derived from a central warehouse and tuned for their dashboard questions.

Method and reasoning

Define the decision, choose KPIs, trace required source data, build transformation steps, add quality checks and publish dashboards with clear refresh timing and ownership.

Risk and limitation

Dashboards can create false certainty. A bright chart based on stale, incomplete or poorly defined data may mislead faster than a spreadsheet.

Degree-level deep dive

Dashboards are the end of a pipeline

A dashboard is only as trustworthy as the source data, transformations, quality checks, ownership and refresh schedule behind it. Visual display does not make weak data stronger, so students must reason backwards from the chart to the data lineage.

KPIs must support decisions

A Key Performance Indicator should connect to action. A metric can be interesting without being a KPI if no one can interpret it, trust it or use it to make a decision. Good KPI design names the decision and the threshold for concern.

What excellent work shows

A strong answer defines the decision, maps each KPI to source data, names pipeline stages, and adds quality and refresh checks. It should include caveats where missing, stale or biased data could mislead users.

Concrete example to study

Student engagement pipeline

Sources: attendance records, virtual learning activity, assessment submissions
Transform: standardise student_id, calculate missed_submissions, calculate last_activity_days
Mart: StudentEngagementMart
Dashboard KPI: students with no activity for 14 days

Reasoning

The KPI is actionable because it identifies students who may need support.
The transformation must handle missing data and duplicate identifiers.
The dashboard should show refresh time and definitions so staff interpret it correctly.

Reveal takeaway

A dashboard is a teaching-quality example only when students can trace the metric back to source and decision.

Worked example

Scenario

A student-success team wants a dashboard of engagement risk.

Worked solution

Define KPIs such as missed submissions, low attendance and inactivity.
Extract source data from learning systems and assessment records.
Transform into a student-support data mart with agreed definitions.
Display dashboard metrics with refresh time, caveats and drill-through to evidence.

Reveal model result

The dashboard should support intervention, not surveillance by vague score. Data quality, definitions and ethics are part of the pipeline.

\text{Define the data, choose the representation, apply the method, interpret the result.}

Applied retrieval task

Plan a dashboard for course leaders.

Deliverables

Choose three KPIs and justify decisions they support.
Map each KPI to source data.
Add one quality check and one refresh rule.

Success checks

KPIs are actionable.
Refresh and quality are visible to users.

Common misconception

A dashboard is not automatically business intelligence. It becomes useful only when metrics are defined, trusted and connected to decisions.

Quick checks

1. In this lesson, why does data mart matter?

It is only a presentational label and has no effect on retrieval quality.A subject-specific analytical store for a team or business area.It replaces the need to understand user tasks or data representation.

2. Which answer best shows degree-level understanding of data marts, analytical pipelines and dashboards?

Define pipeline, apply it to the scenario, then explain the trade-off or risk.Give a final recommendation without showing how the data or query was represented.Name as many technologies as possible without explaining the scenario.

Digital exam practice

Example exam task

A business builds a dashboard showing customer churn risk, but different teams disagree about the numbers. In your answer, define the relevant objects or data structures, use course-specific vocabulary, show the method rather than only the result, and finish with a decision about the storage or retrieval system.

Notation and technical toolkit

source -> transform -> mart -> dashboardanalytical pipeline Use to show delivery flow.

KPImetric tied to decision Use only when the measure supports action.

refresh timelast update timestamp Use to judge currency of analytics.

What a good answer is expected to show

A strong answer for Data Marts, Analytical Pipelines and Dashboards the answer should discuss data mart scope, analytical pipeline steps, KPI definitions, dashboard refresh and data quality checks. It should connect the formal or technical representation to the user's information need instead of listing terms without using them.

How to solve it

Identify the decision the dashboard supports.
Define KPIs and their source data clearly.
Describe the pipeline into a subject-specific data mart.
Add quality checks, ownership and refresh information to make the dashboard trustworthy.

Model answer

Reveal model answer

The churn dashboard needs agreed KPI definitions before it can be trusted. A customer analytics data mart should receive data through a documented pipeline from subscriptions, usage logs and support records. Transformation steps must standardise customer identifiers and churn definitions, then validate missing values and duplicate records. The dashboard should show refresh time and ownership so teams know whether a discrepancy is caused by stale data, different definitions or genuine business change.

Practise next

Design one KPI that is actionable and one that is merely interesting.
Explain why dashboard refresh timing matters for operational decisions.

Self-marking criteria

Defines the scenario objects, data forms or system components before solving.
Uses the lesson vocabulary accurately and in context.
Shows a clear method with enough working for a marker to follow.
Connects the result back to retrieval, storage, analytics or governance.
States a limitation, trade-off or quality risk rather than presenting the answer as absolute.

Extension

Compare a team-specific data mart with a central warehouse for one organisation.

Study route

Save a short worked answer from this lesson using this pattern: define the need or data, choose the representation, show the method, state the result and interpret the implication for the system.

Next lesson: NoSQL Models for Retrieval and Scale.