Marketing Analytics Study Guide

Data cleaning

The process of fixing data issues; it does not include collecting new data.

Marketing analytics

Measuring, managing, and analyzing marketing performance to improve ROI.

Three components of marketing analytics

Descriptive, Predictive, Prescriptive analytics.

Steps to prepare data

Data collection, cleaning, integration.

Data integration

Combining data from multiple sources.

Predictive analytics question

What will happen?

Structured data example

Spreadsheet or database table.

Least valuable analytics

Descriptive analytics.

Diagnostic analytics

Explains why something happened.

IDE

Coding environment.

Integer

A whole number.

Programming

Solving problems using code/algorithms.

Popular languages

Python and R.

Numeric variable

Measured in numbers.

R & Python

Both used for analysis and visualization.

NoSQL

Not Only SQL.

NoSQL types

Key-value, Document, Wide-column, Graph.

Non-relational DB use

Large unstructured data.

Primary key

Unique identifier for each record.

Relational DB

Stores data in tables.

Relational advantage

Complex joins.

Referential integrity

Relationships remain consistent.

SQL get ages

SELECT Age FROM Customers;.

SQL search

LIKE '%red colored pencils%'.

Find names containing ena

LIKE '%ena%'.

Matching values from 2 tables

INNER JOIN.

LEFT JOIN

All from A plus matches from B.

ORDER BY

Sort selected data.

AVG by category

AVG(price) GROUP BY CategoryID.

Inbound marketing

Attract, engage, delight customers.

First-party data

Your customers.

Second-party data

Another company’s first-party data.

GDPR

Data privacy regulation.

Cookie

Small file tracking behavior.

Nielsen data

Third-party data.

Sentiment analysis

Uses NLP.

Social listening

Monitoring online conversations.

Content analysis

Examining digital content.

Brand versatility

Multiple usage scenarios reputation.

Conversion metric

Conversions.

Recommend metric

NPS.

CTR

Ratio of clicks to views.

Marketing metrics

Evaluate performance.

Hardest analytics step

Define the problem.

A/B testing design

Experimental design.

Participant assignment

Random assignment.

Local maxima

Best nearby solution.

After-only advantage

Simple, avoids testing effects.

Natural experiment

Calorie menu study.

Within-subjects design

Participants see all conditions.

Type I error

False positive.

T-test

Compare means of two groups.

ANOVA limitation

Not only two groups.

Type II error

False negative.

Post-hoc tests

Tukey HSD and Contrasts.

AI definition

Machines act like humans.

Machine learning

Systems learn automatically.

Applied AI

Real-world AI use.

Deep learning

Neural networks.

Supervised vs unsupervised

Labeled vs unlabeled data.

AI categories

Narrow AI and General AI.

Deep learning modeled after

Human brain.

Heat map

Shows page attention.

GIS

Geographic Information Systems.

Radar chart

Compare multiple variables.

Box plot

Quartiles and variability.

Gantt chart

Project schedule bar chart.

Infographic

Story visualization.

Dashboards

KPI visualizations in Tableau/Power BI.