Course Outline
Programme outline for analytical foundations
Module 1 Foundations of Applied Data Science and Business Analytics
Purpose: Build a shared understanding of analytics and define useful customer and operational questions.
- Data science, statistics, analytics and machine learning in plain business language.
- Descriptive, diagnostic, predictive and prescriptive analytics and the questions each addresses.
- The analytics lifecycle from business understanding to evaluation and action.
- Translating service challenges into questions, outcomes, scope and success measures.
- Data types, observation units, time periods and basic data dictionaries.
- Responsible interpretation, appropriate data access and human accountability.
Practical activity Frame a customer-service challenge, identify the decision to support and draft an analysis plan.
Participant output A one-page problem definition with outcome measures, data needs and assumptions.
Module 2 Preparing Exploring and Interpreting Customer and Operational Data
Purpose: Develop a repeatable approach to preparing data and drawing defensible descriptive insights.
- Understanding dataset structure, identifiers, categories and date fields.
- Checking missing values, duplicates, inconsistent labels and unusual observations.
- Combining related records while checking matching keys and record counts.
- Counts, rates, proportions, averages, medians, percentiles and measures of spread.
- Comparing service categories, channels, periods and customer groups using appropriate denominators.
- Exploring distributions, trends, seasonality and exceptions through clear charts.
- Introductory sampling, uncertainty and the limitations of small or incomplete datasets.
- Correlation versus causation and common interpretation errors.
Practical activity Prepare a service dataset, reconcile key totals and investigate variations in demand, resolution time and customer feedback.
Participant output A preparation log and an insight sheet containing findings, evidence and limitations.
Programme outline for customer insight and prediction
Module 3 Understanding Customer Behaviour and Performance Drivers
Purpose Identify meaningful customer and service patterns and develop explanations to investigate.
- Business-rule segmentation by behaviour, contact frequency, service type or channel.
- Comparing customer journeys, repeat contacts, complaint categories and satisfaction patterns.
- Separating volume effects from differences in service outcomes.
- Exploring relationships between workload, waiting time, resolution time and repeat contact.
- Generating root-cause hypotheses and distinguishing plausible explanations from proven causes.
- Interpreting segments carefully and avoiding unsupported generalisations.
Practical activity Compare service segments and identify one priority group or process requiring further investigation.
Participant output A segment profile, supporting comparisons and a prioritised set of hypotheses.
Module 4 Predictive Analytics and Forecasting for Business Outcomes
Purpose Apply selected predictive methods and assess whether their outputs are useful for a business decision.
- Choosing between forecasting, regression and classification based on the question and target.
- Defining the prediction time and using only information available at that point.
- Preparing inputs and targets; avoiding data leakage and overfitting.
- Training and evaluation data, including chronological splits for time-dependent problems.
- Demand forecasting with simple baselines, moving averages and an introductory smoothing method.
- Guided regression for a numerical outcome and simple classification for an event such as repeat contact.
- Comparing forecast or regression errors using mean absolute error and root mean squared error.
- Interpreting a confusion matrix, precision and recall in relation to operational costs.
- Explaining uncertainty, model limitations, changing patterns and the need for ongoing review.
Practical activity Create a demand forecast and follow a guided predictive-model exercise; compare results with a simple baseline.
Participant output A forecast or model summary documenting evaluation results, assumptions and appropriate business use.
Programme outline for decisions and business narratives
Module 5 Data Driven Decisions and Continuous Improvement
Purpose Translate findings into feasible improvements and define how to evaluate their effects.
- Connecting evidence to operational performance and customer-experience objectives.
- Prioritising improvement opportunities by expected value, feasibility, effort and risk.
- Using forecasts to inform workload and resource discussions without assuming a forecast determines the best action.
- Comparing scenarios and testing sensitivity to key assumptions.
- Designing an intervention with ownership, milestones and measurable success criteria.
- Defining baseline performance, outcome measures and guardrails against unintended effects.
- Using a controlled pilot where feasible and interpreting before-and-after comparisons cautiously.
- Applying a plan–do–check–act cycle and monitoring whether gains are sustained.
Practical activity Compare two improvement options and develop a pilot plan with measurable business and customer outcomes.
Participant output A prioritised recommendation, scenario comparison and improvement measurement plan.
Module 6 Communicating Insights and Developing Business Recommendations
Purpose Present analysis as a clear business narrative that supports a specific decision.
- Identifying the audience, decision and level of detail required.
- Structuring a narrative around the business question, evidence, implications and action.
- Selecting charts, titles and annotations that communicate the main finding accurately.
- Explaining uncertainty and separating observations, predictions and assumptions.
- Writing recommendations that specify action, rationale, owner and expected benefit.
- Preparing a concise executive briefing and responding to stakeholder challenge.
- Presenting the integrated case and revising recommendations following feedback.
Practical activity Deliver a short decision briefing using the analysis developed throughout the programme.
Participant output A three-to-five-slide presentation and a one-page recommendation memo.
Requirements
Participants should be comfortable with spreadsheets, tables, basic charts, percentages and routine business KPIs. Prior programming or machine-learning experience is not required.
The intended outcome is guided practitioner capability in selected methods. Independent production-model development requires further study and experience
Testimonials (2)
Power Bi Dashboards
Rolien Struwig - Vodacom
Course - Applied Data Science and Advanced Analytics
Learning how to use data to make evidence-based decisions. I gained a better understanding of the data science process, including data preparation, analysis, visualization, and predictive modelling.