Data Analytics for Business Decision Making Training Course

Duration 10 days 17 modules · 65 contact hours
Level Intermediate Certificate on completion
Who attends Business Analysts, Data Analysts and others
Fee from USD 2,200 KSh 190,000 for Kenyan organisations
Where 52 cities + online See dates by city
Next start 9 Nov 2026 Course code SD-DA-064

Before you join: Participants should be comfortable using spreadsheets for sorting, filtering, basic formulas and charts.

Dates & prices

Two start dates a month in every city, and live online on weekdays or weekends. Fees are per participant; classroom fees include tuition, materials, lunch and refreshments, and the certificate.

Suggest a date or city

Showing the next six months. Later dates, other cities and private deliveries for your team are available on request.

Course overview

Business decisions often depend on fragmented operational data, inconsistent metrics and reports that describe activity without explaining what requires action. Managers need evidence they can trust when choosing where to invest, which customers to retain, how to improve service levels or whether a proposed intervention is working. Data analysts must turn a business question into a measurable analysis, test the quality of available information and communicate a recommendation with appropriate confidence and limits.

This course teaches the practical analytical methods used to support business decisions. Participants define decision questions, map data sources, design KPIs, extract and transform data with SQL and Excel, assess data quality, conduct exploratory analysis and create clear visual reports. They apply descriptive statistics, hypothesis tests, forecasting, segmentation, contribution analysis and root-cause methods. The course also addresses governance, privacy, bias, access controls and the difference between correlation, causation and managerial judgement.

Teaching combines short instructor briefings with realistic business cases from banking, public services, retail, operations and customer-facing functions. Participants work with datasets from the first week, producing queries, cleaned data tables, KPI definitions, analyses, dashboards and decision briefs. The final capstone requires each participant or team to prepare and present an evidence-based recommendation supported by an auditable dataset, visual report, assumptions and implementation measures. Participants receive a certificate of completion after the 10-day instructor-led programme.

The course suits professionals who already work with reports, spreadsheets or operational data and need a disciplined method for producing analysis that decision-makers can use.

What you will be able to do

  • Formulate decision-focused analytical questions, hypotheses and measurable success criteria
  • Assess data sources using grain, lineage, ownership, quality and access criteria
  • Write SQL queries using joins, aggregations, CASE expressions and window functions
  • Prepare auditable datasets using Excel, Power Query and documented transformation rules
  • Design KPI trees, metric definitions and data quality scorecards for management reporting
  • Apply descriptive statistics, hypothesis testing, forecasting and segmentation methods
  • Build Power BI dashboards and decision briefs that communicate evidence clearly
  • Present a complete analytical recommendation with assumptions, risks and implementation measures

Benefits of attending

For you

  • Produces decision briefs supported by traceable evidence and stated assumptions
  • Writes SQL queries instead of relying solely on prebuilt reports
  • Builds KPI definitions that prevent inconsistent management reporting
  • Tests business claims using statistical evidence and forecast accuracy measures
  • Presents data findings confidently to managers and non-technical stakeholders

For your organisation

  • Improves consistency of KPI definitions, calculations and report interpretation
  • Reduces manual reporting effort through reusable queries and transformation steps
  • Identifies revenue, cost, service and customer drivers with structured analysis
  • Strengthens data quality controls before findings reach decision-makers
  • Creates clearer accountability for recommendations, actions and realised benefits

Course modules

17 modules over 10 days

Participants can frame a business decision as a focused analytical question with accountable owners and measurable criteria.

  • Decision questions translated into measurable outcomes, constraints and accountable owners
  • SMART decision objectives distinguish strategic, tactical and operational choices
  • Analytic lifecycle links questions, evidence, insight, action and review
  • Decision rights mapped using RACI for data-led recommendations
  • Hypotheses framed with population, comparison, expected effect and decision threshold
  • Problem statements separate symptoms, causes, assumptions and information needs

Participants can assess whether available data is suitable for a defined business question.

  • Data sources classified by owner, granularity, latency, coverage and access
  • Relational tables interpreted through entities, keys, attributes and relationships
  • Grain defined before joining transactions, snapshots and reference data
  • Data dictionaries document field meaning, units, valid values and refresh schedules
  • Structured, semi-structured and unstructured data matched to analytical uses
  • Source-system lineage traced from operational capture to reporting output

Participants can design metrics that link strategic outcomes to operational drivers.

  • KPI trees connect outcomes, drivers, measures, targets and owners
  • Leading and lagging indicators distinguished using causal timing
  • Metric definitions specify numerator, denominator, filters, aggregation and exclusions
  • Balanced Scorecard perspectives align financial, customer, process and capability measures
  • North-star metrics tested against Goodhart's law and unintended behaviour
  • Targets benchmarked against baselines, variation and planned interventions

Participants can write SQL queries that retrieve, combine and summarise business data accurately.

  • SELECT, WHERE, ORDER BY and LIMIT retrieve decision-relevant records
  • INNER, LEFT and FULL joins preserve intended population definitions
  • GROUP BY aggregations calculate counts, sums, averages and rates
  • CASE expressions create transparent business categories and exception flags
  • Common table expressions structure multi-step queries for audit and reuse
  • Window functions compare periods, ranks and customer-level behaviour

Participants can prepare and reconcile business data using repeatable spreadsheet methods.

  • Excel Tables, XLOOKUP and SUMIFS reconcile operational datasets
  • PivotTables profile revenue, volume, mix and service performance
  • Power Query records repeatable import, cleaning and combination steps
  • Formula auditing exposes broken references, inconsistent logic and hidden errors
  • Named ranges and structured references improve model readability and maintenance
  • Version control conventions protect approved files and calculation assumptions

Participants can identify, correct and document data quality issues before analysis.

  • Missingness classified as MCAR, MAR or MNAR before treatment
  • Duplicate records detected through keys, fuzzy matching and survivorship rules
  • Outliers investigated using IQR, z-scores, boxplots and operational context
  • Validation rules test ranges, formats, referential integrity and permitted values
  • Standardisation converts dates, currencies, codes and measurement units consistently
  • Cleaning logs record source issues, transformations, owners and unresolved risks

Participants can apply governance and ethical controls when using business and personal data.

  • Data minimisation limits fields to stated analytical purpose
  • Personal data handling follows GDPR principles of lawfulness and purpose limitation
  • Role-based access separates data viewing, editing, approval and publication permissions
  • Bias checks examine sampling, measurement, proxy variables and disparate outcomes
  • Retention schedules define archival, deletion and legal-hold requirements
  • Data quality scorecards report accuracy, completeness, timeliness, consistency and validity

Participants can investigate patterns, anomalies and drivers before drawing conclusions.

  • Distribution plots reveal skew, spread, multimodality and data-entry anomalies
  • Descriptive statistics summarise centre, dispersion, percentiles and frequency
  • Segmentation compares customer, product, channel, region and cohort patterns
  • Cross-tabulations expose associations between categorical variables and outcomes
  • Time-series decomposition separates trend, seasonality, cycles and irregular variation
  • Drill-down paths move from aggregate exceptions to contributing records

Participants can create visual reports that support accurate and timely management decisions.

  • Chart selection matches comparison, composition, distribution, relationship and time questions
  • Bar, line, scatter and box plots avoid misleading encodings
  • Axis scales, baselines and annotations support accurate visual interpretation
  • Pre-attentive attributes direct attention to material differences and exceptions
  • Accessible design applies colour contrast, labels and non-colour cues
  • Dashboard wireframes prioritise decisions, metrics, filters and action prompts

Participants can assess uncertainty and test whether observed differences support action.

  • Sampling frames define populations, inclusion criteria and selection bias
  • Confidence intervals express uncertainty around estimated rates and averages
  • Hypothesis tests compare groups using null, alternative, alpha and p-values
  • Effect sizes distinguish material differences from statistically detectable differences
  • Correlation coefficients quantify association without asserting causation
  • A/B test design sets randomisation, guardrail metrics, duration and stopping rules

Participants can produce and evaluate forecasts for planning and resource decisions.

  • Baseline forecasts use naïve, seasonal naïve and moving-average methods
  • Exponential smoothing estimates level, trend and seasonal components
  • Forecast accuracy assessed with MAE, RMSE, MAPE and bias
  • Driver-based forecasts link demand assumptions to operational capacity plans
  • Scenario analysis compares base, upside and downside business conditions
  • Prediction intervals communicate forecast uncertainty and planning ranges

Participants can identify commercially important segments and performance drivers.

  • RFM analysis ranks customers by recency, frequency and monetary value
  • Cohort analysis tracks retention, repeat purchase and revenue behaviour
  • Pareto analysis identifies high-value customers, products and defect sources
  • Contribution analysis separates volume, price, mix and cost drivers
  • Customer lifetime value assumptions distinguish acquisition, retention and margin effects
  • Segment profiles combine behavioural data, needs, value and service requirements

Participants can investigate performance gaps and distinguish plausible causes from unsupported claims.

  • 5 Whys traces observed performance gaps to testable causal factors
  • Fishbone diagrams organise people, process, technology, policy and measurement causes
  • Process maps identify handoffs, queues, rework loops and control failures
  • Pareto charts focus investigation on high-frequency or high-impact causes
  • Confounding variables tested before assigning responsibility for observed outcomes
  • Root-cause evidence distinguishes correlation, mechanism and managerial judgement

Participants can build governed Power BI reports from structured analytical data.

  • Star schemas separate fact tables, dimensions, measures and calculated columns
  • DAX measures calculate filter-aware totals, ratios and period comparisons
  • Power BI relationships control cross-filtering and prevent ambiguous model paths
  • Report pages use slicers, drill-through and tooltips for guided analysis
  • Row-level security restricts report views by authorised user attributes
  • Publication checklists verify refresh dates, filters, definitions and data warnings

Participants can communicate analytical findings as concise, decision-ready recommendations.

  • Pyramid Principle structures recommendation, evidence, implications and requested decision
  • Executive summaries state decision, confidence, financial exposure and next action
  • Narrative flow connects context, finding, cause, options and recommendation
  • Sensitivity analysis shows decisions affected by uncertain assumptions
  • Decision memos distinguish facts, inferences, risks and management judgement
  • Stakeholder objections anticipated with evidence, trade-offs and escalation paths

Participants can translate analytical findings into owned actions and measurable follow-through.

  • Analytic recommendations converted into actions, owners, milestones and measures
  • Benefits cases quantify value, costs, confidence ranges and dependencies
  • Experiment backlogs prioritise assumptions by impact, uncertainty and test effort
  • Control plans assign monitoring thresholds, response actions and review cadence
  • Adoption measures track use, compliance, exceptions and realised benefits
  • Post-implementation reviews compare expected outcomes with actual performance

Participants can build, review and present a complete decision analysis containing an auditable dataset, evidence-based recommendation and implementation plan.

  • Complete decision brief combines business question, data preparation, analysis and recommendation
  • Source inventory documents provenance, quality checks, definitions and access constraints
  • SQL extracts and spreadsheet transformations create an auditable analytical dataset
  • Dashboard presents KPIs, segments, trends, drivers and decision options
  • Statistical or forecasting evidence quantifies uncertainty, assumptions and limitations
  • Peer review tests logic, visual clarity, traceability and action feasibility
  • Executive presentation defends recommendation, implementation plan, risks and success measures

Exercises and examples are adapted to each cohort's sectors and roles.

What you take back to work

  1. Decision question and hypothesis template
  2. KPI tree and metric definition sheet
  3. Data quality assessment checklist
  4. SQL query and transformation log
  5. Power BI dashboard wireframe
  6. Evidence-based decision brief template

Who should attend

  • Business Analysts — need defensible analysis for operational and strategic recommendations
  • Data Analysts — need stronger business framing, statistical reasoning and stakeholder communication
  • Finance Analysts — need repeatable methods for performance, variance and forecast analysis
  • Operations Managers — need to diagnose service, process and capacity performance using data
  • Product Managers — need evidence for customer, adoption and feature-prioritisation decisions
  • Programme and Policy Officers — need to assess outcomes, risks and intervention performance

Skills and tools covered

SQL data analysisKPI designData quality assessmentBusiness forecastingDashboard developmentDecision communication

Tools and standards

Microsoft Excel, SQL, Microsoft Power BI, CRISP-DM

How the course runs

Training methodology

The course uses short instructor briefings followed by case-based analysis, group discussions, guided tool use and structured review. Participants work with realistic datasets to define business questions, inspect source data, write SQL, prepare spreadsheet models and develop visual reports. Simulations require participants to interpret uncertain evidence, respond to stakeholder challenges and choose between competing options. Individual application planning connects course methods to a current reporting, service, customer or operational decision. Trainers adapt cases, datasets and discussion emphasis to the cohort's sector and roles.

Requirements and prerequisites

Participants should be comfortable using spreadsheets for sorting, filtering, basic formulas and charts. They should understand common business measures such as revenue, cost, volume, customer numbers or service levels, and should have experience reading management reports or working with operational data. Prior exposure to SQL, Power BI, statistics or programming is helpful but not required. The course introduces the SQL, Excel and statistical techniques used during delivery. Participants do not need a mathematics degree, data science background or prior experience building dashboards.

A typical training day

08:30 – 10:30First session
10:30 – 10:45Refreshment break
10:45 – 12:30Second session
12:30 – 13:30Lunch and networking
13:30 – 15:00Third session
15:00 – 15:15Refreshment break
15:15 – 16:30Workshop and daily review

Live online deliveries follow the same structure in East Africa Time, with shorter screen blocks and longer breaks.

What the fee includes

  • Instruction by a practitioner facilitator
  • Full course workbook and materials
  • Exercise files, templates and case studies
  • Refreshments and lunch on classroom deliveries
  • Certificate of completion
  • Post-course application plan

Frequently asked questions

You should be comfortable using spreadsheets and interpreting business reports. No prior SQL, Power BI, statistics or programming experience is required, although familiarity with any of these will help.

Bring a laptop capable of running spreadsheet and browser-based applications, plus a current business reporting or decision problem if you have one. Training datasets and course materials are provided.

It is designed for analysts, managers and specialists who use operational, financial, customer or programme data to support decisions. It is particularly suitable for professionals moving beyond routine reporting into structured analysis.

This course focuses on business questions, data quality, SQL, descriptive analysis, forecasting, dashboards and decision communication. It does not focus on advanced programming, model deployment or machine learning algorithm development.

Participants can use the methods to improve an existing report, investigate a performance issue, define stronger KPIs or prepare a business case. The templates and capstone structure provide a repeatable approach for future analysis.

Training runs from 08:30 to 16:30 each day, with about 6.5 contact hours. The first week establishes data, SQL, quality, KPI and visualisation skills; the second week applies statistical, forecasting, dashboard and recommendation methods before the capstone.

Questions about this course?

Ask about dates, fees, group bookings or running it for your team. Our training team replies by email.

Ask on WhatsApp info@skillsetdevelopment.com

Related courses in Data Analytics