Business Analytics

1. KPIs, Dashboards & Reporting Basics

Business analytics is about using data to make better decisions, not just collecting numbers. A Key Performance Indicator (KPI) is a specific, measurable figure that tells you whether things are going well.

• Good KPIs are specific: "Number of new clients this month" is a KPI; "how the business is doing" is not.
• Leading vs lagging indicators: Leading indicators predict future results (e.g. number of sales calls made); lagging indicators report what already happened (e.g. last month's revenue).
• A dashboard brings several KPIs together on one screen so decision-makers can see the full picture at a glance.
• Choose KPIs that connect directly to a real decision someone will make — tracking a number nobody acts on wastes effort.

Practical Skill: For a real or fictional small business or NGO program, define 3 KPIs that would genuinely help its manager make decisions, explaining why each one matters.

=== KPI DEFINITION WORKSHEET ===

Business/Program: [e.g. "Community Health Outreach Program"]

KPI 1: ________________________________________________
  Why it matters: ______________________________________
  How often measured: [ ] Daily [ ] Weekly [ ] Monthly

KPI 2: ________________________________________________
  Why it matters: ______________________________________
  How often measured: [ ] Daily [ ] Weekly [ ] Monthly

KPI 3: ________________________________________________
  Why it matters: ______________________________________
  How often measured: [ ] Daily [ ] Weekly [ ] Monthly

Example (Health Outreach):
  KPI: "Number of households visited per week"
  Why: Shows whether outreach targets are being met in time
  Frequency: Weekly

2. Turning Questions Into Measurable Metrics

Managers ask vague questions like "are we doing well?" — a business analyst's job is to turn that into a specific, measurable question that data can actually answer.

• Vague question: "Are customers happy?" → Measurable metric: "What percentage of customers rate their experience 4 or 5 out of 5?"
• Vague question: "Is the program working?" → Measurable metric: "What percentage of participants completed the program and passed the final assessment?"
• Always define exactly how a metric will be measured and where the data will come from before starting to collect it.
• Avoid vanity metrics (numbers that look good but don't drive decisions) — a metric should always connect to an action someone will take.

Practical Skill: Take 3 vague business questions and rewrite each as a specific, measurable metric with a clear data source.

=== VAGUE QUESTION -> MEASURABLE METRIC ===

Vague: "Is our social media doing well?"
Metric: "What is our average engagement rate (likes+comments
         / followers) per post this month?"
Data source: Meta Business Suite Insights

Vague: "Are sales growing?"
Metric: "What is month-over-month revenue growth, in SSP,
         for the last 3 months?"
Data source: Sales record spreadsheet

Now try your own:
Vague: _______________________________________________
Metric: ______________________________________________
Data source: _________________________________________

3. Data Quality Checks & Simple Analysis

Analysis built on bad data produces confident, wrong answers. Before drawing conclusions, check that the underlying data is trustworthy.

• Completeness: Are there missing values that would skew the results?
• Consistency: Is the same thing recorded the same way every time (e.g. "Juba" vs "juba" vs "JUBA")?
• Plausibility: Do the numbers make sense? A reported "500 sales in one day" for a small shop deserves a second look.
• Simple analysis techniques: comparing periods (this month vs last month), comparing groups (region A vs region B), and looking at trends over time.

Practical Skill: Review a small sample dataset for completeness, consistency, and plausibility issues before analyzing it, listing every issue found.

=== DATA QUALITY CHECK ===

Dataset: [e.g. "March Sales Records"]

[ ] Any missing values in key columns? Where?
[ ] Any inconsistent spelling/formatting of the same category?
[ ] Any numbers that look implausibly high or low?
[ ] Are dates in a consistent, correct format?
[ ] Are duplicate entries present?

=== SIMPLE COMPARISON ANALYSIS ===
  This month vs last month:  ______ vs ______  (% change: ____)
  Region A vs Region B:      ______ vs ______
  Trend over last 3 months:  [ ] Rising [ ] Falling [ ] Flat

4. Presenting Insights Clearly (for Leadership)

Good analysis that's poorly presented gets ignored. Leaders are busy — insights need to be clear, specific, and tied to a recommended action.

• Lead with the headline: State the key finding first ("Sales dropped 15% in March"), not the methodology.
• One chart, one message: Each chart should support a single clear point, not require lengthy explanation.
• So what? Always answer "so what should we do about this?" — insights without a recommendation are just trivia.
• Know your audience: A board presentation needs different detail than a working team meeting.

Practical Skill: Take a finding from Module 3's analysis and write a 3-sentence summary for a busy manager: the finding, why it matters, and a recommended next step.

=== ONE-PAGE INSIGHT SUMMARY ===

FINDING:  [One sentence — the key number or trend]
WHY IT MATTERS: [One sentence — the business impact]
RECOMMENDATION: [One sentence — what to do next]

Example:
FINDING: New client sign-ups fell 20% in February compared
  to January.
WHY IT MATTERS: At this rate, we will miss our quarterly
  growth target by a wide margin.
RECOMMENDATION: Investigate whether the drop follows a
  specific marketing channel pausing, and reallocate budget
  toward whichever channel is still performing.

5. Case Examples: NGO Programs, Sales & Inventory

Business analytics looks slightly different depending on the setting. Seeing a few real-style examples helps connect the concepts to actual work.

• NGO program example: Tracking beneficiaries reached per month against a target, and investigating gaps by region or activity type.
• Sales example: Tracking revenue and units sold by product line to identify what to promote or discontinue.
• Inventory example: Tracking stock levels against sales rate to avoid running out of fast-moving items or overstocking slow ones.
• In every case, the same core process applies: define the question → find/check the data → analyze → present a clear recommendation.

Practical Skill: Choose one case (NGO, sales, or inventory) relevant to your own context and walk through the full process end to end using a small sample dataset.

=== CASE: INVENTORY ANALYSIS FOR A SMALL SHOP ===

Question: "Which products should we restock this week?"

Sample Data:
  Product        Stock Left   Avg Weekly Sales   Weeks of Stock Left
  Exercise Books      40             25               1.6
  Pens               200             30               6.7
  USB Drives           5              8               0.6

Analysis:
  USB Drives will run out in under a week at current sales rate
  Exercise Books need restocking soon (under 2 weeks left)
  Pens have healthy stock (over 6 weeks left)

Recommendation:
  Restock USB Drives urgently, restock Exercise Books this
  week, hold off on ordering more Pens.

Quick Quiz — Business Analytics Basics