Data Science, Data & Business Analysis: How Insights Drive Desicions

TechCrushHQ avatar
TechCrush.pro
@TechCrushHQSep 29, 2026

The Spaces brought together host Abisola with practitioners Joshua Gideon (Business Analyst), James (Data Scientist; Ubonthaya Jin Sonida Mula), Opayemi (Data Analyst), and Mizo Kwadi (Growth & Insights, Interswitch) to close a mentorship series focused on data science, data analysis, and business analysis. Joshua outlined clear distinctions: business analysts define and solve business problems through elicitation, stakeholder management, process mapping, and documentation, while data analysts tell evidence-based stories of past and present using data, and data scientists extend analytics with predictive modeling. He detailed a BA’s day, emphasizing communication and cross-functional collaboration, and discussed real-world constraints like budgets and power-interest dynamics. James shared a non-CS journey into data science, stressing communities, visible learning, focused skill selection, and problem-first learning; he clarified DA vs DS skills and workflows. Mizo described building confidence via “stacking evidence” through projects and free gigs, networking, and practical delivery. Opayemi highlighted common beginner pitfalls—tool-only focus, weak domain context, and insufficient practice—and recommended domain selection, logic-first learning, and consistent projects. The session closed with Q&A on BA–DA boundaries, timelines for learning (3–6 months for solid foundations), and program logistics for upcoming scholarships.

Data Science, Data Analysis, and Business Analysis: Roles, Skills, Journeys, and Practical Advice — Webinar Recap

Session context and focus

  • This session closed a multi-day mentorship webinar series covering multiple tech tracks. The final topic focused on data science, data analysis, and business analysis — their overlaps, differences, real-world execution, and how to choose a path.
  • Host: Abisola. Co-host support: Faith.
  • Speakers:
    • Joshua Gideon — Business Analyst (≈5+ years), transitioned from data analysis.
    • Mizo Kwadi — Data/Insights professional at Interswitch, works on transaction/business data to drive growth and decisions.
    • Ubonthaya (James) Sonida Mula — Data Scientist at Tiering (US-based), works on ML/data analysis projects and training datasets for AI models.
    • Opeyemi — Data Analyst, active in research/business projects and beginner mentorship.

Core themes and insights

1) How the roles differ and overlap

  • Business Analysis (BA):
    • Primary aim: define, frame, and solve business problems; determine requirements; guide decision-making and change.
    • Key activities: stakeholder identification and management; elicitation (structured communication to gather needs); process mapping; documentation (BRD, FRD, NFR); negotiating trade-offs; aligning teams.
    • Outputs: problem definitions, business requirements, process improvements, solution recommendations, acceptance criteria, and decision records.
  • Data Analysis (DA):
    • Primary aim: describe and explain what has happened or is happening in the business using data.
    • Key activities: data cleaning, exploratory analysis, KPI tracking, trend/pattern identification, visual storytelling, and answering concrete business questions.
    • Outputs: dashboards, reports, analytical narratives (insights) that inform decisions.
  • Data Science (DS):
    • Primary aim: build predictive/ML models and data products; move from insights to prediction/automation.
    • Key activities: feature engineering, model training/evaluation, scenario simulation (e.g., forecasting, churn, fraud detection), and model-driven insights.
    • Outputs: predictive models, scoring systems, experiments, and model-enabled insights.
  • Relationship:
    • DA answers “what happened/what is happening?”; DS answers “what is likely to happen?”; BA answers “what should the business do and how?”
    • In smaller companies, one person might wear multiple hats; in larger firms, roles are more distinct.

2) Day-in-the-life and scope of a Business Analyst (Joshua)

  • BA work is highly collaborative; unlike a DA who can often work solo with a dataset, a BA’s effectiveness hinges on stakeholder interaction and elicitation across business, technical, and customer groups.
  • Example (warehouse/stock-keeping scenario):
    • Problem: delayed customer fulfillment due to poor warehouse organization.
    • Approach: beyond metrics, BAs observe processes, interview warehouse staff, and map current vs. future states; identify root causes (e.g., no standardized placement, insufficient staffing); propose solutions (re-organization schemes, staffing changes, process controls).
    • Tools/techniques: process mapping, stakeholder engagement, requirements documentation; complement with basic analysis/visualization (Excel, Power BI) for evidence.
  • Core BA skills emphasized:
    • Communication & elicitation: extracting accurate information across hierarchies, including customers.
    • Critical thinking & problem solving: defining problems precisely and designing practical solutions.
    • Documentation rigor: BRD, FRD, NFR, change logs; traceability to decisions and approvals.
    • Stakeholder management: identifying influence/interest levels and managing alignment.

3) Implementing solutions: governance, constraints, and conflict resolution (Joshua)

  • Expect pushback: recommendations (e.g., “hire more warehouse staff”) may hit budget or scope constraints from high-power stakeholders (e.g., CEO).
  • Power-interest dynamics: decisions ultimately rest with high-power stakeholders; BAs must document recommendations, negotiate trade-offs, and secure sign-offs.
  • Conflict mediation: BA often brokers compromises (e.g., marketing vs. engineering viewpoints) and forges common ground centered on business outcomes.
  • Success factors: leadership buy-in and stakeholder education can accelerate adoption; absent that, BAs focus on clear documentation, rationale, and pragmatic middle-ground solutions.

4) Data science vs. data analytics in practice (James)

  • Data Analytics:
    • Tools: SQL, Excel, Power BI; Python (pandas) for analysis and visualization; answering targeted questions (e.g., “Which products had the highest sales/profit last quarter?”).
  • Data Science:
    • Builds on DA: data wrangling (pandas), visualization (Matplotlib/Plotly/Seaborn) or R equivalents (dplyr/ggplot2) plus model development (forecasting, churn/fraud models).
    • Practical examples: demand forecasting; detection of anomalous/“fraud-like” transactions; churn prediction.
    • View shared: “A data scientist is a data analyst already” — DS includes DA competencies plus modeling; less emphasis on BI tools if using code-based visualization/workflows.
  • Program note (host): data science training can be beginner-friendly; DA background is a plus but not mandatory.

5) Real career journeys and strategies

  • Joshua (BA):
    • Path: Health sector student → WHO data entry (Excel) → self-taught DA → professional DA roles → pivot to BA (training + practice) → full-time BA leveraging DA skills.
    • Learning: start with DA if possible; BA requires breadth (process, stakeholder, documentation) plus enough data fluency to ground recommendations.
  • James (DS):
    • Path: Agriculture and plant breeding degrees → stats coursework → self-taught R (client gigs) → Python + DS courses/certifications → unpaid ML internship (Africa) used as experience → DS role at Tiering.
    • Strategies:
      • Communities (WhatsApp/Slack/Discord, local events) for leads and learning.
      • Share publicly (LinkedIn/Status) to attract opportunities.
      • Choose tools by problem; avoid language-chasing; build a clear roadmap.
      • Take unpaid/low-paid gigs early to gain real experience; keep learning (tech is a marathon).
  • Mizo (Data/Insights at Interswitch):
    • Path: Early exposure to web dev, uninterested; rediscovered Python during lockdown; intensive self-study; sponsored DS program; long unpaid internship with deep exposure; teaching; active outreach (LinkedIn); first paid role; subsequent international move; current cross-functional insights role.
    • Mantras:
      • Build, don’t just consume: end-to-end projects teach more than passive courses.
      • Confidence is “stacking evidence” via repeated delivery and preparation, not affirmations.
      • Audacity: approach SMEs for data, start with your own (e.g., bank statements), ask for mentorship/data access, and don’t self-disqualify.
  • Opeyemi (DA):
    • Path: Surveyed tech roles; considered product management; leaned into data due to opportunity access; training in tools; applied analysis in policy/public affairs, research, and business projects; strong community involvement and teaching.
    • Emphasis: communities accelerate learning and relevance; practice consistently.

6) Common mistakes and how to avoid them

  • Trying to learn “everything” (James, Opeyemi):
    • Symptom: hopping between languages/tools (C, JS, Java, Julia, Python, R) without depth.
    • Remedy: define a real problem to solve and select a minimal, relevant toolset (e.g., Python + pandas + SQL) to execute; follow a roadmap.
  • Tools over thinking (Opeyemi):
    • Mistake: focusing on “how to click” vs. “why/when to use” a method.
    • Remedy: learn the logic behind tools; know the questions to ask before analysis; develop analytical thinking, not just software skills.
  • Neglecting domain context (Opeyemi):
    • Mistake: ignoring the industry/business domain reduces relevance.
    • Remedy: pick a domain (e.g., finance, logistics, policy), study its KPIs/processes, and tailor projects accordingly.
  • Avoiding projects or delaying practice (all speakers):
    • Mistake: staying in course mode without applying.
    • Remedy: ship projects early; do free/low-paid work for real users to stack evidence.
  • Building in silence (James):
    • Mistake: no visibility; fewer inbound opportunities.
    • Remedy: share your learning, projects, and results publicly; attend events; build relationships.

7) Tools, methods, and artifacts mentioned

  • DA/DS tooling: SQL; Excel; Power BI; Python (pandas, Matplotlib/Plotly/Seaborn); R (dplyr, ggplot2) — for wrangling, analysis, and visualization.
  • BA practices: stakeholder mapping; elicitation; process mapping; documentation (BRD, FRD, NFR); power–interest grid; decision records and sign-offs.

Q&A highlights

  • Can a BA do a DA’s work (and vice versa)? Is DA still worth studying separately?
    • Mizo’s scenario: “Transaction volume dropped 20%.”
      • DA asks: when did it start, which customers/channels changed, what patterns exist, how does this period compare to prior periods? Output: analytical diagnosis (e.g., 3 customers left, channel X declined).
      • BA asks: why this matters commercially, why customers left (data won’t fully answer this), and which interventions should be prioritized to recover growth.
    • Conclusion: Roles are distinct but complementary. In small companies one person may cover both; in larger firms, separation is common. DA is valuable and remains a distinct, employable skill.
    • Opeyemi added: applicability depends on company structure, your background, and the business problem.
  • Beginner timeline to a solid foundation (DA focus):
    • Opeyemi: depends on dedication and learning speed. With consistent study and practice, 3–6 months can build a solid foundation; faster or slower is possible based on available time and depth.

Career and confidence — practical playbook

  • Start where you are: analyze your own data (e.g., bank statements) to practice.
  • Build end-to-end projects: real, messy data and delivery to real users accelerates skill and confidence.
  • Leverage communities: join online/offline groups; attend events; ask for feedback and opportunities.
  • Share publicly: LinkedIn/other platforms — progress posts can attract gigs.
  • Take experiential opportunities: unpaid/low-paid internships/gigs can be stepping stones; document outcomes.
  • Choose with purpose: pick your domain and problem, then match the minimal toolset.
  • Prepare relentlessly: confidence comes from “stacking evidence” of delivery, not from pep talks.

Personal impact of a tech/data career (speaker reflections)

  • Opeyemi: opens continuous learning and exploration; compels you to keep growing.
  • Joshua: transformative professionally and personally (from introvert to frequent stakeholder engagement); a defining career path.
  • Mizo: expanded analytical mindset, cross-industry mobility (e.g., construction to fintech), financial uplift, and access to high-caliber peers/opportunities.

Program notes and next steps (from host)

  • This webinar concludes the mentorship series for this cohort.
  • Upcoming: award emails to confirm the next Tech Africa Scholars; course-change window opens (watch deadlines and instructions); onboarding will include more details.
  • Additional webinars will continue beyond the mentorship series timeline.

Key takeaways

  • BA, DA, and DS are complementary: DA describes and explains; DS predicts; BA prescribes business actions and steers change.
  • Communication, elicitation, stakeholder management, and documentation are BA cornerstones; analytical rigor and storytelling are DA/DS cornerstones.
  • Don’t chase tools; define problems, pick a domain, and practice deliberately.
  • Confidence and opportunity come from shipping real work, engaging communities, and consistent preparation.
  • Early career momentum often involves unpaid or low-paid experiential roles — treat them as investments in skill and evidence.