Video summary

How I Would Learn to be a Data Analyst

Main summary

Key takeaways

Educational

Main ideas / lessons conveyed

  • The video presents a data-backed, systematic approach to learning the skills needed to become a data analyst.
  • It frames the work of a data analyst using the data analytics lifecycle and explains how data engineers and data scientists fit upstream/downstream.
  • It ranks the most in-demand skills for data analyst roles using analysis from job postings, then groups skills into learning “buckets.”
  • It gives a 3-step method for mastering those skills effectively and building employable proof (projects/portfolio).
  • It provides estimated learning timelines for different tools.

Career context: what a data analyst does (via lifecycle)

A 4-step data analytics lifecycle:

  1. Collect data
  2. Clean data (notably takes “most of your time”)
  3. Analyze data to answer questions for a boss/stakeholder
  4. Report results (communicate findings)

Three practitioner roles in this ecosystem

  • Data Analyst
    • Uses historical data to answer business questions
  • Data Engineer
    • Upstream; builds and maintains data infrastructure so others can access data
  • Data Scientist
    • Downstream; focuses on predictive models using statistics and machine learning

Skills approach: data-backed ranking + categorization

Data used for demand/ranking

  • Based on 80,000+ job postings
  • Each job averages ~4 skills requested
  • Median worldwide salary mentioned: ~$100,000
  • The video uses this to determine which skills are most worth focusing on.

Top skills (ranked by demand frequency)

The video identifies six top skills (after which demand “drops off” and is said to be less worth focusing on):

  1. SQL — appears in about 1 in 2 job postings
  2. Excel (spreadsheet software) — appears in about 1 in 3
  3. Python — tied with Excel in postings, but the speaker argues it’s overhyped
  4. Power BI — business intelligence tool (ranked among BI tools; Power BI is “in the lead”)
  5. Tableau — business intelligence tool
  6. R — another programming language (less demanded than the others)

Four skill buckets (what they are, conceptually)

  • Bucket 1: SQL (language for communicating with databases)
  • Bucket 2: Excel (spreadsheet software)
  • Bucket 3: BI tools — Power BI and Tableau (dashboards/visual reporting)
  • Bucket 4: Programming — Python and R

Tier system used for recommendations

  • S tier = highest recommendation (required)
  • A tier = recommended after S
  • B tier = later/less priority
  • C tier = lowest recommendation

Detailed breakdown of each learning bucket (including what to focus on + tier)

1) SQL (Tier: S — required)

What SQL does

  • Used to create SQL queries that are sent to a database
  • The database returns results as tables

Three major command types

  • DQL (Data Query Language): read data; no modifications
  • DML (Data Manipulation Language): update/delete records
  • DDL (Data Definition Language): create/drop entire databases (careful!)

Learning progression (recommended)

  • Beginners
    • Focus on DQL basics
    • Understand DML
  • After that
    • Advanced concepts such as CTEs and window functions

Rationale for ranking

  • SQL connects to databases, and databases are where most data lives (most foundational).

2) Excel (Tier: S — required)

What Excel is used for

  • Spreadsheet-based analysis and sharing; broadly used by non-data analysts too.

Historical/positioning point

  • Excel is portrayed as the dominant spreadsheet tool (introduced 1985; includes features like Power Query).

Learning progression (recommended)

  • Beginners
    • Formulas, functions, charts, tables
  • Next
    • Pivot tables
    • Power Query (connects to various data sources, including databases)

Rationale for ranking

  • Excel is described as universally used, and stakeholders/co-workers can more easily verify results there.
  • “Likely only” SQL + Excel are enough to land a first role (speaker’s view).

3) BI Tools: Power BI and Tableau (Tier: A for either; learn one)

Purpose / why they exist

  • Designed to solve dashboarding/reporting problems that spreadsheets (Excel) can struggle with.
  • BI tools support:
    • Drag-and-drop dashboards
    • Better stakeholder interaction (e.g., slicers)
    • Centralized sharing via services (Power BI service mentioned)

Which BI tool to choose

  • Learn only one (Power BI or Tableau), since dashboard concepts transfer.

Learning progression (recommended)

  • Start with
    • The range of visualizations each tool can produce
  • Then progress to
    • More advanced concepts, including Power Query
    • Tableau equivalent noted: Tableau Prep

Rationale for tier

  • BI tools are powerful/visual, but speaker advises not starting with them.
  • Recommended order: database/spreadsheet tools first, then BI.

4) Python and R (Tier: Python B, R C; learn one)

High-level distinction

  • Python
    • Multi-purpose: analytics, machine learning, even web development
  • R
    • More specialized: statistics, data visualization, academic research

Speaker’s recommendation

  • Prefer Python over R (but overall says programming is not top priority early).
  • Emphasizes only learning one: “anything you can do in Python you can do in R.”

Why Python is ranked lower

  • Biggest practical issue on the job (as described):
    • Co-workers may not know Python, so:
      • They can’t easily double-check results
      • They may be less likely to use Python outputs

Learning placement

  • Python: B tier (after BI tools)
  • R: C tier (even lower)

Methodology for mastering skills (3-step approach + how courses apply it)

The video’s central teaching/mastery approach has three steps:

  1. Pick a skill

    • Focus first on the top-demanded skills from the rankings.

  2. Build a project

    • Implement data analytics concepts using the chosen tools.
    • Learning should emphasize analytics concepts, not just tool mechanics.

  3. Share the work

    • Post to places like LinkedIn and GitHub to make the project visible.

How the speaker’s courses follow this method

  • Courses for SQL, Excel, Power BI, and Python use the same 3-step approach:
    • Part 1: learn alongside the instructor on the basics using real-world datasets
    • Part 2: create a portfolio project applying what was learned
    • Outcome: produce a shareable project demonstrating skills

Course offering and support (as described)

  • Courses are free
  • Optional “extra perks” for supporters include:
    • Interview-level practice problems
    • Community to work/interact with others
    • Certificate of completion

Learning timelines (time-to-skill estimates)

Based on student completion data (speaker’s interpretation):

  • Power BI: fastest, typically around 2 months
  • Excel and SQL: about 3.5 months each (“neck and neck”)
  • Python: longest, upward of 6 months

Broader transition estimate:

  • Often 6–12 months to transition into data analytics
  • Can be as fast as 3 months if full-time-focused

Speakers / sources featured

  • Luke (the primary speaker/creator of datanerd.tech; also shares personal background and course details)
  • Data from:
    • Interviews with data analysts and data managers (speaker’s “countless interviews” claim)
    • Job posting dataset aggregated via the speaker’s app:
      • datanerd.tech
      • 4 million jobs (platform scale mentioned)
    • 80,000+ job postings analyzed for skill demand (as stated in the video)

Original video