5 Python Skills You Need to Get Hired as a Data Analyst in 2026 (With Code Examples)

Table of Contents

The demand for data analysts in Nigeria is growing faster than the supply of qualified candidates. Fintechs, banks, telecoms companies, consulting firms, and e-commerce platforms are all actively hiring. Almost every one of them wants candidates who can work with Python.

The problem most beginners face is not a lack of motivation. It is not knowing exactly which Python skills to focus on. There are hundreds of tutorials online, but very few of them tell you what employers are actually testing for when they interview data analyst candidates.

This article closes that gap.

Based on what Nigerian employers are looking for in 2026, here are the 5 Python skills that will move you from "still learning" to "job-ready." Each one comes with a real code example you can run in Jupyter Notebook or Google Colab today.

New to Python? Start with our guide on 5 Steps to Perform Exploratory Data Analysis in Python before working through this article.

Why Python Is the Most In-Demand Skill for Data Analysts in 2026

Before we get into the list, it helps to understand why Python matters so much right now.

Python is increasingly required beyond entry-level roles, particularly at tech companies, fintechs, and consulting firms in Nigeria. And according to career research from Profolio, the core skills Nigerian employers look for include Python with Pandas for data manipulation, which is increasingly expected, especially in fintech and consulting.

The good news is that you do not need to know everything. You need to know the right things, practised well enough that you can use them under pressure in a technical interview or assessment.

Skill 1: Data Cleaning with Pandas

Why Employers Care About This

Data cleaning is the most time-consuming task in any real data analyst role. Data rarely comes perfect, so data cleaning skills will always be in demand. Cleaning and transforming datasets, removing duplicates, handling missing values, and standardising formats are often the most time-consuming but essential parts of the job.

If you cannot clean data, you cannot analyse it. It is that simple.

When companies run technical assessments for data analyst roles, they almost always include a dirty dataset and ask you to clean it. Candidates who freeze at this stage rarely move forward in the process.

What You Need to Know

The four most important data-cleaning operations are the following

• Finding and handling missing values
• Removing duplicate rows
• Fixing incorrect data types
• Renaming columns to standard formats

Code Example

import pandas as pd

# Simulate a messy sales dataset (common in Nigerian business data)
data = {
    "order_id":    [101, 102, 103, 102, 104, 105, 106],
    "customer":    ["Ada", "Emeka", None, "Emeka", "Fatima", "Chidi", "Ngozi"],
    "amount":      [15000, 32000, 8000, 32000, None, 21000, 18500],
    "order_date":  ["2026-01-05", "2026-01-06", "2026-01-07",
                    "2026-01-06", "2026-01-08", "2026-01-09", "2026-01-10"],
    "region":      ["lagos", "abuja", "KANO", "abuja", "Lagos", "PHC", "ibadan"]
}

df = pd.DataFrame(data)
print("Before Cleaning:")
print(df)

# Step 1: Remove duplicate rows
df = df.drop_duplicates()
print(f"\nRows after removing duplicates: {len(df)}")

# Step 2: Handle missing values
print("\nMissing values before fix:")
print(df.isnull().sum())

df["customer"] = df["customer"].fillna("Unknown")
df["amount"] = df["amount"].fillna(df["amount"].median())

# Step 3: Fix inconsistent text formatting in the region column
df["region"] = df["region"].str.title()

# Step 4: Convert order_date to a proper date type
df["order_date"] = pd.to_datetime(df["order_date"])

print("\nAfter Cleaning:")
print(df)
print(f"\nData types:\n{df.dtypes}")

Notice the region column had "lagos", "KANO", and "Lagos" all meaning the same thing. The .str.title() method standardises all of them to "Lagos", "Kano", and so on. This kind of inconsistency is extremely common in real business data.

How to Practise This Skill

Download a real dataset from Kaggle and practise cleaning it from scratch. Real data is always messier than anything you create yourself, which is exactly why practising on it is so valuable.

Skill 2: Exploratory Data Analysis (EDA)

Why Employers Care About This

EDA is the process of examining a dataset to understand what it contains before drawing conclusions or building models. Analysts who skip this step regularly produce insights that are wrong, misleading, or based on data they did not fully understand.

Analysts who structure their work clearly create trust and accelerate alignment across teams. EDA is how you build that structure from the very beginning of any project.

In technical interviews, you will often be given a dataset and asked to "tell me what you see." EDA is how you answer that question with confidence.

What You Need to Know

The core EDA operations every data analyst must be able to perform are:

• Checking the shape and data types of a dataset
• Identifying missing values and outliers
• Summarising numerical columns with statistics
• Counting and understanding categorical columns

Code Example

import pandas as pd
import numpy as np

# Load a sample employee dataset
data = {
    "employee_id": range(1, 11),
    "department":  ["Sales", "IT", "HR", "Sales", "Finance",
                    "IT", "HR", "Sales", "Finance", "IT"],
    "salary":      [120000, 250000, 110000, 135000, 200000,
                    230000, 115000, 128000, 195000, 270000],
    "years_exp":   [2, 5, 1, 3, 4, 6, 2, 3, 4, 7],
    "performance": [3.2, 4.5, 3.8, 4.1, 4.0, 4.7, 3.5, 3.9, 4.2, 4.8]
}

df = pd.DataFrame(data)

# Step 1: Basic inspection
print("Shape:", df.shape)
print("\nData Types:")
print(df.dtypes)

# Step 2: Statistical summary
print("\nStatistical Summary:")
print(df.describe().round(1))

# Step 3: Department breakdown
print("\nEmployees per Department:")
print(df["department"].value_counts())

# Step 4: Average salary by department
print("\nAverage Salary by Department (₦):")
avg_salary = df.groupby("department")["salary"].mean().sort_values(ascending=False)
print(avg_salary.apply(lambda x: f"₦{x:,.0f}"))

# Step 5: Is there a relationship between experience and salary?
correlation = df["years_exp"].corr(df["salary"])
print(f"\nCorrelation between experience and salary: {correlation:.2f}")

A correlation of 0.80 or above between experience and salary tells you that more experienced employees tend to earn significantly more. A number close to zero tells you there is no strong pattern. This kind of finding is what employers pay analysts to surface.

Want to go deeper on EDA? Read our full guide: 5 Steps to Perform Exploratory Data Analysis in Python.

Skill 3: Data Visualization

Why Employers Care About This

Learning Python for advanced analysis and automation significantly boosts your chances of getting hired as a data analyst. But technical analysis alone is not enough. You also need to communicate what you found to people who are not data analysts themselves.

That is what data visualization does. A chart communicates in seconds what a table of numbers takes minutes to understand. Analysts who can present their findings clearly are far more valuable to a business than those who can only produce the numbers.

In assessments and interviews, you will regularly be asked to visualise a finding and explain it to a non-technical audience.

What You Need to Know

Focus on mastering four chart types first:

• Line charts for trends over time
• Bar charts for comparing categories
• Scatter plots for relationships between two variables
• Heatmaps for showing correlation across multiple columns

Code Example

import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

# Monthly revenue data for a Nigerian e-commerce business
data = {
    "Month":    ["Jan", "Feb", "Mar", "Apr", "May", "Jun"],
    "Lagos":    [850000, 920000, 880000, 1050000, 1200000, 1150000],
    "Abuja":    [520000, 580000, 610000, 640000, 700000, 720000],
    "Kano":     [310000, 340000, 320000, 380000, 400000, 430000]
}

df = pd.DataFrame(data).set_index("Month")

# Chart 1: Line chart — revenue trend by city
plt.figure(figsize=(10, 5))
for city in ["Lagos", "Abuja", "Kano"]:
    plt.plot(df.index, df[city] / 1000, marker="o", label=city, linewidth=2)

plt.title("Monthly Revenue by City — Jan to Jun 2026", fontsize=13)
plt.xlabel("Month")
plt.ylabel("Revenue (₦ Thousands)")
plt.legend()
plt.tight_layout()
plt.show()

# Chart 2: Bar chart — total revenue comparison
total_revenue = df.sum()
plt.figure(figsize=(7, 5))
bars = plt.bar(total_revenue.index, total_revenue / 1_000_000,
               color=["steelblue", "coral", "mediumseagreen"])
plt.title("Total Revenue by City (Jan to Jun 2026)", fontsize=13)
plt.ylabel("Total Revenue (₦ Millions)")
for bar in bars:
    plt.text(bar.get_x() + bar.get_width() / 2,
             bar.get_height() + 0.02,
             f"₦{bar.get_height():.1f}M",
             ha="center", fontsize=10)
plt.tight_layout()
plt.show()

Want to master all the major Python charting libraries? Read our guide: 5 Best Python Libraries for Data Visualization in 2026.

🎯 Stop Guessing and Start Practising

If you want to practise all three skills above with structured, real-world exercises designed specifically for data analysts, check out the 50 Python Exercises for Data Analysts on Selar. It gives you 50 hands-on coding challenges that mirror what employers actually test during technical assessments, complete with solutions you can check your work against.

Skill 4: Statistical Analysis with NumPy and Pandas

Why Employers Care About This

Data analysis without statistics is just counting. The real value an analyst provides is being able to take numbers and explain what they mean for the business. Strong statistical analysis knowledge separates analysts who report numbers from those who generate insights. Skills like regression, correlation, hypothesis testing, and A/B testing help you interpret trends accurately.

You do not need a degree in statistics to do this well. You need to understand a small set of core concepts and be able to apply them in Python.

What You Need to Know

For entry-level and mid-level roles, focus on these:

• Measures of central tendency: mean, median, and mode
• Measures of spread: standard deviation and variance
• Correlation between variables
• Identifying outliers using the interquartile range (IQR)

Code Example

import pandas as pd
import numpy as np

# Delivery time data (in hours) for an e-commerce platform
delivery_times = np.array([24, 18, 36, 22, 120, 19, 25, 21,
                            48, 23, 17, 26, 24, 22, 200, 20])

# Central tendency
mean_time   = np.mean(delivery_times)
median_time = np.median(delivery_times)

print(f"Mean delivery time:   {mean_time:.1f} hours")
print(f"Median delivery time: {median_time:.1f} hours")

# The mean is much higher than the median — a sign of outliers
# Identifying outliers using the IQR method
Q1  = np.percentile(delivery_times, 25)
Q3  = np.percentile(delivery_times, 75)
IQR = Q3 - Q1

lower_bound = Q1 - 1.5 * IQR
upper_bound = Q3 + 1.5 * IQR

outliers = delivery_times[(delivery_times < lower_bound) |
                           (delivery_times > upper_bound)]

print(f"\nQ1: {Q1}, Q3: {Q3}, IQR: {IQR}")
print(f"Outlier boundary: below {lower_bound:.1f} or above {upper_bound:.1f} hours")
print(f"Outliers detected: {outliers}")

# Clean data without outliers
clean_times = delivery_times[(delivery_times >= lower_bound) &
                              (delivery_times <= upper_bound)]
print(f"\nMean without outliers: {np.mean(clean_times):.1f} hours")
print(f"Standard deviation:    {np.std(clean_times):.1f} hours")

In this example, 120 hours and 200 hours are outliers that make the average look much worse than reality. Removing them and recalculating gives the business a far more accurate picture of its actual delivery performance. This is the kind of finding that makes an analyst indispensable.

Want to strengthen your NumPy foundation? Read our guide: 5 NumPy Functions Every Data Analyst Should Know.

Learn at Your Own Pace with Free Video Lessons

If you prefer learning through video, the JacobIsah Programming Hub YouTube channel covers Python for data analysts with practical tutorials designed for Nigerian learners. Subscribe and turn on notifications so you never miss a new lesson.

Skill 5: Building Real-World Portfolio Projects

Why Employers Care About This

This is the skill that actually gets you hired. Not because it is the most technical, but because it is the most visible proof that you can do real work.

Most data analyst job seekers in Nigeria take 2 to 6 months to land their first role after building foundational skills. Candidates with strong portfolios and targeted applications tend to find positions significantly faster than those relying on qualifications alone. A portfolio project is a complete piece of analysis that you built yourself, documented clearly, and published publicly so anyone can review it. It answers the question every interviewer is really asking: "Can you actually do this job?" Many analysts land their first role without a single line of professional experience on their resume as their projects did the talking instead.

What Makes a Strong Portfolio Project

A good portfolio project for a data analyst role includes these elements:

• A real or realistic dataset (not a toy dataset from a textbook)
• Data cleaning applied before any analysis
• At least two to three visualisations that tell a clear story
• A written summary of what you found and what it means for the business
• Code hosted on GitHub so employers can review it

Code Example: A Mini Portfolio Project

This is a complete, short analysis you can expand into a full portfolio project:

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns

# Dataset: Supermarket sales across Nigerian cities
data = {
    "City":      ["Lagos", "Abuja", "Kano", "PHC", "Ibadan"] * 6,
    "Month":     ["Jan"] * 5 + ["Feb"] * 5 + ["Mar"] * 5 +
                 ["Apr"] * 5 + ["May"] * 5 + ["Jun"] * 5,
    "Category":  (["Food", "Electronics", "Clothing", "Pharmacy", "Household"] * 6),
    "Revenue":   np.random.randint(200000, 1500000, 30),
    "Customers": np.random.randint(100, 800, 30)
}

df = pd.DataFrame(data)

# Finding 1: Which city generates the most revenue?
city_revenue = df.groupby("City")["Revenue"].sum().sort_values(ascending=False)
print("Revenue by City (₦):")
print(city_revenue.apply(lambda x: f"₦{x:,}"))

# Finding 2: Which product category performs best overall?
cat_revenue = df.groupby("Category")["Revenue"].sum().sort_values(ascending=False)
print("\nRevenue by Category (₦):")
print(cat_revenue.apply(lambda x: f"₦{x:,}"))

# Finding 3: Average revenue per customer by city
df["Revenue_per_Customer"] = (df["Revenue"] / df["Customers"]).round(0)
avg_rpc = df.groupby("City")["Revenue_per_Customer"].mean().sort_values(ascending=False)
print("\nAverage Revenue per Customer by City (₦):")
print(avg_rpc.apply(lambda x: f"₦{x:,.0f}"))

# Visualise: Top city revenue comparison
plt.figure(figsize=(8, 5))
sns.barplot(x=city_revenue.index, y=city_revenue.values / 1_000_000,
            palette="Blues_d")
plt.title("Total Revenue by City — Jan to Jun 2026 (₦ Millions)")
plt.xlabel("City")
plt.ylabel("Revenue (₦ Millions)")
plt.tight_layout()
plt.show()

Once you have this analysis working, write a short paragraph explaining what the data reveals and what you would recommend to the business. That combination of code plus written insight is what a portfolio project looks like in practice.

Ready to build 5 complete projects with Python and SQL? Download the free guide: 5 Real-World Data Projects You Can Do with Python and SQL — no payment required. It shows you step by step how to build projects that impress employers.

How to Know When You Are Job-Ready

Many beginners make the mistake of thinking they need to know everything before applying. You do not. Here is a simple checklist to tell you when you are ready:

Technical readiness:

• You can clean a messy dataset without looking up every step
• You can perform a basic EDA and summarise what you found in plain language
• You can produce at least three types of charts and explain what each one shows
• You understand mean, median, standard deviation, and correlation

Portfolio readiness:

• You have at least two to three projects on GitHub with a clear README for each
• Each project uses a real or realistic dataset and includes both code and written findings
• You can explain what you did, why you did it, and what the results mean

Application readiness:

• Your LinkedIn profile lists your Python and Pandas skills with project links
• Your CV mentions the tools you use and the projects you have completed
• You have practised answering the question: "Walk me through a data project you have worked on"

If you can tick most of these boxes, you are ready to start applying.

Take the Fast Track to Your First Data Analyst Role

If you want a structured, step-by-step path that takes you from Python basics all the way to building your first real data project and preparing for job interviews, the Python for Data Analysts course on Selar covers everything in this article and more. It is designed specifically for Nigerian learners who want practical, job-ready skills, not just theory.

And if you are still figuring out which tools and platforms to use for writing your Python code, the Student Coding Platforms Guide breaks down Google Colab, Replit, Programiz, and VS Code so you can pick the right environment from the start.

Conclusion

Getting hired as a data analyst in Nigeria in 2026 is very achievable for anyone willing to build real, practical skills. The job market is strong, the salaries are growing, and employers are actively looking for candidates who can prove their abilities with real work.

The five Python skills covered in this article are the exact ones that show up in technical assessments and interviews across Nigerian companies:

1. Data Cleaning with Pandas: the foundation of all real-world analysis
2. Exploratory Data Analysis: the ability to understand any dataset quickly
3. Data Visualisation: the skill that communicates your findings to decision-makers
4. Statistical Analysis: the thinking that separates a good analyst from a great one
5. Portfolio Project Building the proof that makes employers confident enough to hire you

You do not need to master all five at once. Start with data cleaning and EDA, because they appear in almost every technical interview. Add visualization next. Then sharpen your statistics. And build your portfolio as you go, because every project you complete adds to the evidence that you can do this job.

The market is waiting. Your next step is to practise.

Related Articles on This Blog

• 5 Steps to Perform Exploratory Data Analysis in Python
• 5 NumPy Functions Every Data Analyst Should Know
• 5 Best Python Libraries for Data Visualization in 2026
• 5 Python Projects for Beginner Data Analysts to Build Their Portfolio

References

1. 13 Data Analyst Skills That Get You Hired in 2026 — Great Learning — mygreatlearning.com
2. 12 Data Analyst Skills That Will Get You Hired in 2026 — Dataquest — dataquest.io
3. How to Get a Data Analyst Job in Nigeria — Profolio — profolio.ng
4. 50 Companies Hiring Data Analysts in Nigeria 2026 — Profolio — profolio.ng
5. 7 In-Demand Data Analyst Skills to Get You Hired in 2026 — Coursera — coursera.org
6. 13 Data Analyst Skills That Will Get You Hired in 2026 — Learning Saint — learningsaint.com
7. Pandas Official Documentation — pandas.pydata.org
8. NumPy Official Documentation — numpy.org

---

Published on JacobIsah Programming Hub | enemzy.blogspot.com