Analyzing 6 Years of My Blogging Journey (2018–2024)

Table of contents

• 1 Introduction
• 2 Dataset Overview
• 3 Summary Statistics
• 4 Exploratory Data Analysis (EDA)
  • 4.1 Posts Over Time
    • 4.1.1 Interpretation
  • 4.2 Category-Wise Post Distribution
    • 4.2.1 Observation
  • 4.3 Views by Category
    • 4.3.1 Insight
  • 4.4 Word Count vs. Views
    • 4.4.1 Observation
• 5 Hypothesis Testing
  • 5.1 Hypothesis 1: Android-related posts attract higher views
  • 5.2 Hypothesis 2: Post length correlates with views
• 6 Key Findings
• 7 Reflection

1 Introduction

In August 2018, I published my first blog post on a small Hindi tech website called YouPays.in.
Over the next six years, I wrote 68 posts that reached over 2 million readers — covering topics like Android apps, gaming, and social media.

Now, as I transition into the world of Data Science, I wanted to revisit this blogging journey not with nostalgia, but with data.
What kind of topics resonated most with my readers? Did post length affect engagement? Did trends shift over time?

To answer these, I analyzed my complete Blogger dataset — a collection of 68 posts between 2018 and 2024 — using Python for exploratory and statistical analysis.

---

2 Dataset Overview

The dataset contains 13 columns, each representing key information about a post.

Code

import pandas as pd

df = pd.read_csv("D:/Website/notebooks/blogger_dataset.csv")
df.head(3)

	Title	Published_Date	Updated_Date	URL	Labels	Word_Count	Views	Comments	Content	Real_Publish_Date	Year_Real	Month_Real	Primary_Label
0	एंड्रॉयड के लिए बेस्ट मल्टीप्लेयर गेम 2023	2023-08-05T09:25:00+05:30	2024-01-04T22:28:22+05:30	http://nafeesdata.blogspot.com/2023/08/multipl...	Android, Gaming	2879	3770	0	<p>एंड्रॉयड मल्टीप्लेयर गेम्स समय के साथ बेहतर...	2023-08-01	2023	8	Android
1	Top 5 Best Running Game Kaun Se Hai 2023	2023-08-05T08:24:00+05:30	2023-08-06T02:06:21+05:30	http://nafeesdata.blogspot.com/2020/11/Best-ru...	Android & iOS, Gaming	3090	17500	0	<p dir="ltr" id="docs-internal-guid-1d3b8256-7...	2020-11-01	2020	11	Android & iOS
2	Top 5 Sniping Games For Android In Hindi	2023-08-05T08:23:00+05:30	2023-08-05T08:23:40+05:30	http://nafeesdata.blogspot.com/2019/09/top-5-s...	Android & iOS, Gaming	2134	15450	0	<div dir="ltr" style="text-align: left;" trbid...	2019-09-01	2019	9	Android & iOS

Column	Description
Title	Blog post title
Published_Date	Original Blogger publication timestamp
Updated_Date	Last update timestamp
URL	Blog post link
Labels	Original multiple category tags
Word_Count	Total words in the post
Views	Lifetime page views
Comments	Comment count
Content	Post HTML body
Real_Publish_Date	Parsed actual date
Year_Real	Year of publication
Month_Real	Month of publication
Primary_Label	Dominant category assigned

---

3 Summary Statistics

Code

df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 68 entries, 0 to 67
Data columns (total 13 columns):
 #   Column             Non-Null Count  Dtype 
---  ------             --------------  ----- 
 0   Title              68 non-null     object
 1   Published_Date     68 non-null     object
 2   Updated_Date       68 non-null     object
 3   URL                68 non-null     object
 4   Labels             68 non-null     object
 5   Word_Count         68 non-null     int64 
 6   Views              68 non-null     int64 
 7   Comments           68 non-null     int64 
 8   Content            68 non-null     object
 9   Real_Publish_Date  68 non-null     object
 10  Year_Real          68 non-null     int64 
 11  Month_Real         68 non-null     int64 
 12  Primary_Label      68 non-null     object
dtypes: int64(5), object(8)
memory usage: 7.0+ KB

Code

df.describe(include='all')

	Title	Published_Date	Updated_Date	URL	Labels	Word_Count	Views	Comments	Content	Real_Publish_Date	Year_Real	Month_Real	Primary_Label
count	68	68	68	68	68	68.000000	68.000000	68.000000	68	68	68.000000	68.000000	68
unique	68	63	68	68	25	NaN	NaN	NaN	68	28	NaN	NaN	7
top	एंड्रॉयड के लिए बेस्ट मल्टीप्लेयर गेम 2023	2023-07-07T07:53:00+05:30	2024-01-04T22:28:22+05:30	http://nafeesdata.blogspot.com/2023/08/multipl...	Internet	NaN	NaN	NaN	<p>एंड्रॉयड मल्टीप्लेयर गेम्स समय के साथ बेहतर...	2019-01-01	NaN	NaN	Android & iOS
freq	1	3	1	1	8	NaN	NaN	NaN	1	9	NaN	NaN	27
mean	NaN	NaN	NaN	NaN	NaN	2995.647059	23631.029412	0.235294	NaN	NaN	2019.544118	6.235294	NaN
std	NaN	NaN	NaN	NaN	NaN	1497.428002	60647.904004	0.601432	NaN	NaN	1.151687	3.574861	NaN
min	NaN	NaN	NaN	NaN	NaN	586.000000	1900.000000	0.000000	NaN	NaN	2018.000000	1.000000	NaN
25%	NaN	NaN	NaN	NaN	NaN	1873.000000	4417.500000	0.000000	NaN	NaN	2019.000000	3.000000	NaN
50%	NaN	NaN	NaN	NaN	NaN	2804.500000	6945.000000	0.000000	NaN	NaN	2019.000000	6.000000	NaN
75%	NaN	NaN	NaN	NaN	NaN	3509.000000	15745.000000	0.000000	NaN	NaN	2020.000000	9.000000	NaN
max	NaN	NaN	NaN	NaN	NaN	7121.000000	357650.000000	3.000000	NaN	NaN	2023.000000	12.000000	NaN

Let’s summarize key insights from the metadata.

Code

summary = {
    "Total Posts": len(df),
    "Total Views": f"{df['Views'].sum():,}",
    "Average Views per Post": round(df['Views'].mean(), 2),
    "Average Word Count": round(df['Word_Count'].mean(), 2),
    "Most Common Label": df['Primary_Label'].mode()[0],
    "Active Years": f"{df['Year_Real'].min()} - {df['Year_Real'].max()}"
}
pd.DataFrame(summary, index=["Summary"])

	Total Posts	Total Views	Average Views per Post	Average Word Count	Most Common Label	Active Years
Summary	68	1,606,910	23631.03	2995.65	Android & iOS	2018 - 2023

---

4 Exploratory Data Analysis (EDA)

4.1 Posts Over Time

Code

import matplotlib.pyplot as plt

yearly = df.groupby('Year_Real').size()
yearly.plot(kind='bar', figsize=(7,4), title='Posts Published per Year')
plt.xlabel('Year'); plt.ylabel('Number of Posts')
plt.grid(True)
plt.show()

4.1.1 Interpretation

Posts were most frequent in 2019–2020, aligning with peak engagement years. From 2021 onward, posting slowed, but those posts generally saw higher average views.

---

4.2 Category-Wise Post Distribution

Code

labels = df['Primary_Label'].value_counts().head(10)
labels.plot(kind='barh', figsize=(8,5), title='Top 10 Categories by Post Count')
plt.xlabel('No. of Posts'); plt.ylabel('Category')
plt.show()

4.2.1 Observation

The most frequent category is Android & iOS, followed by Blogging and Gaming. Posts about iOS-only topics or social media tutorials were less popular overall.

---

4.3 Views by Category

Code

views_by_label = df.groupby('Primary_Label')['Views'].mean().sort_values(ascending=False)
views_by_label.plot(kind='bar', figsize=(8,4), title='Average Views by Category', color='steelblue')
plt.ylabel('Average Views per Post')
plt.xticks(rotation=45)
plt.show()

4.3.1 Insight

Posts about Android games and “Earn Money” tutorials gained substantially higher average views. In contrast, iOS and Social Media posts consistently underperformed.

---

4.4 Word Count vs. Views

Code

import seaborn as sns

sns.scatterplot(x='Word_Count', y='Views', data=df)
plt.title('Word Count vs. Views')
plt.xlabel('Word Count'); plt.ylabel('Views')
plt.grid(True)
plt.show()

4.4.1 Observation

There’s no strong linear correlation between word count and views. It suggests topic popularity matters more than length — a key insight for content strategy.

---

5 Hypothesis Testing

We’ll now test a few hypotheses based on the observed patterns.

5.1 Hypothesis 1: Android-related posts attract higher views

• \(H_0\): Mean views of Android posts = Mean views of other posts
• \(H_1\): Mean views of Android posts > Mean views of other posts

Code

import scipy.stats as stats

android = df[df['Primary_Label'].str.contains('Android')]['Views']
others = df[~df['Primary_Label'].str.contains('Android')]['Views']

t_stat, p_val = stats.ttest_ind(android, others, equal_var=False)
t_stat, p_val

(0.07003686273462072, 0.944376799412921)

Code

if p_val < 0.05:
    print("✅ Reject H₀: Android posts received significantly more views.")
else:
    print("❌ Fail to reject H₀: No significant difference found.")

❌ Fail to reject H₀: No significant difference found.

---

5.2 Hypothesis 2: Post length correlates with views

• \(H_0\): No correlation between word count and views
• \(H_1\): There is a correlation

Code

corr = df['Word_Count'].corr(df['Views'])
corr

0.26046113392834486

Interpretation:

The correlation coefficient indicates how strongly (or weakly) the length of a post influences its popularity. In this case, the correlation is expected to be weak — confirming that content topic > post length.

---

6 Key Findings

• Android gaming and “Earn Money” tutorials consistently outperformed all other topics.
• iOS-related and social media tutorials had notably low engagement.
• Views increased significantly after 2019, likely due to improved SEO and audience growth.
• Post length was not a major factor — trending keywords and timing mattered more.

---

7 Reflection

Analyzing my own blog data taught me that content creation and data analysis share a common principle — pattern recognition. The same curiosity that drove me to write about technology now drives me to explore data.

This marks the beginning of nafeesdata.me — a space where storytelling meets analytics.

---