Web Metrics & Analytics: Common metrics: Hits, Page views, visits, unique page views, Bounce,

Bounce rate & its improvement, Average time on site, Real time report, traf c source report, custom
campaigns, content report, Google analytics; Key Performance Indicator: Need, characteristics,
perspective and uses. Graphs and Matrices- Basic measures for individuals and networks. Random
graphs & network evolution, Social Context: Af liation & Identity Web analytics tools: A/B testing,
online surveys, Web crawling and Indexing. Natural Language Processing Techniques for Micro-
text Analysis. Explain each topics in detail in easy english language

ChatGPT said:

Updated saved memory

Here is a detailed explanation of each topic in easy English to help you understand better:

1. Web Metrics & Analytics (Common Metrics)

These are measurements used to understand how people use a website. Let’s go through them one
by one:

• Hits:

◦ A hit is a request made to the server. Every time someone loads a webpage
(including images, CSS, or JavaScript), it's a hit.

◦ Example: One page with 5 images = 6 hits (1 for the page, 5 for images).

• Page Views:

◦ The total number of times a speci c page is loaded by users.

◦ Example: If 1 user opens the same page 5 times, it counts as 5 page views.

• Visits (Sessions):

◦ A visit is when a user comes to the website and interacts with it for some time.

◦ It includes multiple page views.

◦ Ends after: 30 minutes of inactivity.

• Unique Page Views:

◦ Counts how many different users viewed a page, even if they viewed it many times
in one session.

◦ Example: 1 user views the same page 5 times = 1 unique page view.

• Bounce:

◦ A bounce happens when a user visits only one page and leaves without any further
interaction.

• Bounce Rate:
fi
fi
fi
 ◦ Percentage of single-page visits (bounces) out of total visits.

◦ Formula:
Bounce Rate
=
Single-page visits
Total visits

×
100

\text{Bounce Rate} = \frac{\text{Single-page visits}}{\text{Total visits}} \times 100

Bounce Rate=Total visitsSingle-page visits ×100

◦ Improvement Tips:

▪ Use engaging content.

▪ Add internal links.

▪ Improve website speed.

▪ Make your website mobile-friendly.

• Average Time on Site:

◦ The average time users spend on your website.

◦ Shows how engaging your content is.

• Real-Time Report:

◦ Shows what’s happening on your website right now.

◦ How many users are active, which pages they’re visiting, etc.

• Traf c Source Report:

◦ Tells you where your visitors are coming from:

▪ Direct (typed URL)

▪ Organic Search (Google)

▪ Social Media

▪ Email

▪ Paid ads

• Custom Campaigns:
fi
 ◦ You can track the success of marketing campaigns by adding UTM codes to URLs.

◦ Helps understand which ad or email got the most traf c.

• Content Report:

◦ Shows which content/pages are most popular.

◦ Helps you know what your audience likes.

2. Google Analytics

A free tool by Google that helps track and report website traf c and user behavior.

Features:

• Tracks users, sessions, bounce rate, traf c sources, etc.

• Provides dashboards and real-time data.

• Supports goal tracking and e-commerce analysis.

3. Key Performance Indicators (KPIs)

KPIs are the most important metrics that help you measure the success of your business or
website.

• Need:

◦ To check if your marketing or website goals are being met.

◦ Helps improve decision-making.

• Characteristics:

◦ Clear and simple.

◦ Relevant to business goals.

◦ Measurable.

◦ Time-bound.

• Perspective:

◦ Marketing KPI: Conversion rate, Cost per click.

◦ Sales KPI: Revenue, No. of sales.

◦ Customer KPI: Satisfaction score, Retention rate.

• Uses:
fi
fi
fi
 ◦ Measure performance.

◦ Identify issues.

◦ Plan improvements.

4. Graphs and Matrices

Used to analyze relationships and connections between people or web pages.

• Basic Measures for Individuals:

◦ Degree: Number of direct connections.

◦ Closeness: How easily someone can reach others.

◦ Betweenness: How often a person connects others.

• Basic Measures for Networks:

◦ Density: How many connections exist compared to the maximum possible.

◦ Centralization: Whether a few nodes are highly connected.

5. Random Graphs & Network Evolution

• Random Graphs:

◦ Networks formed by randomly connecting nodes.

◦ Used to study and compare with real-life networks.

• Network Evolution:

◦ How networks grow over time (like social media, web pages).

◦ Some nodes become more popular ("rich get richer").

6. Social Context: Af liation & Identity

• Af liation: People are part of different groups (e.g., clubs, companies, online forums).

• Identity: People’s roles, interests, and backgrounds affect their behavior online.

• Helps in targeted marketing.

7. Web Analytics Tools

• A/B Testing:
fi
fi
 ◦ Compare two versions (A and B) of a webpage or ad to see which performs better.

◦ Example: Different headlines, buttons, images.

• Online Surveys:

◦ Ask users for feedback to understand their behavior and satisfaction.

• Web Crawling and Indexing:

◦ Web Crawling: Bots (like Googlebot) scan the web to nd and collect web pages.

◦ Indexing: Organizing collected pages so they can appear in search results.

8. Natural Language Processing (NLP) for Micro-text Analysis

• NLP is used to analyze text data like tweets, comments, reviews.

• Micro-text: Very short text like tweets, SMS, etc.

Techniques include:

• Sentiment Analysis: Detect if the text is positive, negative, or neutral.

• Topic Extraction: Find the main subject or idea.

• Keyword Extraction: Identify important words.

• Spam Detection: Filter unwanted messages.

Would you like a summary chart or visual diagram for easy revision?

You said:

more detailed explanation

ChatGPT said:

Great! Here's a more detailed and easy-to-understand explanation of each topic under Web
Metrics & Analytics, KPIs, Graphs & Matrices, and NLP Techniques for your MBA studies:

✅ 1. Web Metrics & Analytics – Common Metrics (Explained

in Detail)

🔹 Hits

• Every element (image, video, script) on a webpage counts as a hit when the page is loaded.
fi
 • Not a good indicator of user activity because one page can create many hits.
✅ Example: A page with 1 HTML le + 5 images = 6 hits.

🔹 Page Views

• Total number of times a page is loaded or reloaded.

• Includes repeat views by the same user in the same session.

✅ Example: If 1 person views a page 3 times = 3 page views.

🔹 Visits (or Sessions)

• A session begins when a user visits your website and ends after:

◦ 30 minutes of inactivity, or

◦ Midnight (if the user stays past 12 AM)

• ✅ Includes: All activities during that visit: page views, clicks, conversions, etc.

🔹 Unique Page Views

• Counts only one view per page per session, even if the user visits the page multiple times.
✅ Example: 1 user visits the same page 5 times in one session = 1 unique page view.

🔹 Bounce

• A bounce is when a user lands on a page and leaves without clicking anything or visiting
another page.

🔹 Bounce Rate

• Measures the % of visitors who bounce.

📌 Formula:
Bounce Rate
=
(
Single Page Visits
Total Visits

)
fi
 ×
100

\text{Bounce Rate} = \left( \frac{\text{Single Page Visits}}{\text{Total Visits}} \right)

\times 100

Bounce Rate=(Total VisitsSingle Page Visits )×100

• High bounce rate can mean:

◦ Irrelevant content

◦ Slow loading time

◦ Poor user experience

✅ Ways to Improve Bounce Rate:

• Improve page speed

• Use engaging visuals and headlines

• Add related internal links

• Optimize for mobile

🔹 Average Time on Site

• Average time a visitor spends on your website during a session.

✅ Longer time = More interest and better content.

🔹 Real-Time Report (in Google Analytics)

• Shows live data about:

◦ Current visitors

◦ Pages they are viewing

◦ Traf c sources (how they found you)

◦ Locations (countries/cities)

🔹 Traf c Source Report

Shows where your traf c is coming from:

fi
fi
fi
 • Direct: Typed your URL directly.

• Organic Search: Found via search engines like Google.

• Referral: Clicked from another website.

• Social: From Facebook, Twitter, Instagram, etc.

• Paid Search: From ads (e.g., Google Ads).

• Email: From email campaigns.

🔹 Custom Campaigns (UTM Tracking)

• Track speci c marketing campaigns using tagged URLs (with UTM codes).

• Helps you know which campaign or ad brought the most users.

🔹 Content Report

• Shows performance of each page:

◦ Page views

◦ Bounce rate

◦ Average time

◦ Exit rate

• Helps identify top-performing and low-performing content.

✅ 2. Google Analytics
A free web analytics tool from Google to track and analyze website performance.

📌 Main Features:

• Tracks users, sessions, bounce rate, page views, conversions, etc.

• Shows user demographics and behavior

• Tracks campaigns, traf c sources, devices used

• Offers custom dashboards, real-time data, and goal setting

✅ Example: You can set a "Goal" to track how many users completed a purchase or lled a contact
form.
fi
fi
fi
✅ 3. Key Performance Indicators (KPIs)

🔹 What are KPIs?

• KPIs are speci c and measurable metrics that show how well a website or business is
doing.

🔹 Need for KPIs

• Monitor performance

• Align goals with results

• Make data-based decisions

• Improve marketing strategies

🔹 Characteristics of Good KPIs

1. Relevant – Matches your business goal

2. Measurable – Can be calculated

3. Achievable – Realistic targets

4. Time-bound – Includes a deadline

5. Clear – Easy to understand

🔹 Perspective (Types of KPIs)

Perspectiv
Example KPIs
e
Cost Per Click, Conversion Rate,
Marketing
Impressions
Sales Revenue, No. of New Customers
Website Bounce Rate, Average Session Duration
Customer Satisfaction Score, Retention Rate

🔹 Uses of KPIs

• Measure progress
fi
 • Identify weak areas

• Make improvements

• Evaluate employee performance

✅ 4. Graphs and Matrices

Used in network analysis to understand relationships (e.g., in social media or web linking).

🔹 For Individuals (Node-level Measures)

• Degree Centrality: Number of direct connections a node (person/page) has.

• Betweenness Centrality: Measures how often a node lies on the shortest path between
others. Shows in uence.

• Closeness Centrality: How quickly a node can reach others.

🔹 For Networks (Overall Network-level Measures)

• Density: Proportion of possible connections that actually exist.

• Centralization: Whether a few nodes dominate the network.

✅ 5. Random Graphs & Network Evolution

🔹 Random Graphs

• Graphs where connections between nodes are made randomly.

• Used to study average behavior of large networks.

🔹 Network Evolution

• Over time, some nodes gain more links (like in uencers or viral posts).

• “Rich get richer” effect – new nodes tend to connect with already well-connected nodes.

✅ 6. Social Context: Af liation & Identity

🔹 Af liation
fi
fl
fi
fl
 • People are linked through groups (e.g., fan pages, clubs, forums).

• Helps in community detection and marketing.

🔹 Identity

• Each person’s identity (gender, age, location, interests) affects how they interact online.

• Useful for targeted advertising and personalization.

✅ 7. Web Analytics Tools

🔹 A/B Testing

• You create two versions of a page (A and B).

• Show them to different users.

• Measure which one performs better (clicks, sign-ups, etc.)

✅ Example: Test two different headlines to see which gets more conversions.

🔹 Online Surveys

• Ask users directly for feedback (via forms, pop-ups).

• Helps understand their likes, dislikes, and satisfaction.

🔹 Web Crawling & Indexing

• Web Crawling: Search engine bots (like Googlebot) go through web pages to collect data.

• Indexing: Collected data is stored in databases and used to show search results.

✅ 8. Natural Language Processing (NLP) Techniques for

Micro-text Analysis
Micro-text = Short and informal text (tweets, reviews, WhatsApp messages)

🔹 Techniques:

1. Tokenization – Splitting text into words or phrases.

2. Sentiment Analysis – Detecting emotional tone (positive/negative/neutral).

 3. Named Entity Recognition (NER) – Identifying names of people, brands, places.

4. Topic Modeling – Finding common themes or subjects.

5. Spam Detection – Identifying unwanted messages.

6. Language Detection – Recognizing the language of the text.

7. Hashtag Analysis – Tracking trending topics.

📊 Common Web Metrics

Metric Description
Total number of les requested from a server (not reliable – every image,
Hits
CSS counts).
Page Views Total number of pages viewed by users.
Visits (Sessions) Number of visits by users – includes multiple page views during one session.
Counts only one view per user per session, regardless of repeat visits to the
Unique Page Views
same page.
Bounce When a user leaves the site after viewing only one page.
Bounce Rate % of single-page sessions. High bounce rate indicates poor engagement.
Average Time on How long users stay on the site (average session duration).
Site
Real-Time Report Shows current visitors, their location, pages they are on, traf c sources.
Traf c Source Shows where visitors came from (e.g., Organic search, Direct, Referral,
Report Social, Paid).
Custom Campaigns tracked using UTM parameters to analyze marketing efforts.
Campaigns
Content Report Shows top-performing pages, time spent, bounce rate per page, etc.

✅ Key Performance Indicators (KPIs)

Need
fi
fi
fi
 • KPIs help measure performance against business goals, like engagement, lead generation,
or conversions.

Characteristics

• SMART: Speci c, Measurable, Achievable, Relevant, Time-bound

• Actionable: Leads to decision-making

• Aligned: Tied to business strategy

Perspective

• Marketing KPIs: Click-through Rate (CTR), Conversion Rate

• Sales KPIs: Cost per Lead (CPL), Revenue

• Customer KPIs: Customer Lifetime Value (CLV), Retention Rate

Uses

• Track performance of campaigns

• Identify what’s working

• Guide future strategies

• Report to stakeholders

📈 Graphs and Matrices

Basic Measures for Individuals (Nodes)

• Degree: Number of connections

• Centrality: Importance in the network (e.g., betweenness, closeness)

• Clustering coef cient: Measure of how connected the node’s neighbors are

Basic Measures for Networks

• Density: Proportion of possible connections that are actual connections

• Diameter: Longest shortest path in the network

• Average Path Length: Average distance between nodes

🔀 Random Graphs & Network Evolution

fi
fi
 • Random Graphs: Networks generated by randomly connecting nodes.

• Evolution: Social networks evolve over time; e.g., users join, connections grow, some links
break.

• Models like Barabási–Albert show how networks grow through preferential attachment
(popular nodes get more links).

🌐 Social Context: Af liation & Identity

• Af liation: User connections through shared groups, events, or interests.

• Identity: User traits like name, age, location, gender, etc., used for targeting and
personalization.

• Both help in community detection and recommendation systems.

🛠 Web Analytics Tools

1. A/B Testing

• Compares two versions of a webpage to see which performs better.

• Example: Two landing page versions tested for higher conversion.

2. Online Surveys

• Collect user feedback directly.

• Useful for qualitative insights: satisfaction, usability, etc.

3. Web Crawling & Indexing

• Crawling: Automated programs (bots) go through websites to collect data.

• Indexing: Organizing this data for fast retrieval (used in search engines).

🧠 Natural Language Processing (NLP) Techniques for Micro-

text Analysis
• Micro-text: Short texts like tweets, comments, posts

• NLP Techniques:

◦ Tokenization: Breaking sentences into words

fi
fi
 ◦ Sentiment Analysis: Positive, negative, or neutral sentiment

◦ Hashtag & Keyword Extraction

◦ Topic Modeling: Discovering hidden topics (e.g., LDA – Latent Dirichlet

Allocation)

◦ Text Classi cation: Categorizing text (e.g., spam vs not-spam)

fi