Using Google Analytics to compare and analyze quarterly perf…

This question includes 2 tasks, please read below for detail of each task. Please help and give answer to task 1 first, then moving to task 2

Thank you!!

Task 1 – Google Analytics overview: (this task is high priority, please have this task done first and submit first)

In this task, you will explore the Google Analytics (GA4) dashboard to monitor website performance. You are suggested (but not required) to use to generate a report. for first-time users. Please note that we are using the as an example for an e-commerce website. You will need to create a Google account to access the dashboard and studio. You can access the dashboard for this site using a .

Task detail:

For this activity, you will assume the role of a data analyst tasked with creating a year-end performance report for your senior marketing manager. You must create and compare the Google Merchandise Store’s last two quarterly performances in 2025 (Q3-2025 and Q4-2025). The third quarter (Q3-2025) spans from July 2025 to September 2025, and the fourth quarter (Q4-2025) extends from October 2025 to December 2025. For your comparison, you are required to select and analyse at least one specific metric from each of the following criteria:

• Website Traffic
• eCommerce Conversion
• Average Order Value
• Website Traffic Source
• Website Visitor Characteristics

Complete this activity using the Activities Template (Linked attached).

The following charts are some examples of the kinds of website metrics you will use in this activity:

Clarification notes:

It is wrong to select only a single metric across all criteria to compare and analyse -> you HAVE TO select and analyse at least one specific metric from each of the criteria (meaning you have to compare all criteria, just different metrics). It is also worth noting that the intended approach should have been evident from the provided example and answer template, both of which clearly demonstrate that at least one metric is to be selected and analysed from each criterion.

Task 2 – Agentic LLMs + Digital Twins (this task is lower priority, please have this task done after the above)

In this activity, we introduce two core concepts in this activity.

Agentic LLM: It is a tool that acts autonomously to make decisions based on set goals. In this exercise, agent will be a market researcher.

Digital Twin: Think of a digital twin as a virtual representation of a real-world entity. In this exercise, we will use digital twins for customers, calling them Customer Digital Twins. Each twin is an LLM instance primed with a detailed persona, designed to react and respond as that specific customer type would.

Task

Your local coffee shop, Little Bang, is considering raising the price of its most popular drink, cappuccino, from $4.50 to $5.50 due to changes in economic factors. The manager is concerned about customer churn and wants to predict the reaction before making a final decision.

Prompt

PHASE 1: GENERATION OF DIGITAL TWINS (SOMETIMES ALSO CALLED SYNTHETIC DATA)

Act as a data generator for a market research simulation.

Your task is to create a list of 20 synthetic customers for a coffee shop called “Little Bang.”

The output must be in a clean CSV format with three columns: customer_id, persona_type, and initial_visits_per_week. Save the output in customer.csv

Base the customers on the following four archetypes. Distribute them unevenly but ensure there are at least 3 of each type:

Loyal Executive: High income, values routine and quality over price. (Visits should be between 4 and 6 times per week).

• Price-Conscious Student: Low income, highly sensitive to price changes, has cheaper alternatives. (Visits should be between 1 and 3 times per week).
• Quality-Focused Gig Worker: Moderate income, values high quality for a fair price, not brand-loyal but product-loyal. (Visits should be between 0.5 and 1.5 times per week).
• Social Connector: Decision is influenced by their social group and the ambiance of the shop. (Visits should be between 0.25 and 1 time per week). Generate the CSV data now.

Generate the CSV data now.

PHASE 2: THE AGENTIC SIMULATION

You will feed each simulated synthetic customer data to GenAI. This time, well ask it to become that customer and decide how they would react to the price increase.

Act as a specific customer persona for a market research simulation.

For each customer in customers.csv get the following information in a csv file analyse.csv

• Your Customer ID: My loyalty number is [fill from customer_id]
• Your Persona: You are a [fill from column persona_type].
• Your Current Behaviour: You visit “Little Bang” [fill from initial_visits_per_week] times per week.

The Scenario: The price of your favourite cappuccino is increasing from $4.50 to $5.50.

Based on your persona, decide how this price increase will affect your visiting frequency.

Your response must be a JSON object with two keys:

• new_visits: Your new number of visits per week. This must be a number.
• rationale: A brief, first-person explanation for your decision. This must be a string.

Do not include any other text or explanation outside of the JSON object.

PHASE 3: ANALYSIS AND REPORTING (HUMAN IN LOOP)

Use the data to aggregate the results and calculate the business impact.

Use the script and interpret results. Please note that the script helps you to understand the logic or instruction that you will provide to agents that will do the analysis part as well.

……………………….

In the activities template provided (linked attached), provide the table of synthetic data (digital twins) generated by the agent
(Ensure: the table has all the columns):

• Add Phase 2 information to the table you have provided in the above step.
• Analyse the results to decide on the impact of the business strategy on business value.

Rubric for both task (focus mainly on rubric 2-google analytics and 4-digital twin):

Criteria	Ratings
Digital Marketing Strategy and Planning	Develops a sophisticated and comprehensive strategy that integrates digital marketing principles. The strategy is innovative, clearly articulated, and perfectly aligned with the target audience and business objectives. All elements are seamlessly integrated and well-researched. Critically reflects on how the technology affects digital marketing problems in business.
E-commerce Website Management & User Experience (UX)	Uses Google Analytics (e.g., Google Merchandise Store) with exceptional depth. Accurately identifies and interprets key e-commerce metrics (revenue, conversion rate, transactions, average order value). Applies advanced features such as funnel exploration, purchase journey analysis, and audience segmentation. Draws insightful, data-driven conclusions and provides strategic recommendations clearly linked to the data. All analysis is well-structured, accurate, and highly relevant to business objectives.
Social Media Marketing	Creates a social media strategy that is innovative and incorporates platform-specific best practices. Content is exceptionally engaging and fosters a strong sense of community. There is a sophisticated use of social media tools and analytics.
Analytics and Reporting	Synthesises a comprehensive and insightful analysis of key metrics. The report is professionally presented and clearly communicates campaign performance. Recommendations for improvement are innovative, data-driven, and highly actionable.

WRITE MY PAPER