This report provides a comprehensive analysis of ABC Retail's sales performance over the past 5 years. It covers descriptive analytics to summarize historical sales trends, predictive analytics to forecast future performance, and prescriptive analytics to recommend strategies for growth.
Key Findings: The sales show a consistent pattern across the years, with average sales over the past 5 years resting at around ₱15,875. There is a noticeable increase in sales from February to April — late Q1 — each year. Customer feedback scores tend to decline from October to February but rise again from February through October. Importantly, stock levels remained healthy and never fell below 450 units, which was the minimum inventory level observed the periods.
Key Recommendations: It's recommended to investigate customer feedback trends from October to February to identify the factors contributing to lower satisfaction during this period. Based on these findings, strategies can be implemented to improve service or consider removing products that may be affecting customer satisfaction.
This report improves ABC Retail's historical data to uncover key insights, predict future outcomes, and provide actionable recommendations that can help drive business growth and smarter decision-making. The analysis uses descriptive, predictive, and prescriptive methods, assuming past trends will continue under stable conditions.
This section analyzes past sales performance, monthly trends, and average performance to understand business patterns over the past five years.
Total Sales Per Year:
Average Monthly Sales:
| Year | Sales (Php) |
|---|
The annual growth rate remained fairly consistent from 2019 to 2023, averaging about 3–3.4%. Specifically, sales grew by 3.36% from 2019 to 2020, 3.25% from 2020 to 2021, 3.15% from 2021 to 2022, and 3.05% from 2022 to 2023. The slight slowdown each year suggests a gradual leveling of growth.
| Month | Sales (Php) |
|---|
The Monthly Trend Chart shows a clear pattern in sales performance. Each December typically reaches a peak — the highest point of the year — but then drops sharply in January. However, from January onwards, sales gradually rise again, reflecting a consistent pattern of recovery and growth each subsequent year. This highlights a seasonal dip followed by a steady upward trend across years.
| Month | Feedback Score (1-5) |
|---|
The Feedback Chart shows a clear pattern: Customer feedback scores tend to decline from October to February, reflecting a seasonal dip in satisfaction. However, from February through October, scores rise again, indicating recovery and improvement in customer sentiment.
| Month | Inventory Level (units) |
|---|
The Inventory Chart shows that stock levels remained healthy and stable, never falling below 450 units, indicating effective inventory management and consistent product availability.
The descriptive analysis reveals a steady growth in sales from 2019 to 2023, with annual sales consistently increasing by about 3-4%. The pattern shows a dip in January following a peak in December, and this pattern repeats each subsequent year.
Customer feedback scores typically decline from October to February, then rise again from February through to the following October. Nonetheless, this fluctuations are temporary and follow a consistent seasonal pattern.
Inventory levels remained healthy and stable, never falling below 450 units, which highlights effective stock management and a strong ability to meet customer demand without experiencing shortage.
Overall, the descriptive analysis underscores a positive trend in sales, a predictable seasonal pattern in customer feedback, and stable inventory levels, reflecting a well-managed and growing retail business.
This section presents a sales forecast for the upcoming quarter using a linear regression model trained on recent historical data.
Forecast for Next Quarter:
| Month | Forecasted Sales (Php) |
|---|
We prepared the data by cleaning and aggregating historical sales. The regression model uses time (in months) as the main feature to capture trends. By applying this approach, we were able to reflect the ongoing pattern of sales growth accurately.
The data was split into 80% training and 20% testing. The regression shows a Mean Absolute Error (MAE) of Php 500, and an R² score of 0.95, indicating a strong fit and reliable performance on unseen data.
The slope suggests that sales increase by about Php 500 each subsequent period, reflecting a steady upward trend. This highlights a consistent growth pattern, which can aid in future business decisions and forecasting.
Based on the patterns and trends we identified in the historical data, we can implement strategies to maximize opportunities and drive growth.
The data we analyzed reveals a clear and insightful pattern in the business's performance. From the descriptive analysis, we observed that sales typically peak in December and drop in January, while customer feedback tends to fluctuate, declining from late in the year into February and then rebounding afterwards. Our inventory remained healthy and well-managed, never falling below 450 units, which suggests strong control over stock levels.
The diagnostic analysis highlights the key drivers of these trends. The drop in sales and feedback during certain months can be tied to seasonal factors and reduced promotions, while the upward trends in subsequent months reflect a recovery, driven by promotions, holidays, and growing customer loyalty.
Our predictive analysis, employing regression techniques, underscores these patterns. The regression shows a strong upward trend, with an average increase of Php 500 in sales each subsequent period. The high R² score of 0.95 and low Mean Absolute Error (MAE) of Php 500 demonstrate that our model accurately captured this pattern and can be relied upon for future forecasting and decision making.
Based on these findings, the prescriptive analysis advocates for a range of strategies to maximize performance. By identifying slow periods in advance, we can proactively implement promotions, special offers, and collaboration campaigns with influencers to boost visibility and drive sales upward. Furthermore, understanding when customer feedback drops lets us promptly address issues, improve products, and leverage promotions to reconnect with our base.
Taken together, these analyses enable the business to respond proactively to future trends, minimize weaknesses, and maximize opportunities. The combination of a strong, data-informed approach and a clear understanding of customer behavior will help drive sustained growth and success in the future.
The following appendices provide additional depth and context to our analysis. They include raw data, methodological details, and supplementary materials that support and validate the main findings presented in this report.
Here are snippets of code we used for regression and data transformation to produce our forecasts.
// Combine sales, feedback, and inventory into a unified array
const mergedData = salesData.map(sale => {
const year = sale['Year'];
const month = sale['Month'];
const feedback = feedbackData.find(f => f['Year'] === year && f['Month'] === month);
const inventory = inventoryData.find(i => i['Year'] === year && i['Month'] === month);
return {
year,
month,
sales: sale['Sales (Php)'],
feedback: feedback ? feedback['Feedback Score (1-5)'] : null,
inventory: inventory ? inventory['Inventory Level (units)'] : null
};
});
Explanation: Here we merged different datasets into a unified array of
data points. Each object contains sales, feedback, and inventory for a specific
Year and Month.
// Perform simple regression on the last 6 months to forecast future sales
const x = recent.map((_, i) => i + 1);
const y = recent.map(d => d.sales);
const n = x.length;
const avgX = x.reduce((a, b) => a + b) / n;
const avgY = y.reduce((a, b) => a + b) / n;
const slope = x.map((xi, i) => (xi - avgX) * (y[i] - avgY)).reduce((a, b) => a + b) /
x.map(xi => (xi - avgX) ** 2).reduce((a, b) => a + b);
const intercept = avgY - slope * avgX;
Explanation: Here we applied a simple linear regression formula to find
the slope and intercept, which lets us extrapolate future
sales
trends based on past data.
These appendices aim to provide full transparency and depth to the analytical process, allowing stakeholders to further explore, validate, or challenge the findings presented in this report.