ONLINE ADVERTISING MODELING TO INCREASE SALES (CASE STUDY: LO TEXTILE BANDUNG)
The textile industry in Indonesia is rapidly growing, influenced by fashion trends, lifestyle changes, product quality awareness, and the impact of social media. In the fierce competition in this industry, companies strive to win market share through various marketing strategies, one of which is onl...
Saved in:
| Main Author: | |
|---|---|
| Format: | Theses |
| Language: | Indonesia |
| Subjects: | |
| Online Access: | https://digilib.itb.ac.id/gdl/view/80926 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | The textile industry in Indonesia is rapidly growing, influenced by fashion trends, lifestyle changes, product quality awareness, and the impact of social media. In the fierce competition in this industry, companies strive to win market share through various marketing strategies, one of which is online advertising.
Online advertising has become a key element in the marketing strategy of the dynamic textile industry in Indonesia. In a competitive environment, companies like Lo Textile Bandung, one of the players in textile retail in Indonesia, face challenges in maintaining and increasing sales. One of the strategies used by Lo Textile Bandung to achieve sales growth is online advertising, particularly keyword search advertising. To achieve the desired sales targets, Lo Textile Bandung requires a predictive model for online advertising that can help in forecasting sales and identifying crucial variables to achieve those targets.
This research utilizes historical data from the Lo Textile Bandung’s advertising on the Shopee online marketplace platform to develop predictive models for keyword advertising using Multiple Linear Regression. Multiple Linear Regression is a statistical method used to analyze the relationship between two or more independent variables (predictors) and one dependent variable to predict future outcomes.
To develop the keyword advertising predictive model, the author goes through four stages: data collection, data pre-processing, model development, and evaluation of the model's results. The online advertising modeling process using Multiple Linear Regression is assisted by the Orange applications, an open-source data visualization, machine learning, and data-mining application. With this application, calculations such as gradient descent are done automatically. This model enables the analysis of correlations between various variables such as Impressions, Click-Through Rate, Conversion Rate and Gross Merchandise Value.
The results of this developing predictive model are used to forecast sales conversions and recommend keywords with good conversion performance. Keywords such as "diamond premium", "wollycrepe caltri”, and "toyobo deluxe" are some of the top keywords recommended for increased advertising budgets.
|
|---|