Commercial photo for project, and graph on the Social Media Sentiment Analysis for La Roche-Posay on Instagram.
Project Scope
The project aimed to perform an in-depth analysis of social media sentiment regarding La Roche-Posay on Instagram. The goal was to leverage this analysis to inform and enhance the brand's marketing strategies. The primary objectives included understanding public perception of La Roche-Posay products, identifying key themes in customer feedback, and providing actionable recommendations to improve marketing campaigns and customer engagement.
My Role
As the lead data analyst, my responsibilities encompassed defining the project scope and objectives, meticulously collecting and preprocessing Instagram data, conducting comprehensive sentiment analysis using advanced NLP techniques, interpreting the results, and formulating strategic recommendations. Additionally, I was tasked with presenting these findings to the marketing team to facilitate data-driven decision-making. My role required a blend of technical expertise, strategic thinking, and effective communication to ensure the project's success.
Post Engagement Table on La Roche-Posay Instagram.
Python Code, and the result giving the sentiment score.
Work Process
The project began with the collection of social media data from Instagram using the platform's APIs. I focused on gathering posts, comments, and reviews mentioning La Roche-Posay over the past year to ensure a robust dataset. The data collection phase involved handling vast amounts of unstructured data, requiring careful planning and execution.
Once the data was collected, I embarked on the preprocessing phase. This involved cleaning the data to remove noise and irrelevant content, such as spam and unrelated posts. The text data was then tokenized, and stop words, punctuation, and special characters were removed to streamline the analysis process. Using Natural Language Processing (NLP) techniques, I normalized the text to prepare it for sentiment analysis.
Sentiment Analysis Results, and the two page comprehensive report.
For the sentiment analysis, I employed tools such as TextBlob to classify the sentiment of each post as positive, negative, or neutral. This step was crucial in understanding the overall sentiment distribution and identifying specific posts that influenced public perception. To visualise sentiment trends over time, I created detailed line charts and word clouds, which highlighted the most frequently mentioned keywords and their associated sentiments. Additionally, I performed topic modeling to uncover key themes and topics in customer feedback, providing deeper insights into customer concerns and interests.
Finally, I generated a comprehensive report summarizing the findings, insights, and strategic recommendations. The report included an analysis of the impact of major marketing campaigns on social media sentiment and provided suggestions for targeted marketing strategies. These strategies were designed to address areas of negative feedback, enhance positive sentiment, and ultimately improve customer satisfaction and engagement.
The outcome and results
The sentiment analysis revealed that major marketing campaigns significantly influenced public perception, highlighting product quality, customer service, and pricing as key themes. Based on these insights, I recommended strategies to address negative feedback and amplify positive sentiment. The marketing team implemented these suggestions, resulting in a 20% increase in positive sentiment over the next quarter. This project demonstrated the power of data-driven marketing and my ability to provide actionable business insights.
