Project information
- Category: Data Analytics
- Client: REFASH Singapore
- Project date: Sept 2020 - Nov 2020
Social Analytics | REFASH Singapore
REFASH Singapore is the largest thrift store brand for women in Singapore with thousands of new styles from both local and international brands at up to 70% off estimated retail.
In light of the COVID-19 situation, they faced a decrease in demand for both buying and selling of secondhand clothes, coupled with a relatively low engagement rate on their social media platforms, their business has been affected. To tackle this issue, we aim to help improve their social media presence and engagement by analysing their social media channels. This would in turn improve their branding and be useful in amplifying demand for the brand.
We use a series of tools to perform the analysis, from which, we provided suitable recommendations to REFASH Singapore.
The following are a few examples of tools used:
| Tools/technologies | Purpose and rationale |
|---|---|
| Selenium: An open source web automation tool | Mainly for heavy web scraping of the Instagram data |
| Python Reddit API Wrapper (PRAW): Python package that allows for simple access to Reddit's API | Scraping of Reddit Data that we require on REFASH |
| Instaloader: Tool to download pictures (or videos) along with their captions and other metadata from Instagram | Light scraping of Instagram data |
| Gephi: Open-source software for network visualization and analysis | Visualising the hashtag network graph and calculating the various centrality measures |
| Google Analytics: Free web analytics tool offered by Google to help analyze website traffic | Connected it to REFASH’s Shopify page to view the website traffic |
| Google Data Studio: Free tool from Google to make custom reports with data from Google’s marketing services and external sources | Used to build an interactive dashboard for an in-depth look at the data, making it understandable for users |
| Word Cloud: Word cloud generator in Python | Generate word cloud of popular hashtags used by thredUP |
| OCRSpace: Service converts scans or (smartphone) images of text documents into editable files by using Optical Character Recognition (OCR) | Converts Instagram posts with text to editable text, used mainly on refash.reveal Instagram page where they post posts with text |
| Sentiment Analysis (using VADAR lexicon): Valence Aware Dictionary for Sentiment Reasoning (VADER) is a model used for text sentiment analysis that is sensitive to both polarity (positive/negative) and intensity (strength) of emotion | Used to determine the sentiment analysis of posts’ captions |
| Topic Modelling with LDA (Latent Dirichlet Allocation): LDA is an example of a topic model and is used to classify text in a document to a particular topic. It builds a topic per document model and words per topic model, modelled as Dirichlet distributions | Used to generate topics on the Reddit thread comments |
| Natural Language ToolKit (NLTK): Open-source library for building Python programs that deal with natural language | Removal of stopwords when pre- processing text data |
| TextBlob: Simple API for processing textual data | Lemmatization of words when pre- processing text data |
| Netlify: Web hosting and serverless backend services for web applications and static websites. | Used to host our deployed dashboard on the web |
As most of the information surround the confidentiality of REFASH's data, the specific insights could not be listed here, however, our recommendations surround these main areas:
- Use of recommended hashtags
- Optimal post timings
- Search Engine (SE) & Mobile Optimisation
- Topics to Address
The hashtags are generated after using various centrality measures to provide a list of recommended hashtags for the best outreach. As for the optimal post timings, we cross referenced with other competitors that REFASH Singapore has, such as thredUP, and we also looked at the posts with highest engagement rate. Optimal timings are generated and suggested to REFASH Singapore to post their content during specified timings for the most reach.
As we also realise, from their Shopify data that most users tend to only stick around their website and online store for a short while. By implementing better mobile optimisiation (specific call-to-action buttons), this would improve the overall user experience when using the online store. Improving Search Engine optimisation would also lead users to their online stores more successfully. As for the topics to address, we look through data on Instagram and Reddit to find important topics that customers are unhappy about with the brand, and using LDA, we manage to generate topics that REFASH Singapore could focus on and improve customer experiences and thus their brand overall.