Predictive and Comparative Analysis of Social Media Platform Performance for SMEs in Ireland and Nigeria

Abioro, Olamide Isaac (2026) Predictive and Comparative Analysis of Social Media Platform Performance for SMEs in Ireland and Nigeria. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (499kB) | Preview

Abstract

Small and medium sized enterprises increasingly rely on social media platforms to reach customers, yet it is still unclear which platforms and content strategies deliver the strongest engagement across different markets. This project builds a comparative prediction framework for Irish and Nigerian SMEs on Instagram, TikTok and Facebook. Using Apify based scrapers and a shared PostgreSQL database, a final sample of 40,217 public posts was collected from roughly 750 SME accounts across beauty, fashion and retail sectors.

After cleaning, cohort sizes ranged from roughly 650 to 2,700 posts per country sector platform combination. For each post, an engagement rate per 1,000 followers was constructed, and Random Forest and XGBoost classifiers were trained to predict whether a post would fall into the top quartile of engagement within its cohort, using temporal splits for validation. On held out test sets, the best cohorts achieved area under the ROC curve values between 0.80 and 0.93. Nigerian retail on Facebook and Nigerian fashion on TikTok achieved top decile lift values above 7, meaning the models captured more than seven times as many high engagement posts as a random selection of the same size.

The results show consistent gains from data driven posting strategies, with clear differences across platforms and sectors. Fashion and beauty SMEs benefit most from short video and mid length captions on Instagram and TikTok, while retail SMEs obtain comparatively stronger returns on Facebook. The framework is reusable for other markets and can support practical decision making for SME social media planning.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Khan, Sallar UNSPECIFIED
Subjects:	H Social Sciences > HF Commerce > Marketing Z Bibliography. Library Science. Information Resources > ZA Information resources > ZA4150 Computer Network Resources > The Internet > World Wide Web > Websites > Online social networks T Technology > TK Electrical engineering. Electronics. Nuclear engineering > Telecommunications > The Internet > World Wide Web > Websites > Online social networks H Social Sciences > HD Industries. Land use. Labor > Small Business Sector
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	09 Nov 2026 19:05
Last Modified:	09 Nov 2026 19:05
URI:	https://norma.ncirl.ie/id/eprint/9951

Actions (login required)

View Item