University of Oulu

Nuortimo, K., Karvonen, E. & Härkönen, J. Establishing social media firestorm scale via large dataset media analytics. J Market Anal 8, 224–233 (2020).

Establishing social media firestorm scale via large dataset media analytics

Saved in:
Author: Nuortimo, Kalle1; Karvonen, Erkki1; Härkönen, Janne1
Organizations: 1University of Oulu, Pentti Kaiteran katu 1, 90570, Oulu, Finland
Format: article
Version: accepted version
Access: open
Online Access: PDF Full Text (PDF, 0.5 MB)
Persistent link:
Language: English
Published: Springer Nature, 2020
Publish Date: 2021-06-19


A social media (SoMe) firestorm can present a liability for personal brands via the loss of reputation, as well as for the organisational brand image. The drastic measures often taken in these situations, especially in cases of negative media attention or a scandal, usually involve dismissal of the related persons. Hence, predicting, monitoring, analysing and measuring SoMe firestorms related to organisations or individuals can be beneficial. This paper describes SoMe firestorms and their effect, using media analysis involving opinion mining. The analysis focuses on the human trash (ihmisroska) scandal that was caused by a local centre party politician in Finland. The politician caused a SoMe firestorm by describing homeless people and substance addicts as ‘human trash’. The analysis utilises machine learning to classify 3300 media hits in the Finnish language to analyse their sentiment during the SoMe firestorm. General conclusions are drawn about the spread and influence of the SoMe firestorm to form a basis for wider global generalisation. The study formulates a scale for quantifying and analysing the influence of SoMe firestorms. The scale includes three classes relating to the exponential rise of the effect, starting from 1, with 3 being the highest. This scale aligns with the literature, which states that these events usually follow the same pattern. The case example provides further direction for the presented 1–3 level scale.

see all

Series: Journal of marketing analytics
ISSN: 2050-3318
ISSN-E: 2050-3326
ISSN-L: 2050-3318
Volume: 8
Pages: 224 - 233
DOI: 10.1057/s41270-020-00080-w
Type of Publication: A1 Journal article – refereed
Field of Science: 222 Other engineering and technologies
Copyright information: © Springer Nature Limited 2020. This is a post-peer-review, pre-copyedit version of an article published in J Market Anal. The final authenticated version is available online at