Instagram, a main photo-sharing social network, plays a prime role in affecting consumer purchase decisions. Specifically, those accounts able to engage more effectively with their audience have the power to sway their beliefs. Moreover, people have been subjected to an increasing mistrust of paid advertising, which is pushing the relevance of organic marketing tools. In this context, hashtags become key within social media growth strategies. They are used to add metadata to a post and increase its reach, prefixing a word or phrase by a hash symbol (e.g. #hashtag). The current literature, on the one hand, lacks a uniquely accepted hashtag taxonomy (e.g. descriptive, emotive, etc.). On the other hand, it either focuses on alternative scopes of hashtags or, when considered as a driver for user engagement, just their total count was reckoned meaningful. The hypothesis behind this thesis is that distinct hashtag classes affect differently post engagement. Thus, the objectives of this study are 1) to develop a comprehensive hashtag taxonomy and 2) to assess the impact of each hashtag class on user engagement. The reference dataset was built collecting over a three month period the daily post information from 12 Instagram influencers, split among 3 different industries: food, fitness and fashion. It resulted in 1,406 posts and 3,135 hashtags collected in total. The development of several regression models led to the following conclusions: hashtags describing activities, generating communities, or related to popular feature accounts proved to be effective in stimulating engagement; however, those including time, space and camera information, related to a brand or a famous personality, and personal hashtags are harmful to post engagement levels. Using highly popular, shorter hashtags, relevant to the profile industry is also beneficial. Besides hashtag-related results, the study provided extra insights: having more followers, posting frequently, replying to comments, mentioning and tagging other profiles positively impact engagement; while the post caption does not affect tangibly engagement. This thesis provides researchers and managers with 1) a solid taxonomy to understand and classify the hashtag multidimensionality and 2) practical advice, both qualitative and quantitative, to support social media engagement strategies.
Instagram, un social network per la condivisione di immagini, impatta fortemente le decisioni d’acquisto del consumatore. Nello specifico, i profili che coinvolgono maggiormente i propri seguaci hanno più influenza sulle loro opinioni. Inoltre, la gente ha perso fiducia nella pubblicità a pagamento, favorendo l’adozione di strumenti di marketing organici. In questo contesto, gli hashtag diventano un fattore chiave per le strategie di crescita sui social network. Essi sono usati per etichettare i post e aumentare la loro portata, prefissando una parola o frase con un ‘cancelletto’ (#hashtag). La letteratura attuale, da un lato, manca di una classificazione degli hashtag univoca (descrittivi, emotivi, ecc.), dall’altro, o si focalizza su funzioni alternative degli hashtag o, quando legati al coinvolgimento utente, viene considerata solo la loro quantità. L’ipotesi dietro questa tesi è che diverse classi di hashtag impattino diversamente sul coinvolgimento degli utenti verso un post. Dunque, gli obiettivi di questo studio sono 1) lo sviluppo di una classificazione esaustiva degli hashtag e 2) la valutazione dell’impatto di ogni classe di hashtag sul coinvolgimento dell’utenza. Il set di dati usato è stato generato raccogliendo giornalmente per tre mesi i post di 12 profili di influencer su Instagram, divisi in tre settori: alimentare, fitness e moda. In totale, sono stati raccolti 1,406 post e 3,135 hashtag. Lo sviluppo di diversi modelli di regressione ha portato alle seguenti conclusioni: gli hashtag che descrivono attività, generano comunità, o connessi a feature accounts stimolano positivamente l’utenza; tuttavia, quelli con informazioni di tempo, spazio, tipo di scatto, relativi ad un marchio o una personalità, e gli hashtag personali sono dannosi al coinvolgimento degli utenti. Ancora, utilizzare hashtag popolari, corti e pertinenti al settore del rispettivo profilo è vantaggioso. Oltre ai risultati relativi agli hashtag, lo studio ha fornito delle considerazioni addizionali: avere più followers, postare di frequente, rispondere ai commenti, menzionare e taggare altri profili impattano positivamente il coinvolgimento; invece, la descrizione del post non lo fa in modo tangibile. Questa tesi fornisce a ricercatori e manager 1) una classificazione solida per comprendere e catalogare le multidimensionalità degli hashtag e 2) consigli pratici, sia qualitativi sia quantitativi, per supportare strategie di social media engagement.
The impact of different hashtag classes on user engagement on Instagram
D'AMATO, MICHELE ITALO
2017/2018
Abstract
Instagram, a main photo-sharing social network, plays a prime role in affecting consumer purchase decisions. Specifically, those accounts able to engage more effectively with their audience have the power to sway their beliefs. Moreover, people have been subjected to an increasing mistrust of paid advertising, which is pushing the relevance of organic marketing tools. In this context, hashtags become key within social media growth strategies. They are used to add metadata to a post and increase its reach, prefixing a word or phrase by a hash symbol (e.g. #hashtag). The current literature, on the one hand, lacks a uniquely accepted hashtag taxonomy (e.g. descriptive, emotive, etc.). On the other hand, it either focuses on alternative scopes of hashtags or, when considered as a driver for user engagement, just their total count was reckoned meaningful. The hypothesis behind this thesis is that distinct hashtag classes affect differently post engagement. Thus, the objectives of this study are 1) to develop a comprehensive hashtag taxonomy and 2) to assess the impact of each hashtag class on user engagement. The reference dataset was built collecting over a three month period the daily post information from 12 Instagram influencers, split among 3 different industries: food, fitness and fashion. It resulted in 1,406 posts and 3,135 hashtags collected in total. The development of several regression models led to the following conclusions: hashtags describing activities, generating communities, or related to popular feature accounts proved to be effective in stimulating engagement; however, those including time, space and camera information, related to a brand or a famous personality, and personal hashtags are harmful to post engagement levels. Using highly popular, shorter hashtags, relevant to the profile industry is also beneficial. Besides hashtag-related results, the study provided extra insights: having more followers, posting frequently, replying to comments, mentioning and tagging other profiles positively impact engagement; while the post caption does not affect tangibly engagement. This thesis provides researchers and managers with 1) a solid taxonomy to understand and classify the hashtag multidimensionality and 2) practical advice, both qualitative and quantitative, to support social media engagement strategies.| File | Dimensione | Formato | |
|---|---|---|---|
|
2018_07_DAmato.pdf
non accessibile
Descrizione: Thesis text
Dimensione
9.35 MB
Formato
Adobe PDF
|
9.35 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/141363