Obrona pracy szczegóły

Thesis title and aim

"Social Media Language Guide for GetHero." The aim is to develop a practical, scientifically grounded communicative instrument that helps GetHero's creators integrate commercial messages into their organic content without compromising their linguistic identity or audience trust. It is an applied linguistics thesis — the product itself is the main contribution.

Why this topic matters

Traditional vertical brand-to-consumer advertising is losing effectiveness due to advertising clutter. Influencer marketing works through peer-mediated persuasion, but commercial demands constantly threaten the authenticity that makes creators effective. The thesis solves a real, concrete agency problem.

Language Guide — full definition

A structured, evidence-based instrument with three sequential components: (1) the Tone of Voice Matrix (diagnostic), (2) the Engagement Playbook (tactical), and (3) the Visual-Linguistic Protocol (final check). It is a framework, not a rigid rulebook — creators keep their own voice.

GetHero — key facts

Wrocław-based influencer marketing agency, founded 2015, part of the international Webedia Group since 2018. Operates across four platforms (TikTok, YouTube, Instagram, Twitch) and three niches (Gaming, Lifestyle, Tech). Manages both macro-influencers and micro/nano-influencers.

Why micro-influencers matter

They have smaller but more engaged audiences; their communication feels more personalised and less commercialised, which is closely tied to perceived authenticity (Audrezet et al. 2018).

Central research problem

The communicative tension between the commercial requirements of brand partnerships and the authenticity demands of digital audiences — how to pursue commercial goals without breaking the parasocial bond.

Linguistic Engineering (my term)

The intentional, strategic modification of language to fulfil specific communicative functions in the influencer-brand ecosystem. Unlike copywriting (which aims at aesthetic/persuasive appeal), it uses pragmatic theory (Levinson 2010) and discourse analysis (Cook 2001) to build a linguistic environment that minimises audience resistance before it arises.

Tonal rupture (my term)

The moment in a sponsored post where the brand's corporate register suddenly displaces the creator's natural voice. It is the single most damaging event in a sponsored post because it signals commercial intent and triggers audience scepticism. Operationally detectable via Cook's participant model (the dominant "voice" shifts from addresser to sender).

Pragmatic dissonance

Instances where rigid, brand-mandated advertising structures conflict with the influencer's established idiolect (their individual, recognisable way of using language), making the creator's voice feel "foreign" and undermining social capital.

Methodology — overview

A qualitative, multidisciplinary approach combining pragmatic linguistics and discourse analysis. The primary method is qualitative discourse analysis (Cook 2001); a small cross-cultural audience survey provides triangulation. The methodology forms a closed loop: diagnosis → categorisation → prescription → implementation → evaluation.

Triangulation — why I use it

Combining multiple methods (content audit + audience survey + established theory) to cross-check findings. If the survey had contradicted the audit, I would have revised the recommendations; instead they converged, which lowers the risk that the Guide rests on a single flawed source.

Content audit — what and how

A qualitative analysis of selected GetHero sponsored posts across all three niches and three platforms, including both macro- and micro-influencers. Each post was analysed with Cook's (2001) discourse participant model, asking whether the brand-as-sender or the influencer-as-addresser dominated at each moment. Posts were treated as communicative events, not isolated texts.

Content audit — main finding

The most damaging moment is the tonal shift — when personal storytelling drops into corporate brand-speak (e.g. "Buy now via the link in bio") or an uncontextualised discount code. This breaks the parasocial contract: the audience's expectation that the creator speaks as themselves.

Content audit — niche differences

Gaming audiences are the most sensitive to tonal shifts (they detect scripted language instantly). Tech audiences tolerate formal language because authority is expected. For Lifestyle, sequencing is decisive — the brand mention must not appear before personal context.

Audience survey — design

Anonymous Google Forms survey in four language versions (Polish, Russian, Ukrainian, English). 25 valid responses, ages 18-35 (the core social media demographic). Hypotheses were withheld from participants to minimise social desirability bias. Ten questions: two scale (1-5), several choice questions.

Survey finding — authenticity

64% rated the importance of authentic language at 4 or above out of 5 (mean 3.84). This establishes authenticity as the dominant variable through which audiences evaluate sponsored content, and it held across all four languages, suggesting it is not culturally localised.

Survey finding — tone of voice

60% rated the influence of tone of voice at 4 or above (mean 3.80). How something is said — register, consistency with the creator's usual voice — is a central determinant of audience reaction. This justifies placing the Tone of Voice Matrix at the core of the Guide.

Survey finding — humour

92% said humour either increases (72%) or sometimes increases (20%) engagement. Humour is a low-resistance channel for commercial messaging; it lowers critical resistance (linked to Chiaro 1992 in the data table).

Survey finding — emojis

Only 32% view emojis positively; 44% are neutral or negative. Emojis used decoratively (rather than as tone signals) actively undermine credibility — this sets the Visual-Linguistic Protocol's ceiling of 2-3 emojis per post.

Survey finding — direct address

52% indifferent, but a combined 28% react neutrally or negatively to direct address ("you", "your"). This justifies replacing commands with opt-out constructions in the Politeness Protocol.

Survey finding — transparency

40% named transparency as the top trust factor. Audiences accept disclosure when it is framed as a shared benefit (e.g. "I managed to get us a small discount") rather than a corporate announcement.

Cross-cultural finding — detail

Russian-speaking respondents rated tone sensitivity at mean 4.1 vs 3.3 for the Polish-language group. The Polish subsample also showed the highest negative reaction to direct address. This corroborates Stawarz-García (2015) and justifies the Cultural Calibration module.

Parasocial interaction (Lou & Yuan 2019)

A psychological bond in which audience members develop a sense of intimacy and "friendship" with a media figure, despite the one-sided, non-reciprocal nature of the medium. The foundation of influencer trust.

Human brands (Lou & Yuan 2019)

Influencers function as "human brands" where an individual's personal life and personality are the core commercial product, creating a relatable identity that contrasts with traditional corporate marketing.

Authenticity as cornerstone (Audrezet et al. 2018)

"Perceived authenticity is the cornerstone of the influencer-follower relationship." Influencers must balance commercial requirements with personal expression; authenticity is a managed linguistic performance, not an accident.

Socially-situated identity (Gee 1999, 2011)

Identity is built through language in specific contexts. Gee: "Who we are… and what we are doing and the ways we are being and doing them" (1999: 13).

Big-D Discourse (Gee 1999)

A "Discourse" with a capital D is not just a style of speaking but a complete way of being in the world — integrating words, acts, values, beliefs, attitudes, and social identities. This grounds the persona concept in the Matrix.

Speech acts (Yule 2017)

Utterances that perform functions (advising, requesting, directing). "Pragmatics is the study of invisible meaning… how we recognise what is meant even when it isn't actually said or written" (Yule 2017: 128).

Social distance (Yule 2017, Levinson 2010)

The perceived relational distance between speaker and listener, which governs linguistic choices. Creators minimise it with deictic expressions like "we", "us", "here with you" to create an illusion of intimacy. It is dynamic, not fixed.

Politeness Theory (Brown & Levinson 1987)

Communication manages "face" (a person's public self-image). Positive face = desire to be liked/included; negative face = desire for autonomy. Creators use positive politeness (in-group markers, nicknames) to build common ground.

Face-Threatening Act / FTA

A communicative act that imposes on the listener's autonomy or self-image. A direct imperative ("Buy this now", "Subscribe") is a high-severity FTA because it presupposes a power asymmetry that contradicts the creator's peer identity.

Negative Politeness strategy

A way of making requests that explicitly acknowledges the listener's right to decline, using hedging and opt-out clauses (e.g. "If you're looking for something new…", "Let me know what you think if you try it"). Protects the follower's autonomy.

Discourse of Advertising (Cook 2001)

"Discourse is text and context together in interaction: only when we have all of these can we say that we have an act of communication" (Cook 2001: 4). Advertising is dynamic, not a static text.

Sender vs addresser (Cook 2001)

The sender originates the message (the brand); the addresser physically delivers it (the creator). Persuasion depends on how the addresser's persona is perceived by the addressee (the audience). The core of identifying tonal rupture.

Substance and paralanguage (Cook 2001)

Substance = the digital medium itself; paralanguage = the visual, auditory, and interactive elements (emojis, hashtags, layout). Allows a holistic, multimodal analysis of the post.

Liking (Cialdini 2016)

People are more inclined to comply with requests from those they know, like, and perceive as similar. Operationalised as "Affiliative Speech Acts" — language emphasising shared experiences and common ground, turning a CTA into a peer recommendation.

Authority (Cialdini 2016)

Credibility achieved by framing recommendations within a discourse of expertise, so product mentions read as expert advice rather than paid scripts. Especially relevant to the Expert persona.

Multimodality (Kress & van Leeuwen 1996, 2001)

"Meaning is made in many different ways, and always through more than one semiotic mode at the same time." Text, image, typography, and layout all carry meaning; the Visual-Linguistic Protocol applies this "grammar of visual design."

Emojis as digital gestures (Ge & Gretzel 2018)

Emojis are rhetorical devices that establish social presence and reduce ambiguity; paralinguistic cues equivalent to a change in vocal tone or a knowing pause. Crucially, they confirm tone — they do not introduce it — so they belong after a verbal cue.

Lateral Marketing (Kotler & Trias de Bes 2003)

When traditional segmentation and vertical messaging saturate, a "lateral" shift introduces the product as a solution to a relatable everyday problem inside the creator's organic content. The conceptual basis for adapting messages per persona.

Advertising clutter (Kotler 2003)

The saturation of the media environment with promotional messages, so individual communications lose distinctiveness and fail to reach awareness. The structural problem the Guide is engineered to bypass.

Webwriting / copywriting (Jabłoński 2018)

Digital readers scan content before reading, processing visual structure (breaks, emphasis, spacing) first. The Visual-Linguistic Protocol turns this into rules: a line break before the brand mention is a communicative act, not a stylistic choice.

Credible dialogue (Wójcik 2013)

Professional communication must rest on transparency, consistency, and responsiveness to build lasting trust. Grounds the implementation phase — the Guide is a foundation for credible dialogue, not a rigid set of rules.

Polish market sensitivity (Stawarz-García 2015)

Polish audiences are particularly sensitive to overtly aggressive sales tactics and value high-quality, informative content with genuine added value. Corroborated by the survey and built into the Cultural Calibration module.

Content analysis (Krippendorff 2004)

A systematic methodology for analysing communication content. Used in the evaluation phase to assess the "authenticity score" — how well creators integrate commercial messages into organic discourse.

Tone of Voice Matrix — definition

The diagnostic entry point. It captures a creator's "vocal fingerprint" by placing them on two axes and assigns one of four personas, preventing "identity dilution" (the erosion of a distinctive voice when every campaign uses the same corporate language).