Does Nvidia speak differently to engineers than to everyone else?
Nvidia tunes complexity by audience: technical detail and a formal tone for engineers, plain words, visuals, and metaphor for everyone else, with different effects on trust and engagement.
Hypotheses.
What the paper argues, written so it can be tested.
- H1
Nvidia’s engineer-facing posts carry more technical specificity, direct product utility, and formal tone than its general-audience posts, which carry more metaphor, visual storytelling, and plain wording.
- H2
Engineers rate technical, formal excerpts as more credible and engaging, and non-engineers rate plain, visual excerpts that way, as central and peripheral routes of persuasion predict.
The model.
One announcement goes out as two streams. The engineer stream uses central-route cues like specs, how-to steps, and a formal tone. The general stream uses peripheral-route cues like visuals, metaphor, and plain words. The two link to each other so a reader can step up or down in complexity.
Partial abstract.
A comparison of Nvidia’s engineer-facing and general-audience communication across channels. The paper argues the company moves between central-route credibility for engineers and accessibility for everyone else, and that each earns trust differently.
Audience Voice and Directed Levels of Complexity: How Nvidia Speaks to Engineers vs. Non-Engineers
Key sources
- Petty, R. E., and Cacioppo, J. T. (1986). The elaboration likelihood model of persuasion.
- Kincaid, J. P., et al. (1975). Derivation of new readability formulas for Navy enlisted personnel.
- Neuendorf, K. A. (2017). The content analysis guidebook.
How we’ll test it.
No results yet. This is the plan: the data, the measure, and what would count as support.
The data
Posts from the Nvidia Developer Blog, the Newsroom, the GeForce blog, and GTC keynotes from January 2024 to June 2025, coded on body text, plus a survey that shows excerpts to engineers and non-engineers.
The measure
Acronyms per 100 words, passive-voice ratio, and Flesch Reading Ease per post, and how readers rate credibility, engagement, emotional resonance, and brand trust.
What would count as support
Engineer-facing posts score clearly higher on the technical cues, and each audience rates the stream written for it above the other one.
How it connects to Pulp.
Pulp’s interaction research: one message at different levels of complexity.
Also in Interaction
Doing work like this?
Students from any school or field, and mentors from academia or industry: get in touch.