How media made for the many best targets the individual

Linda Margaret
7 min readJul 2, 2023

When I worked as a contractor in marketing for Big Pharma companies (and most of us marketers were contractors to ensure our clients maintained plausible deniability), the research was clear: people are fairly homogenous. We’re ninety-nine percent chimps and the remaining one percent that makes us homo sapiens varies, but not by much.

What really separates one sapiens from another are the stories s/t/he(y) tell themselves and others.

In market research, we call this the ‘profit margin.’

Online and off, evolving and deceptively malleable narratives provide the requisite emotional triggers underpinning lucrative content-fueled (and fueling) algorithms. Every pulse of endorphin that any aspect of data helps deliver to a human brain is informative — if not to the human, then to the machines and the AI-aided analysts trying to figure out how to exploit mass resources (human, financial, and otherwise.)

Just keep swimming, swimming, swimming… #DevilwearsPrada

For most market analysts, work like this pays the rent…at least until AI takes over. Then my plan is to become a battery a la ‘The Matrix.’

I’ve already submitted my resume to Siri AND Alexa. I’m not picking sides.

Yet.

Weapons of Mutually Assured Division

The point is that successful market researchers can cleverly categorize complex individuals living in dynamic environments into simple and stagnant ‘target audience’ containers and then use some sort of co-created ‘Master narrative’ to keep these otherwise unique individuals contained.

I say ‘co-created,’ of course, because the consumer contributes to maintaining and updating his/her/their market classification.

Essentially, consumers are encouraged to collaborate when they decorate their containers for the benefit of corporate revenue. (#BumperStickerNation)

Marketing or strategic engagement is a lot like LinkedIn — it works best if each targeted individual oversees their own profile(s) and, ideally, actively contributes to data processing using predefined tools and options.

Like. Subscribe. Share.

Consume.

Accept the cookies.

Or not.

It doesn’t really matter at this point.

Okay, so this isn’t news….or rather, it encompasses the news, and we’re limited in our (re)actions.

We’re each of us a diverse grab bag of ̶e̶v̶o̶l̶v̶i̶n̶g ̶ adapting data points.

This touches on a cornerstone of my fifth-grade democratic ̶p̶r̶o̶p̶a̶g̶a̶n̶d̶a̶ m̶a̶r̶k̶e̶t̶i̶n̶g̶ ̶c̶o̶m̶m̶u̶n̶i̶c̶a̶t̶i̶o̶n̶ theory (?) class: cross-cutting cleavages. Cross-cutting cleavages hypothetically ensure that no one ‘self-determining’ group dominates another in all areas of public policy.

Cross-cutting cleavages assume that each individual voter is multifaceted and, you know, flexible. No voter carries the various voting models that they build inside their heads using disparate datasets from one area of inquiry and decision-making to another because, obviously, that would be overfitting.

Also overfitting.

Creating or Training a (machine and/or human) learning model

For example, my encyclopedic knowledge of reality television is useful when voting down the show 90 Day Fiancé on Rotten Tomatoes and engaging in the…um….germane critical media discourse….on Reddit, but this same mental model that I carefully constructed from hours of wasted time is not applicable to US foreign policy and trade with Central America.

Therefore I must build a new, more applicable mental model using more relevant, possibly less anecdotal data. I must retrain my brain at least in this subject (and subjective) area before making any decisions and/or endorsements. I must gorge on new data.

To the digital algorithms and more pertinent clickbait!

90-Day Fiance. Best foot forward.

How do we escape incentivized circular reasoning?

We can’t.

We’re doomed.

Just kidding.

Maybe.

This is called a ‘split point.’ (90-day Fiance.)

Let’s reconceptualize our options.

For this, I’m going retro, back to the late 1980s and Hallin’s spheres.

Blame my algorithms.

A little nerdy nostalgia…

Hallin’s spheres are three concentric spheres used as a framing device in the last century to indicate audience complacency or diffidence with topics and topical packaging (e.g., what is said, who says it, how it is said, etc.)

Hallin’s spherical premise is that ideas within widely consumed media move from the outer sphere of deviance to the inner sphere of controversy to the innermost sphere of consensus.

Like a toilet, we humans as eighties media consumers accepted all shit delivered directly to us and, eventually, sucked it all down the same hole.

From nerd to jock to preppy, we all wanted to do Tom Cruise in Risky Business. #80SKID

Gather round, young whippersnappers. Granma’s gonna school you.

In the dark eighties…

Prior to widely-available algorithms, machine learning, AI, and personal cellular devices, popular and commercially viable media existed, according to Hallin, in the ‘central’ sphere, the sphere of consensus.

This, of course, did not always fully investigate the hierarchical structure behind the creation of the spheres - who and what runs media construction and dissemination, motivation, resource allocation, external influences, internal dissent, dissonance, etc. We weren’t high enough yet.

After all, the eighties drug of choice was a stimulant, not a depressive. I mean, just look at our fashion.

My god…the hair….It’s so beautiful.

Anyone born in the eighties remembers that the TV told you what you could watch and when, much like my father’s phone tells him how he should vote today. Our media algorithms were constructed from economically endorsed humans instead of just their tech. We hadn’t yet taught the machines to think, but we were working on it.

In between episodes of Coke.

Yeah. Sure. That Coke.

Scholar Piers Robinson updated Hallin’s simplistic framework in 2001, evaluating environments in which the media can be accused of manufacturing consent either for the elites or within elite circles or on behalf of selected elites, specifically within politics or public policy.

You know, the way Coke does.

Today shit flows both ways. Progress?

Underlying Robinson’s reboot of Hallin is the understanding of how offline and online filters that individuals, organizations both public and private, and intangible machines apply to what is consumed and how it is understood.

Subject largely to financial incentives, modern algorithms have reshaped and repositioned Hallin’s spheres into individually oriented Venn diagrams to which media can emotionally appeal and, at strategic, mathematically mappable moments, move concepts from one or two overlapping spheres to another. Gay marriage reflects the consensus? Add in a trans-rights controversy and for some unseen but statistically valid reason, gay marriage is now fluctuating between controversy and deviancy.

For those interested in entomology and life sciences, we chimps (adjacent) are synchronous in consensus and conflict; we flash our emotions together in beautiful (to marketing/engagement algorithms) togetherness over specific issues at specific moments, incorporating any resulting controversy or consensus into a constructed master narrative that corporate entities can co-opt to create our respective collective audience containers.

Ain’t science grand?

Serendipitious coincidences made cash-friendly.

When the above is done successfully for a critical mass of individuals – what is acceptable (and profitable) to any ‘mainstream audience’ changes.

(See my image below - note that controver$y is understood to be less long-term profitable, as indicated by the single $, than con$en$u$. What draws an initial audience does not necessarily maintain them or, market research shows, the credibility of their creator. This is why, for example, when planning to research and eventually sell a female arousal pill, many of the pharma companies seeking to create this market and a corresponding, financially willing consumer base decided to pool resources and fund an NGO, since discontinued, that promotes female empowerment in the bedroom. Often the most lucrative spontaneity is carefully planned and sponsored, at least initially, by corporate.)

Investment Frameworks.

Give this framework to any quality local media anthropologist, like those found at a reputable media mapping firm, and the experts at the firm will be able to immediately highlight influencers and topics that merit additional investment on the part of a specific info-awareness aka marketing campaign. For a sizeable fee, these experts will tell you where and how to invest your resources in recruiting consumers to your preferred consumption pattern.

This approach can also be helpful in finding the local 'experts' in digital media advocacy and education, often media-critical individuals who investigate, for the audience, where 'news' and 'entertainment' collide and why and how to support audience members in understanding their own media landscapes (and who often, in my albeit limited experience of the US market, have Patreon accounts as well as YouTube, etc.)

But that won’t win you in points (or pensions) at corporate HQ.

Thanks for watching!

Further reading:

https://www.amnh.org/exhibitions/permanent/human-origins/understanding-our-past/dna-comparing-humans-and-chimps#:~:text=These%20three%20species%20look%20alike,98.8%20percent%20of%20their%20DNA.

https://newlearningonline.com/literacies/chapter-8/kress-and-van-leeuwen-on-multimodality

https://www.nps.gov/grsm/learn/nature/fireflies.htm#:~:text=Why%20synchrony%3F,to%20one%20of%20her%20kind.

--

--

Linda Margaret
Linda Margaret

Written by Linda Margaret

I write academic grants etc. in Europe's capital. Current work: cybersecurity, social science. https://www.linkedin.com/in/lindamargaret/

No responses yet