From Cigarette-Symbol to Snippet-Symbol: PR as Meaning-Making, Then and Now

In the late 1920s, the most effective advertising stopped describing products.

It started describing people.

Not the people using a product—the people becoming someone by choosing it. A cigarette could be framed as independence. A refrigerator could be framed as upward mobility. A brand could quietly turn into a shorthand biography: who you are, what you value, which tribe you belong to.

That shift was the real invention. Not “PR” as a job title, but PR as a system for manufacturing context—making a thing mean something in public, and then letting the public do the selling.

We still use the same principles. Often without noticing.

The difference is the audience.

In the 1920s, you were persuading people through newspapers, radio, and the emergent machinery of mass culture. In the 2020s, you’re persuading people through systems that interpret the world for them: search, recommendations, feeds, and increasingly, AI summaries.

The symbol moved.

The cigarette was the symbol. Now the snippet is the symbol.

The 1920s insight: people don’t buy the object

The “direct sale” is usually boring because it’s literal.

Literal language is friction. It forces the buyer to do all the meaning-making: Why do I need this? What does it say about me? How does it fit into my life?

PR and modern marketing learned to do that work upstream.

Instead of “here is what the thing does,” the message became:

• Here is the kind of person who chooses this.
• Here is the world where choosing it makes sense.
• Here is the feeling you get to borrow.

That’s why early brand campaigns were never only about features. They were about roles. A product became a prop in a story where the consumer was the protagonist.

This is where marketing and culture started merging.

When a brand “works,” it becomes a compression algorithm for identity:

• A logo becomes an affiliation.
• A tagline becomes a worldview.
• A design language becomes a class signal.

The object is real. But the reason you want it is symbolic.

How that maps to today: people don’t buy the page either

Now apply the same idea to the web.

Most people don’t experience your website as a “site.” They experience it as a rendered verdict:

• a title in a search result
• a two-line description
• a thumbnail in a feed
• a quote in someone else’s newsletter
• an AI Overview that replaces the click

What you publish is not what gets consumed. What gets consumed is what platforms can extract, compress, and re-present.

So if the 1920s lesson was “sell the meaning, not the product,” the 2020s lesson is:

Sell the meaning in a format that machines can recognize and safely repeat.

That’s where AI and modern SEO converge with PR.

PR used to be about access to editorial distribution. Today it’s increasingly about access to interpretive distribution—the layer where systems decide what to surface and how to summarize it.

The snippet as the new symbol

In a SERP, the snippet is not “preview text.” It’s a social object.

It’s the part of your work that becomes portable:

• copied into Slack
• dropped into a deck
• turned into a screenshot
• quoted by a model

And it has a different job than your article.

Your article can be nuanced. It can be slow. It can be contradictory. It can earn trust over time.

The snippet cannot.

The snippet must be:

• legible in isolation
• specific enough to feel like an answer
• neutral enough to be safe
• structured enough to be extracted

This is why AI changes content strategy even when it doesn’t “take your traffic.”

Because it changes which parts of your work are rewarded: the parts that can be turned into reliable, repeatable fragments.

In other words: meaning still wins, but meaning now needs structure.

PR didn’t die. It became system-facing.

A lot of people talk about AI and SEO as if they’re “new.”

They’re not new in the way the 1920s shift was new.

They’re new in the way printing presses were new to pamphlets: the distribution mechanism changes, so the craft adapts.

If you squint, modern “technical SEO” is just the infrastructure layer of PR:

• canonicalization is reputation hygiene
• structured data is self-description for machines
• internal linking is intentional context, not navigation
• topic hubs are editorial architecture

PR has always been about what the environment believes about you.

Search and AI are just environments with stricter rules.

They don’t “understand” the way humans understand. They infer, classify, compress, and score.

So the game becomes: make the correct inference the cheapest inference.

That is the most underappreciated bridge between PR and SEO.

What actually works now (without turning into spam)

If you want your work to survive this shift, the goal is not “write for AI.”

The goal is to write for humans in a way that produces machine-legible meaning.

Here are practical patterns that translate the 1920s insight into 2020s mechanics.

1) Make one idea unavoidable

PR campaigns of the past didn’t try to communicate everything. They tried to make one frame contagious.

Do the same.

Pick a claim that can be repeated without losing its integrity:

• “Visibility is not clicks; it’s being chosen by systems that summarize.”
• “Indexing is a prioritization decision, not a page-quality verdict.”
• “Brand is semantic architecture: what gets associated with you by default.”

Then build the essay around protecting that claim from misreading.

2) Write “extractable” paragraphs on purpose

If a paragraph cannot survive being copied into a snippet, it’s not a paragraph—it’s atmosphere.

You don’t need to write in bullet points.

But you should have segments that behave like:

• definitions
• crisp distinctions
• short models (3–5 parts)
• “if/then” tradeoffs

This is how systems decide your content is safe to reuse.

3) Create a few canonical entry points and treat them like your front page

Old PR understood something brutal: if you can’t control what people remember, you don’t control what they repeat.

On a modern content site, the pages that shape what systems “think your site is” are:

• the homepage
• /start
• your pillar pages
• topic hubs

Everything else is supporting evidence.

If you want the broader model of modern SEO and how visibility now works, start here:

• New here? Start here
• Pillar: Modern SEO in 2026: Visibility, Indexing, and Why Keywords Are Not the Unit

4) Don’t redirect “dead” meaning into “alive” meaning

One of the fastest ways to destroy trust with systems is to pretend the past never happened.

If an old URL no longer matches what you do, a polite redirect to a generic page is not “clean.” It’s a soft lie.

Clear signals age better:

• redirect only when there is a real successor
• otherwise let the old thing be gone

That’s reputation hygiene in machine language.

5) Build clusters like you’re building a worldview, not a blog

PR in the 1920s succeeded because it created coherent cultural objects—frames that showed up across channels and felt consistent.

Topic clusters are the same move, translated:

• one pillar (the map)
• supporting essays (each answers one intent)
• explicit links between them (the structure)

The payoff is not only “SEO.” It’s interpretability.

Systems prefer coherent worlds because coherent worlds are cheaper to summarize.

The uncomfortable conclusion: the brand is back

For a while, the internet trained people to believe brand is decoration.

The performance era fetishized measurability and treated meaning as fluff.

AI is quietly reversing that.

When answers become commoditized, differentiation comes back to:

• framing
• voice
• trust
• taste
• the ability to be quoted without embarrassment

Which is exactly where PR started: making meaning stable enough that other people will carry it for you.

The cigarette was never just a cigarette.

And your content is no longer just a page.

It’s a unit of meaning that must survive compression.