For decades, advertisers have repeated a line attributed to John Wanamaker: “Half the money I spend on advertising is wasted; the trouble is I don’t know which half.”
In the age of social media, that formulation is no longer adequate. The uncomfortable truth is that the wasted portion is not random — and in many categories it is far larger than half. We now know where it sits and why it was never going to work.
The problem is not creative quality, targeting precision, or attribution models.
It is a misunderstanding of how social platforms actually generate power.
Attention is not authority
The past decade trained marketers to treat attention as a proxy for influence. If a campaign reached enough people, generated sufficient engagement, or went viral, it was assumed to have moved the brand.
But platforms do not create the same kind of power.
Some generate Attention.
Some encourage Participation.
A smaller number confer Legitimacy.
Fewer still shape Commitment — the decisions people organise their lives around.
Advertising works when a brand’s objective matches the kind of power a platform can produce. It fails — expensively — when it does not.
Two brands, one system, opposite outcomes
Consider the contrast between PayPal and e.l.f. Cosmetics.
Both are global, digitally native brands with large social footprints. Yet their relationship with social media could not be more different.
e.l.f. Cosmetics thrives on social platforms because its value proposition is participatory by design. Beauty rewards repetition, imitation, and public routine. Social platforms compress discovery, purchase, and display into a single loop. The brand’s growth is accelerated — not distorted — by this dynamic.
PayPal, by contrast, is financial infrastructure. Its power comes from legitimacy and necessity, not expression. People rely on it because it is embedded in checkout flows, compliance systems, and cross-border commerce. They do not choose it for identity expression; they tolerate it because exit is costly.
When e.l.f. spends on social, behaviour compounds.
When PayPal does the same, risk compounds.
This reminder matters, because the failure is often misdiagnosed as a question of execution. In reality, it is a mismatch of physics.
When platforms invert value
On participation-driven platforms such as TikTok and Instagram, compression is rewarded. Complexity is flattened. Emotion travels faster than explanation.
That dynamic suits beauty, fashion, and entertainment. It is hostile to payments, healthcare, insurance, and infrastructure.
For legitimacy-dependent brands, visibility on such platforms does not build confidence; it often surfaces enforcement moments — disputes, freezes, exceptions — and amplifies anxiety. Trust becomes something to be defended rather than accumulated.
This pattern extends well beyond fintech. United Airlines, for example, derives its power from systems: hubs, alliances, loyalty credentials, and corporate booking workflows. Social media does not meaningfully increase trust in those systems. It disproportionately amplifies disruption — delays, outages, service failures — because negativity converts faster than reliability ever can.
The spend appears to work — impressions rise, engagement is healthy — but the underlying emotional ledger moves in the wrong direction.
The problem is not whether platforms create attention. It is whether attention is the kind of power the brand actually needs.
The quiet authority platforms
Now consider where people actually verify, learn, and decide.
They do not do it where content is fastest. They do it where time is allowed to accumulate.
Long-form, archive-based platforms such as YouTube quietly dominate explanation, instruction, and decision-making. They attract less glamour and smaller budgets, yet they are where meaning stabilises.
This is why brands seeking trust, credibility, or long-term decision influence consistently discover — often too late — that authority settles on different platforms than attention.
This asymmetry is striking. Brands overspend on platforms optimised for attention and underspend on those that confer authority. The result is a structurally inverted media mix.
Infrastructure that can — and cannot — be performed
Not all infrastructure brands fail equally on social platforms.
Ford demonstrates the narrow exception. Ford’s core rituals — delivery, modification, work deployment, community replication — are embodied and observable. They can be performed without undermining the brand’s authority because the proof of value is physical and repeatable.
Even here, the balance is fragile. As software, recalls, and service failures become more visible, social amplification can shift from pride to volatility without ever threatening the brand’s underlying dependence.
Luxury shows the opposite danger. Gucci illustrates how visibility can persist even as identity lock and conversion weaken. Social platforms sustain recognition through constant circulation — resale, authentication, dupe discourse, crisis cycles — while quietly eroding exclusivity.
Fame survives; purchase intent fragments.
In other words, social media can keep a brand culturally present while making it harder to sell.
Why the waste persists
Social spend looks productive because platforms are forgiving. Defaults are sticky. Behavioural decay is slow. Damage appears first as fatigue, scepticism, and private routing — not immediate churn.
By the time performance metrics deteriorate, the misallocation has been institutionalised.
The error is not inefficiency. It is category blindness.
Brands that require participation should be social.
Brands that require trust should be procedural.
Brands that require infrastructure should be restrained.
Many are currently trying to be all three — and achieving none.
A new rule of thumb
The old advertising maxim needs updating.
We no longer waste half our social spend.
We waste most of it asking the wrong platforms to do the wrong job.
Social media does not fail brands indiscriminately. It selects. It amplifies those whose value is meant to be performed, and it quietly undermines those whose value must be believed.
The uncomfortable implication is that many large, respectable categories would perform better by being less visibly expressive, not less present.
That is not a fashionable conclusion. But it is increasingly a measurable one.
If Wanamaker were alive today, he would not be asking which half was wasted.
He would be asking why it was spent there at all.
2026 External Signals
- Consumers increasingly treat ads as background noise rather than influence
More than eight in ten U.S. consumers say they try to tune out advertising, and large numbers actively block or avoid ads. This reinforces the argument that visibility does not automatically convert into authority.
Source: Gartner — Gartner Marketing Survey Finds 81% of Consumers Tune Out Ads
Link: https://www.gartner.com/en/newsroom/press-releases/2026-04-13-gartner-marketing-survey-finds-eighty-one-percent-of-consumers-tune-out-ads
(Where to find: ad avoidance and active blocking data) - Trust settles on different platforms than attention
Gartner’s consumer survey found YouTube to be the most trusted social platform for accurate information, ahead of Reddit and well ahead of faster, feed-native platforms. This supports the distinction between attention-rich and authority-rich environments.
Source: Gartner — Gartner Consumer Survey Identifies Most Trusted Social Media Platforms for Information Accuracy
Link: https://www.gartner.com/en/newsroom/press-releases/2025-10-13-gartner-consumer-survey-identifies-most-trusted-social-media-platforms-for-information-accuracy
(Where to find: platform trust rankings) - Messaging and chat environments are increasingly preferred by older working-age adults
Recent global digital data shows that chat and messaging apps are the preferred online destination for many older working-age users, reinforcing the migration of trust and coordination into quieter environments.
Source: We Are Social / GWI — Digital 2026 Mid-Year Global Update Report
Link: https://wearesocial.com/uk/blog/2026/04/digital-2026-mid-year-global-update-report/
(Where to find: sections on chat and messaging app preference by age group)
These signals are consistent with the behavioral patterns observed.
Methodology
This paper is based on behavioral evidence from two locked Fame Index cycles (FY24–FY25). All comparisons are kernel-anchored, reproducible, and HASHLOCK-enforced.
