For the better part of the last decade, online influence has been treated as though it were mostly a question of scale: the person with the larger audience was assumed to have more authority, the person with more views appeared to have more momentum, and the person with more impressions was presumed to have more opportunity. The logic was tidy enough for a full marketing dashboard, which is usually where nuance goes to put on a blazer and die.
But tidy logic has a way of breaking down when actual humans are involved.
My take: the influencer era is ending because we have become suspicious of recommendations that behave way too much like advertising, too much like performance, and too little like genuine human conviction.
The old assumption was that visibility eventually produces trust. I think we are now watching that assumption completely reverse. Visibility may still create awareness, but awareness has become a rather weak currency in a world where everyone is aware of too much already. Trust is what makes visibility matter.
This is inconvenient because trust is not glamorous. Trust is not especially viral. Trust does not wear a headset microphone and pace confidently across a stage telling you that your next level is waiting on the other side of your limiting beliefs. Trust is boring in the way a foundation is boring. Nobody compliments the foundation at a dinner party, but remove it and everyone suddenly becomes an amateur structural engineer.
And right now, the internet is making us all very interested in structural integrity.
The internet has become too polished to believe
Artificial intelligence has made competence cheap.
That sounds harsher than I mean it, but it is difficult to deny. A person can now generate a polished sales page, brand strategy, carousel, email sequence, product description, landing page, and motivational LinkedIn post before finishing a latte. The average quality of digital output is rising quickly, which is useful in obvious ways and unsettling in ways we are only beginning to name.
The unsettling part is not that AI can make things. Humans have always used tools to make things, and the tool panic cycle is one of our most reliable traditions. The unsettling part is that so much of the internet now feels as though it has been passed through the same invisible smoothing machine. Everything is clear. Everything is optimized. Everything has a framework. Everything promises “no fluff,” which is almost always how you know fluff is approaching in a linen suit.
For years, polish worked as a shortcut for legitimacy. A business with beautiful branding seemed more established, a creator with sharp videos seemed more credible, and a professional-looking funnel borrowed authority from its own structure. If something looked expensive, we were more likely to assume it had value.
AI has complicated that arrangement. Polish is no longer proof of taste, depth, effort, experience, or judgment. Sometimes it is simply proof that someone had access to a good tool and typed “make this feel premium.”
This does not mean beauty no longer matters. Beauty matters enormously. Craft matters. Taste matters. Presentation matters. But polish can no longer carry the whole burden of trust. As the internet fills with competent, frictionless, professionally packaged sameness, people begin scanning for the thing the machine cannot quite manufacture: fingerprints.
Caveats, specific details, a weird little observation, or a sentence that has clearly not been approved by nine stakeholders and a compliance department named Brad. A recommendation with enough texture to suggest the person has actually touched the product, lived with it, hated something about it, forgiven it, and decided it was still worth telling you about.
The more perfect the internet gets, the more suspicious perfection becomes.
The influencer became the billboard
The great irony of influencer marketing is that its original power came from not feeling like marketing.
Early influencer culture worked because it collapsed the distance between the commercial and the personal. You were not watching an ad; you were watching someone’s morning routine. You were not being targeted; you were being let in. The mascara was on the bathroom counter. The supplement was beside the kettle. The book was dog-eared. The recommendation appeared to arrive inside a life, not outside it.
Then the machinery arrived, as machinery tends to do whenever something starts working.
The recommendation became a deliverable. The kitchen became a set. The casual “I’ve been loving this” became a line item in a campaign brief. The friend-shaped internet slowly became an advertising channel wearing sweatpants.
This is not an argument against creators being paid. They should be paid. Creative labour is still labour. Audience-building is labour. The ability to make a brand culturally legible inside a messy, fast-moving digital environment is labour. I have no interest in pretending otherwise, mostly because pretending creative labour is not labour is how people end up being paid in exposure, which remains a currency accepted by exactly zero grocery stores.
But there is a difference between monetizing trust and liquidating it.
Audiences can feel the difference. They may not always have the language for it, but they know when affection has been rented for the weekend. They know when “obsessed” has a deadline. They know when someone is reading a caption with the spiritual energy of a hostage note. They know when a product entered the chat.
The problem is not sponsored content. The problem is misalignment.
A genuine recommendation depends on the belief that the person making it had a choice. That they could have said no. That they would have said no if the product were mediocre, irrelevant, overpriced, or secretly terrible. Once an audience stops believing that, influence stops behaving like trust and starts behaving like media inventory.
Media inventory can still be valuable. It can create awareness, reach, and cultural familiarity. It can move numbers, certainly. But it is not the same thing as belief.
And belief is the thing everyone is trying to buy.
The friend remains undefeated
For all our technological sophistication, the most persuasive commercial force on earth may still be one person saying to another, “I think you would like this.”
This is almost embarrassingly simple. It is also the thing modern marketing has spent years trying to manufacture at scale.
We ask friends what to read, what to buy, what to avoid, where to go, which app to use, which supplement helped, which shoes are comfortable, which microphone is worth it, which course was actually useful, which restaurant has not yet been ruined by TikTok, and which jeans will not turn against us emotionally by 2 p.m.
A recommendation from a trusted person does more than transfer information, it transfers confidence. It says, “I have entered this tiny corner of the market, looked around, and returned alive.” That is no small gift in a world where consumers are drowning in reviews, rankings, ads, sponsored lists, affiliate roundups, chatbot summaries, and comparison charts written with suspicious enthusiasm by people who may or may not actually exist.
Nielsen has reported that 88 percent of global respondents trust recommendations from people they know more than any other channel. This should surprise no one and alarm almost everyone in marketing. The industry has spent enormous sums attempting to replicate what the group chat does for free.
The group chat has an advantage no ad platform can fully imitate: consequence.
If I recommend a terrible moisturizer to a stranger, the consequences are abstract. If I recommend it to my sister, I will hear about it. Possibly until one of us leaves this earth. Real relationships create reputational stakes, and reputational stakes are what make recommendations meaningful.
That is why trust is difficult to fake for long. It has memory.
AI did not make humans obsolete. It made humans premium.
There is a lazy version of the AI conversation that assumes machines will replace human judgment because machines can produce answers quickly. The more interesting possibility is that AI increases the value of human judgment precisely because it increases the supply of plausible answers.
When everyone can generate content, the question is no longer who can explain the thing. The question is whose explanation we trust.
When everyone can publish a review, the question is no longer which review exists. The question is which reviewer has earned our confidence.
When every brand can describe itself as thoughtful, values-driven, customer-obsessed, and “rooted in community,” the question becomes who is actually telling the truth and who simply found the correct adjectives in the startup pantry.
This is the context in which trust becomes commercially powerful, not because consumers have suddenly become noble philosophers of discernment (although wouldn’t that be a lovely surprise), but because skepticism has become a survival mechanism.
The public mood is not exactly “sure, I’ll believe whatever the polished person says.” Edelman’s 2026 Trust Barometer describes a world sliding from grievance into insularity, where seven in ten respondents report being unwilling or hesitant to trust someone with different values, backgrounds, approaches to social issues, or information sources. It also notes that trust is increasingly concentrating among those closest to us, including neighbours, family, friends, coworkers, and familiar local voices.
Boom. Maybe read that section again.
People are suspicious. They are tired. They have been over-marketed to, overpromised to, and over-nurtured through funnels that use the word “community” while behaving more like a velvet-rope sales sequence with better fonts.
In that climate, small signals matter. A nuanced review matters. A specific story matters. A person willing to say, “This is good, but not for you,” matters. In fact, that may be one of the clearest markers of trust: the willingness not to sell.
The most trustworthy person in the room is rarely the one recommending everything.
The next influential person may not look like an influencer
This is where the shift becomes particularly interesting for speakers, authors, coaches, consultants, creators, affiliate marketers, and stay-at-home parents. For years, the internet taught people to imagine influence as something performed publicly at scale. Influence was a job, a look, a platform, a category. It required lighting, editing, posting, optimizing, and an apparently inexhaustible willingness to film oneself opening packages.
But influence has always existed outside the performance of influence.
The author who recommends her favourite research tools is influencing. The speaker who tells another speaker which lav mic survived three airports and one emotionally unstable conference ballroom is influencing. The coach who recommends a journal, assessment, book, app, or supplement that genuinely helped a client is influencing. The stay-at-home parent who has become the unofficial neighborhood authority on lunch containers, winter boots, probiotics, grocery delivery, and how to survive a stomach bug without abandoning civilization is absolutely influencing.
None of these people need to become famous to be valuable. In many cases, fame would make them less persuasive.
This is the part the old influencer model struggles to metabolize. We are used to treating audience size as the primary asset, but audience size is only one kind of power. Specificity is another. Context is another. Trust is another. A person with 1,200 followers and a precise, lived understanding of a community may be able to drive more meaningful action than a person with 500,000 followers and the emotional texture of a mall kiosk.
Industry data has been pointing in this direction for a while. HypeAuditor’s 2025 State of Influencer Marketing report notes that nano-influencers account for 87.7 percent of TikTok creators and hold the highest engagement rate in its TikTok creator breakdown. That does not mean large creators are irrelevant. It means scale is no longer the only story.
A celebrity can create awareness. A trusted person can change behaviour.
Those are not the same job.
Affiliate marketing deserves a better publicist
At this point, we have to talk about the phrase that makes everyone’s shoulders migrate toward their ears: affiliate marketing.
I understand why. The term has collected baggage the way a toddler collects rocks: enthusiastically, mysteriously, and from places you did not authorize. For some people, affiliate marketing conjures spammy links, fake scarcity, suspicious income claims, “DM me FREEDOM,” and screenshots from people who seem to spend a great deal of time near rented cars.
But the basic mechanism is not inherently strange. A person recommends something. If someone buys through that recommendation, the person may receive a commission. That is all.
The ethical question is not whether money changes hands. Money changes hands in nearly every form of commerce. The ethical question is whether the recommendation is honest, relevant, transparent, and made with appropriate care for the person receiving it.
There is a grotesque version of affiliate marketing, certainly. It treats people as traffic, products as payout rates, and trust as a resource to be strip-mined before moving on to the next offer. That version deserves its reputation and possibly a small haunted basement of its own.
But there is also a more human version. It sounds like a person saying, “I use this. I trust this. Here is why it helped me. Here is who I do not think it is for. Here is what I tried before. And yes, if you use my link, I may earn a commission.”
That last sentence is not a vibe killer. It is a trust builder. The FTC’s endorsement guidance says material connections between endorsers and marketers should be disclosed, but even apart from regulation, disclosure is simply decent behavior. It tells the person on the other side of the recommendation, “I respect you enough to be clear.”
I think affiliate marketing’s future belongs to people who treat disclosure not as a legal nuisance, but as part of the relationship.
The most underrated business model is borrowing infrastructure
There is another reason this conversation matters, and it has very little to do with influencers.
Starting a product-based business is difficult in ways that motivational entrepreneurship content tends to understate. It is one thing to have an idea for a beautiful supplement, candle, planner, skincare line, children’s product, or household tool. It is quite another to deal with formulation, manufacturing, compliance, packaging, inventory, warehousing, shipping, returns, refunds, customer service, payment processing, quality control, cash flow, and the spiritual devastation of discovering that “just ship it” involves actual shipping.
If your dream is to create a product from scratch, wonderful. Some people absolutely should. The world needs people willing to build difficult things.
But many people do not actually want to become manufacturers, logistics coordinators, fulfillment centres, customer support departments, compliance specialists, and advertising agencies all at once. They want to build income around their expertise, taste, community, and relationships. They want a business model that allows them to contribute what they are actually good at.
That is where affiliate partnerships and brand partnerships become more interesting than they are often given credit for.
A strong brand partner has already built the infrastructure. The product exists. The checkout exists. The operations exist. The shipping exists. The customer support exists. The compliance, packaging, inventory, and payment processing are not waiting for you at 11:43 p.m. with a spreadsheet and a personal vendetta.
What the brand cannot manufacture at scale is your trust.
That is the exchange. They bring infrastructure, you bring context, discernment, and relationship.
Done poorly, this becomes tacky very quickly. Done well, it is one of the cleanest forms of asset-light entrepreneurship available: a way for people to participate in commerce without having to invent, fund, store, ship, and support every product themselves.
EMARKETER projects that affiliate marketing will drive more than $210 billion in U.S. ecommerce sales in 2025. This is not a fringe tactic happening in the darker corners of the internet. It is a major commercial channel. The question is not whether affiliate commerce will grow, the question is whether it will be dominated by faceless coupon sites and commission-chasing content farms, or whether thoughtful, trusted people will learn how to participate with taste.
I am rooting, perhaps stubbornly, for the second future.
Trust has a cost
The catch, of course, is that trust cannot be endlessly monetized without consequences.
A recommendation has to cost you something to mean anything. Not necessarily financially, but reputationally. When you recommend something, you are lending it your name. You are telling your audience, clients, friends, or community that this product has passed through your filter.
If everything passes through your filter, the filter is broken.
This is where many people will get the Recommendation Economy wrong. They will see the opportunity and mistake it for a content strategy. They will create lists, links, roundups, resource pages, storefronts, “favourite things,” and endless soft pitches without doing the slower work of discernment. They will try to monetize trust before they have protected it.
But trust behaves differently than attention. Attention can spike, while trust compounds. Attention can be purchased, while trust has to be earned. Attention can be manipulated for a moment, while trust is what remains after the moment is over.
The most powerful recommendation may not be the one that earns the highest commission. It may be the one where you say, “Don’t buy this yet.” Or, “This worked for me, but I think something else is better for you.” Or, “I know everyone is talking about this, but I found it underwhelming.” These sentences are dreadful for short-term conversion and excellent for long-term credibility.
The future belongs to people who understand the difference.
The Recommendation Economy
This is what I mean by the Recommendation Economy: a shift from monetizing attention to monetizing trust, from performing influence to earning it, from selling harder to recommending better.
It is not new, exactly. Word-of-mouth is ancient. The neighbor, the aunt, the mentor, the colleague, the friend with unusually strong opinions about vacuum cleaners… these people have always shaped buying decisions. What is new is the infrastructure. The internet now allows ordinary trust to travel, track, compound, and, when handled ethically, create income.
That does not mean everyone needs to become an influencer. In fact, I hope most people do not. The world has enough people trying to be broadly appealing to strangers.
What it means is that many people are already sitting on a small, useful, underdeveloped business asset: the fact that people trust their taste.
The better question is not, “How do I get more followers?” The better question is, “What do people already trust me for?”
For a speaker, the answer may be communication tools, books, travel systems, wardrobe staples, and stagecraft. For an author, it may be writing tools, research systems, retreats, editors, and reading lists. For a coach, it may be frameworks, assessments, wellness supports, journals, and learning resources. For a stay-at-home parent, it may be the practical infrastructure of daily life: food, routines, household tools, child development resources, books, supplements, clothing, and sanity.
None of this requires pretending to be someone else. In fact, it works best when you do not.
The Recommendation Economy is not about turning every human interaction into a sales opportunity. That would be dreadful, and I would like to be excluded from that group chat. It is about recognizing that trust, taste, and lived experience already create value, and that people can build ethical, useful income streams around those assets without manufacturing a persona or chasing algorithmic approval.
The future may belong to the person your friend texts you about
The influencer is not disappearing. The celebrity creator will still exist. Massive audiences will still matter in some contexts. Brands will still chase reach because reach is easy to understand and looks lovely in a dashboard.
But I suspect the moat around mass influence is getting weaker. The performance of influence is easier to imitate, easier to automate, and easier to distrust. Meanwhile, the quiet power of specific, human, relational trust is becoming harder to replace.
The next influential person in your life may not be a famous creator, it may just be the person whose opinion you trust. The person who always finds the good book, the useful tool, the better option, the thing that actually works. The person who does not recommend constantly, which is exactly why you listen when they do.
After all the platforms, funnels, automations, launches, pixels, dashboards, and AI-generated content strategies, we may be circling back to the oldest form of marketing in the world: a person you trust saying, “I thought of you.”
Boring? Maybe.
Valuable? Increasingly.