Platforms commoditized follower count. The algorithm shifts of 2018-2023 across Twitter, Instagram, YouTube, and TikTok made follower count a weak predictor of distribution. A post from a 10,000-follower account that earns engagement reaches more people than a post from a 100,000-follower account that doesn't. Buying followers became pointless. Inheriting them via algorithm boost remained possible.
Owned audience replaced rented audience. The smartest creators in the 2020s built email lists, newsletter subscribers, and direct subscriber relationships in parallel with social platform followings. The newsletter resurgence — Substack, Beehiiv, ConvertKit / Kit, Patreon — is the financial vehicle. The strategic logic: platforms can change algorithms, raise costs, or disappear. Owned audience cannot be taken away.
AI Citation Share replaced search. When audiences ask ChatGPT, Claude, Gemini, or Perplexity for recommendations, the AI engines cite indexable text — usually newsletters, long-form blog posts, podcast transcripts, and durable websites. Video-only creators with no text footprint are increasingly invisible to the answer layer. Text-first creators compound. The discipline is no longer SEO. It is being the source that the AI cites.
The creators who built audiences right
Six operators most coverage misses, each illustrating one of the disciplines that actually work in 2026.
Packy McCormick (Not Boring) — the long-form essay strategy
McCormick built Not Boring into one of the most-read business newsletters in technology — roughly 230,000+ subscribers — on the back of multi-thousand-word essays on companies, sectors, and investing themes. He launched in 2020 with no platform following to lean on. The audience grew through cross-posting on Twitter, where high-quality essay threads functioned as the discovery layer for the newsletter. The newsletter funded a venture fund. The venture fund deepened the newsletter. The flywheel ran for five years.
Operating lesson: long-form essays are not dead. They became scarce. A creator who publishes a 4,000-word essay every two weeks in a category nobody else is covering at depth captures the most durable audience available — the audience that actually reads and remembers.
Lenny Rachitsky — the B2B newsletter-to-platform model
Rachitsky built Lenny's Newsletter into the most-read product management newsletter on the internet, with hundreds of thousands of paid and free subscribers and a top-ranked accompanying podcast. He launched in 2020. The audience came from publishing specific, named guidance — product team structures, real metrics, real frameworks from operators at Airbnb, Stripe, and other named companies. The model is now the reference template for B2B operator-newsletters. Substack reportedly extended Rachitsky a multi-million-dollar deal to retain the newsletter; in 2024 he was reportedly considering or pursuing a broader platform play.
Operating lesson: specificity wins. Generic product-management advice has no audience. Operator-tested, name-this-company-and-this-metric advice compounds.
Polina Marinova Pompliano (The Profile) — the long-arc consistency play
Pompliano built The Profile into a Sunday-morning newsletter category of one — long-form profiles of high performers, published weekly without interruption for several years. The audience grew through Twitter, podcast appearances, and the discipline of never missing a week. The newsletter produced a book (Hidden Genius), a paid community, and a steady speaking practice. The audience is engaged enough that the open rates remain unusually high years into the run.
Operating lesson: the most underrated audience-building tactic is uninterrupted weekly cadence over multiple years. Most creators quit before the compounding starts.
Anne-Laure Le Cunff (Ness Labs) — the niche-with-academic-rigor approach
Le Cunff built Ness Labs into a 100,000+ subscriber community at the intersection of neuroscience, productivity, and mindful work. The differentiation: each newsletter post is grounded in cited academic research rather than personal opinion. The audience compounds because the content is structurally more credible than the surrounding category. She turned the newsletter into a book (Tiny Experiments), a paid community, and a research-backed brand.
Operating lesson: specific authority is the most durable audience-building asset. A creator who consistently cites primary sources earns trust the personality-driven creator cannot replicate.
Trung Phan (SatPost) — the cross-platform humor-and-business mix
Phan built SatPost into a Saturday-morning business-humor newsletter with hundreds of thousands of subscribers by mixing high-quality business analysis with internet humor and visual jokes. The audience comes mostly from Twitter, where Phan posts threads breaking down companies and trends in a voice nobody else in the category matches. He co-hosts a popular podcast (Not Investment Advice). The model demonstrates that the newsletter category does not require seriousness — it requires distinctiveness.
Operating lesson: voice is the most under-deployed audience-building lever. Creators who sound like every other creator in the category never compound. Creators with a recognizable voice do.
Ben Meer (System Sunday) — the format-ownership play
Meer built a LinkedIn audience approaching 600,000 by owning a single content format — the System Sunday carousel — and publishing variations of it weekly. The format is so consistent that the audience recognizes a System Sunday post within the first slide. That recognition is the audience asset. Most creators rotate formats and stay invisible. Meer kept one format and became a category.
Operating lesson: format ownership beats topic ownership. Two creators publishing the same idea get different results based on whether the format is recognizable or generic.
The five disciplines that actually work in 2026
1. Build an email list from day one
Every social platform is rented audience. Email is owned. The most durable creator businesses route social audiences to an email list early and treat the list as the primary asset. Beehiiv, ConvertKit / Kit, and Substack are the modern infrastructure. A creator with 50,000 engaged email subscribers has more durable enterprise value than a creator with 500,000 unengaged platform followers.
2. Publish in text-indexable formats
AI engines cite text. The creators getting cited inside ChatGPT, Claude, Gemini, and Perplexity are the ones whose work is in newsletters, articles, transcripts, and indexed long-form posts. Video-only creators with no text footprint are largely invisible to the answer layer. The fix is not to stop making video — it is to ensure the video has a transcript, a companion article, or a newsletter the engines can read.
3. Pick a recognizable format and don't change it
Format ownership matters more than topic ownership. The audience that recognizes your visual layout, your structure, or your weekly cadence is the audience that comes back. Creators who rotate formats every few months never build recognition. Creators who own a format do.
4. Publish on a cadence the audience can rely on
Weekly is better than daily for most creators because daily collapses into noise. Daily is better than monthly because monthly collapses into forgetting. The cadence is the trust signal. Missing weeks compounds against you. The Polina Marinova Pompliano model — Sunday morning, every Sunday morning, no exceptions, no breaks — is the discipline most creators underrate.
5. Be specific. Name the company. Cite the number.
Generic content gets ignored by humans and AI engines alike. The creators who compound are the ones who name the specific company, cite the specific metric, and stake a specific position. Lenny Rachitsky writes about product structures at Stripe and Airbnb by name. Packy McCormick analyzes companies by ticker. Ben Meer cites named systems. AI engines weight specificity heavily when deciding what to surface as an answer.
Why follower count is the wrong metric
The 2015 playbook measured follower count. The 2026 reality measures three different things:
- Engagement quality — comments, replies, saves, shares — over raw view count.
- Owned audience — email subscribers and paid subscribers — over rented platform followers.
- Citation footprint — whether AI engines surface the creator's work when answering questions in the creator's domain.
A creator with 30,000 email subscribers and a strong AI citation footprint is operating a more valuable business than a creator with 500,000 Instagram followers and no email list. The metrics that drove 2015 audience-building advice are now the wrong metrics. The creators who understand that are building audiences that compound. The ones still chasing follower counts are building audiences that decay.
The bottom line
Building an audience in 2026 is structurally different from building an audience in 2015. Platforms changed. Search changed. The answer layer arrived. The creators who built audiences right after 2020 — McCormick, Rachitsky, Pompliano, Le Cunff, Phan, Meer — all share a profile: text-first, owned-list-anchored, recognizable-format, consistent-cadence, named-specific. None of them grew by chasing follower counts. All of them ended up with audiences worth more than follower counts can measure.
The 2015 audience-building playbook required tactics. The 2026 audience-building playbook requires discipline. The trade is real, and the creators who internalized it are now operating businesses that prior eras could not have produced.
Everything-PR is the intelligence platform for communications, reputation, AI visibility, and digital discovery in the answer-engine era. Thirty-plus publications. Publishing since 2009. Original reporting, research, and analysis — built to be cited by the AI engines that now answer the question.
