Your content archive just got a second buyer.

This week, 17 publishers — including the Washington Post and the Associated Press — quietly signed six-figure deals with banks, logistics companies, and communications teams. The buyers aren’t paying to get their names in front of readers. They’re paying to feed publisher journalism into their own internal AI tools. They need a source their systems can actually trust — something reported by real people, specific enough to be useful, with enough depth that it can’t just be swapped out for a web search.

Snowflake built the platform making these deals happen. They take no cut. Publishers and enterprise buyers set their own terms directly. It’s an early market, but the deals are real and the appetite is growing.

At the same time, Axios announced it’s expanding its local news program again — from 30 cities to 43 by year’s end, eventually targeting 100. The funding making this possible is a deal with OpenAI. Axios gives OpenAI access to its content. OpenAI funds the reporters. What’s interesting isn’t the AI partnership — it’s what the AI partnership is built on. Their CRO described the strategy in one phrase: “Premium or die.” No broad paywall. Niche professional audiences in specific cities. An events business growing more than 30% a year. A new C-suite newsletter that hit 4,000 readers before it officially launched.

The content licensing deal is built on that audience model. You can’t license what you don’t have. And what Axios has — deep coverage of a specific reader in a defined geography — is exactly what enterprise buyers are paying for.

One more layer. CEOs told reporters this week they’re hunting for cheaper AI alternatives as costs blow out IT budgets. As the cost of running AI keeps dropping, more content is going to get processed by more systems. Generic content gets summarized and replaced. Specific, original, well-sourced content gets licensed.

Here’s the part I keep coming back to.

The qualities those enterprise buyers are paying six figures for — depth, specificity, a clear voice on a defined subject — are the same qualities that build a loyal reader. The ones who open every issue. The ones who forward it to colleagues. The ones who show up to your events.

I’ve been writing about how the questions media CEOs are asking are changing — away from how big is our audience, toward how well do we actually know them. (A few of those shifts here.) The AI licensing market is asking the same question from a completely different direction, and arriving at the same answer.

The publishers getting paid aren’t the biggest ones. They’re the most specific ones.

One question worth sitting with: if a company had to pay to use your content to power their internal AI tools, would it be worth the check? If yes, you’re building something that compounds in more ways than you might have realized. If the answer is probably not — that’s worth knowing, because it’s the same answer you’d get about your audience relationship.

They’re the same question. They always were.

Sources

• Publishers quietly cut ‘six-figure’ deals via Snowflake’s AI licensing platform — Digiday

• Axios Bets That AI Can Make Local News Pay, One Market at a Time — Adweek

• CEOs go bargain hunting for AI — Axios

Some of that is predictable. AI changed the tenor of every industry conversation. But the more interesting shifts are quieter. They show up in what CEOs are asking about, not what they’re reacting to. Here’s what I’m hearing now that I wasn’t hearing two years ago.

The word “audience” is being used more literally

In 2024, “audience” usually meant reach. Total unique visitors. Email list size. Social following. It was a scale metric dressed up in softer language.

In 2026, CEOs are starting to use the word more precisely. They want to know who their audience actually is — by segment, by behavior, by economic contribution. They’re asking how many of those anonymous visitors are actually known. They’re asking what percentage of their email list is engaged versus dormant. They’re asking who shows up for the flagship newsletter versus the vertical newsletters, and whether those are the same people.

That’s a different question than “how big is our audience.” It’s the question of an operator who has figured out that size without shape is not an asset.

Retention is showing up in CEO conversations for the first time

For most of the last decade, retention lived one or two levels down from the CEO. It was a CMO metric, or an audience ops metric, or a subscriptions team metric. The CEO asked about growth.

That’s changing. More CEOs are asking about churn directly. They want to know the curve, not just the number. They want to know which cohorts disengage in the first 90 days versus the ones that stick for years. They want to know what the leading indicators look like before a subscriber cancels.

The shift makes sense. When acquisition gets harder — and it is getting harder, on every channel — every point of retention is worth more. CEOs who aren’t personally tracking churn by cohort are flying blind on the single most important variable in their long-term economics. A growing number of them have figured that out.

The AI question has matured

In 2024, the AI question from CEOs was usually some version of “what tools should we buy” or “how are other media companies using this.” Vendor-scanning questions.

In 2026, the question I hear more often is “what work should stay human.” Which parts of the editorial operation should be explicitly protected. Which parts of the audience relationship should never be automated. What the brand loses if AI creeps into places it shouldn’t.

That’s a more mature question. The first version assumes AI is a tool you deploy. The second assumes AI is a force you manage.

Board-level conversations now include audience economics

This one surprised me. Two years ago, audience metrics rarely showed up in board packs outside of a marketing appendix. The board talked about revenue, margin, pipeline, maybe a high-level traffic number. Audience was a functional report, not a strategic one.

I’m now seeing audience show up as its own line of discussion. Engaged audience size. Revenue per engaged reader. Retention by cohort. Concentration risk — how much of the audience relationship depends on platforms the company doesn’t own. Some of this is being driven by the fact that attribution itself is getting noisier, which is forcing boards to look at audience in more structural terms than they used to.

The CEOs driving this aren’t doing it because a consultant told them to. They’re doing it because they figured out that their valuation conversation, their capital allocation conversation, and their M&A conversation all need this language. The board members who matter are asking for it.

Nobody is talking about pivots to video anymore

Mostly a joke, but only mostly. The era of strategic pivots driven by platform algorithm changes is fading. CEOs have been burned enough times — by Facebook referral traffic, by pivot-to-video, by Google algorithm shifts — that platform-dependent strategy is out of fashion in the conversations I’m having.

What’s replacing it is a more durable question: what can we build that doesn’t depend on a platform we don’t control. Owned distribution. Direct audience relationships. First-party data. The language is less exciting. The underlying business is more defensible.

The pattern across all five shifts is the same. The industry conversation is maturing faster than the industry’s operating models are. CEOs are asking better questions than their reporting systems can answer. They know their audience is more valuable than their dashboards suggest, but they don’t yet have the instrumentation to prove it.

That gap is where the next few years of advantage gets built. Not in another tool, another channel, another pivot. In finally building the operating picture that lets a CEO talk about their audience the way they talk about the rest of the business — specifically, economically, and with the math to back it up.

Most newsletters go paid too early.

Not because the content isn’t good enough. Because the math isn’t ready.

There’s a moment in every publisher’s arc where the idea of paid subscriptions starts to feel inevitable. You’ve built an audience. You’re getting replies. People are telling you your work is “worth paying for.” And there’s a platform making it very easy to flip the switch — complete with a checkout page and a dashboard full of revenue projections.

So you do it. You launch paid. And here’s what usually happens: a small percentage of your free list converts. The growth you were seeing on the free side slows down because you’re now gating content. Sponsorship conversations get awkward because your open list is suddenly smaller. And three months later, you’re making $800/month and wondering whether you should reverse the whole thing.

This isn’t a knock on paid newsletters. Paid subscriptions can be an excellent business. But the timing of the paywall decision — and the structure behind it — matters more than most operators realize.

The real question isn’t “should I go paid?” — it’s “what am I optimizing for?”

When you introduce a paywall, you’re making a tradeoff that most people don’t fully price in.

On the plus side, you generate direct revenue from readers. You reduce your dependency on advertisers. You get a clearer signal about what your audience truly values. Paying subscribers tend to be more engaged, more loyal, and more vocal advocates.

On the other side of the ledger, you’re shrinking your addressable audience. Free growth slows because your best content — the stuff that gets shared, forwarded, and talked about — is now behind a wall. Sponsorship revenue takes a hit because advertisers pay for reach, and your free list just got smaller. And you’ve introduced a retention problem you didn’t have before: now you need people to actively decide, every month or every year, that your work is worth renewing.

Neither side is wrong. But most operators evaluate this decision based on excitement rather than arithmetic. “People said they’d pay” is not a financial model.

The conversion math most people skip

Here’s the uncomfortable truth about free-to-paid conversion: industry benchmarks sit between 2% and 10%, depending on the niche, the audience relationship, and the content format. Most operators land closer to the low end than the high end.

That means if you have 5,000 free subscribers and you launch a $10/month paid tier, you’re probably looking at 100 to 300 paying subscribers at launch. That’s $1,000 to $3,000/month in gross revenue, before platform fees. Not bad — but probably not enough to replace whatever revenue you were generating (or could generate) through sponsorships, consulting, or other models.

The operators who do well with paid subscriptions tend to share a few traits. Their content is niche enough that readers can’t easily find it elsewhere. Their audience already has a professional or financial reason to stay informed on the topic. And — this is the one people underestimate — they had a large enough free base that even a modest conversion rate produced meaningful revenue.

Going paid with 2,000 subscribers and a 5% conversion rate gives you 100 paying readers. Going paid with 50,000 subscribers and the same 5% gives you 2,500. The strategy is identical. The outcome is not.

Five signals that suggest you’re ready

Instead of asking “can I go paid?” — which almost always has a technical yes — ask whether these signals are present:

Your audience is asking for depth you can’t give away for free. Not “your stuff is great, I’d pay for it” (everyone says that). But specific requests for deeper analysis, exclusive data, direct access, or formats that require more production effort. The demand should feel like pull, not flattery.

You have a free audience large enough to absorb the conversion cliff. There’s no magic number, but a useful gut check: if 3-5% of your free list converted tomorrow, would that revenue be meaningful to you? If the answer is “not really,” you’re not there yet. Keep growing the free list.

Your content has a clear “who” and “why.” Paid works when readers can articulate what they’re getting and why it matters to their work or life. Generalist newsletters with broad appeal struggle with paid because the value proposition is harder to pin down. The more specific your niche, the more defensible a paywall becomes.

You’ve already monetized the free list in other ways. This might sound counterintuitive, but publishers who’ve run sponsorships, launched a product, or sold services to their audience tend to have a much better read on what their subscribers will actually pay for. You’ve already tested the relationship commercially. A paywall is just a different transaction.

You have a plan for what stays free. The worst version of going paid is gating everything and watching your growth engine die. The best operators treat free content as the marketing arm and paid content as the product. That distinction should be clear before you flip the switch — not something you figure out on the fly.

The hybrid model most operators underuse

Here’s what often works better than a hard paywall, especially in the early and mid stages: a hybrid approach where most content stays free and the paid tier delivers something genuinely different.

Not “the same newsletter but with a bonus paragraph.” Genuinely different. Think: raw data sets, operational templates, private community access, extended interviews, or weekly analysis that would be impractical to give away.

The reason this works is simple. Your free content keeps the growth flywheel spinning — new subscribers, shares, forwards, search visibility. Your paid tier captures the segment of your audience that wants more and is willing to pay for it. And your sponsorship revenue stays intact because your free list is still large and growing.

This is harder to execute than a simple paywall because you’re running two products, not one. But it sidesteps the biggest risk of going paid: killing the free growth that got you here.

What the math actually looks like

Let’s be concrete. Imagine a newsletter with 20,000 free subscribers. It’s generating $3,000/month in sponsorship revenue and growing at 8% month-over-month.

Scenario A: Hard paywall at $10/month. You convert 5% of your list on day one. That’s 1,000 paid subscribers, or $10,000/month before fees. Solid. But your free list drops to 19,000 (minus the converters). Growth slows to 3% because your best content is gated. Sponsorships get renegotiated down because your open audience shrank. Twelve months later, you’ve got maybe 1,200 paid subscribers (churn offsets new conversions) and $12,000/month — but your free list has stalled at 22,000. Total revenue with reduced sponsorships: maybe $13,000-$14,000/month.

Scenario B: Hybrid model at $10/month. Same 5% launch conversion — 1,000 paid subs, $10,000/month. But the free list keeps growing at 7-8% because your free content is still out there working. Sponsorships hold at $3,000/month because your free reach is unaffected. Twelve months later, your free list is at 38,000. Paid subs are at 1,500 (ongoing conversion from a bigger funnel). Total revenue: $18,000/month and growing.

These are simplified numbers. But the directional logic holds. The operators who protect their free growth while layering in paid tend to build bigger businesses over time than those who gate everything and hope the content sells itself.

When the answer is “not yet” — and what to do instead

If you’re reading this and realizing you’re not ready for paid, that’s not a failure. That’s clarity.

Here’s what to focus on instead: grow the free list aggressively. Get to a scale where the conversion math works in your favor. Monetize through sponsorships, consulting, events, or affiliate partnerships — models that reward reach rather than restrict it. Build the relationship so that when you do introduce a paid tier, the conversion rate is higher because trust is deeper.

The paywall will still be there in six months. Your window to grow the free audience at the rate you’re growing it now might not be.

The bottom line

Going paid is not a milestone. It’s a business model decision with real tradeoffs — and the timing matters as much as the decision itself.

The publishers who get this right tend to share one trait: they treat the paywall as a financial question, not an emotional one. They know their numbers. They’ve stress-tested the conversion math. They’ve figured out what stays free and why. And they’ve built enough of a free audience that the economics actually pencil out.

If you can do that, a paid tier can transform your business. If you can’t — at least not yet — the best move is to keep building the machine that will make it work when the time comes.

The cliff is real. But you don’t have to jump off it.

Most publishers are still running their ad business on an assumption that stopped being reliable two years ago: that recorded actions in the inbox reflect human intent.

They don’t. Not consistently. And the monetization models built on that assumption are already adjusting — even if most teams haven’t named what’s happening yet.

The measurement layer got noisier, and nobody recalibrated

The problem isn’t that attribution broke overnight. It degraded gradually, and the dashboards kept populating the whole time.

Inbox automation, security scanners, link pre-fetching, privacy proxies, and bot filtering all generate activity before a human ever sees a message. Apple’s Mail Privacy Protection alone distorted open rates for roughly half of most publisher audiences when it rolled out. Click-level data followed a similar path — B2B publishers in particular have reported bot-driven click inflation north of 40% in some segments.

None of this produces obviously bad data. It produces ambiguous data that still looks precise. Reports still get generated. Campaigns still show results. But the signal-to-noise ratio shifted, and most teams are optimizing against numbers that no longer track actual human behavior.

That drift is harder to catch than a sudden drop. Performance doesn’t vanish. It just stops compounding the way it used to. Creative tests stop converging. Audience segments get noisier. CAC gets harder to defend. And marketing teams spend more time explaining the numbers than learning from them.

Performance channels feel the pressure first

Performance marketing runs on tight feedback loops. Test, measure, adjust, repeat. When the measurement layer degrades, those loops stop sharpening. They start wandering.

This isn’t a failure of execution. The system stopped giving clean answers.

And here’s where it gets uncomfortable for publishers: when an advertiser’s confidence in attribution drops, their confidence in the channel drops with it. That doesn’t mean they stop spending. But it changes where they spend and what they expect in return.

How ad dollars are actually moving

The shift toward brand, events, research products, and community-based programs didn’t happen because performance advertising stopped working. It happened because those formats depend less on fragile click-level proof.

Let’s be concrete.

Brand programs operate on memory and repetition. They don’t need a click to justify the spend — they need consistent presence across a buying cycle. As B2B purchasing committees have expanded (Gartner’s been putting the number at six to ten stakeholders for years now) and buying timelines have stretched, brand presence across a longer window holds more weight than a single attributed conversion.

Events produce outcomes you can actually observe. Attendance is human. Conversations happen in person. Pipeline gets generated in a room, not in a funnel report. Sponsorship value is easier for a sales team to articulate because the proof isn’t buried in a dashboard — it’s sitting at the table.

Research and proprietary data deliver utility that’s hard to summarize away. Original analysis and benchmarks are harder to replace and easier to connect to value. When an advertiser sponsors a report that their prospects actually reference in meetings, the attribution question becomes less important. The value is self-evident.

Paid communities compound over time rather than spiking on launch day. Their value comes from ongoing interaction and natural qualification — not from a single tracked action. They’re harder to build but harder to commoditize, and advertisers are starting to notice.

None of this means performance advertising is dead. But the share of the pie that depends on clean click-level attribution is getting smaller, and the share that depends on trust, context, and repeated presence is getting larger.

What the more stable publishers have in common

The organizations navigating this well share a few operational traits worth paying attention to.

They have a defined audience with a clear identity — not “business professionals” but a specific role, problem, or industry vertical. They produce original insight rather than repackaged news. They can explain their value to an advertiser without leaning on a dashboard. And they’re willing to educate their ad partners on what good measurement looks like now, rather than just reporting last month’s numbers and hoping nobody asks hard questions.

Their strength isn’t better attribution. It’s that their value is legible even when attribution is imperfect.

That distinction matters more than most teams realize.

Measurement isn’t gone — it’s shifting from proof to confidence

Attribution still exists. But it’s becoming less about proving causality and more about building directional confidence.

The signals that carry more weight now: lead quality over volume. Depth of engagement over surface activity. Repeat exposure over a longer window rather than a single interaction. Brand recall. Sales influence measured in quarters, not days.

This isn’t a retreat from rigor. It’s alignment with how B2B decisions actually get made. Nobody signs a six-figure contract because they clicked a link in a newsletter once. They sign because they’ve seen your brand, trusted your content, and recognized the name when it came up in a meeting.

If your measurement framework doesn’t reflect that reality, you’re optimizing for a version of the buying process that doesn’t exist anymore.

What this means for your operation

If your revenue model only makes sense when attribution is clean, you’re exposed. That doesn’t mean throwing out measurement. It means being honest about what measurement can and can’t tell you right now.

Operationally, that means a few things.

Align your editorial, revenue, and data teams around outcomes that extend beyond inbox metrics. Build ad products that hold value even when clicks don’t line up cleanly. Invest in formats — research, events, community — where the value is observable without depending on a fragile tracking chain. And have honest conversations with your advertisers about what’s changed, because they’re dealing with the same degradation on their side.

Attribution didn’t break because the industry made mistakes. It broke because the environment changed — automation increased, privacy hardened, discovery fragmented, and buying behavior got more complex.

The publishers who come out of this well won’t be the ones with the most sophisticated dashboards. They’ll be the ones whose value is obvious even when the numbers are noisy.

Having an email address is not the same as owning an audience. The media industry has conflated the two for a decade, and the gap between that assumption and reality is getting wider.

The belief made sense when it formed. Facebook throttled organic reach. Video algorithms flipped incentives overnight. Search traffic became volatile, then opaque. Email felt like solid ground — direct, reliable, portable. Compared to social and search, it felt owned.

“Owned versus rented” became a useful shorthand. Email was owned. Everything else was rented. And for a stretch, that model held. If someone handed you their email address, you could show up in their inbox whenever you wanted. That was power.

Then the environment shifted. Quietly at first. Then all at once.

Why email ever felt like ownership

The industry didn’t arrive here by accident.

As platforms became less predictable, email offered a compelling alternative. Direct delivery. Consistent reach. A sense of permanence. Independence from algorithmic feeds.

If you had the address, you had the relationship. That was the operating assumption.

For years, it was close enough to true that nobody questioned it.

The inbox is no longer neutral territory

Email is no longer a neutral pipe between publisher and reader. Inbox providers have reinserted themselves into the relationship — aggressively.

Today, platforms control things that materially affect outcomes. Gmail sorts newsletters into tabs and decides when to surface an auto-unsubscribe prompt. Apple Mail’s privacy protection masks whether someone actually opened a message. Security filters and bot pre-fetching generate fake clicks that inflate engagement data. And AI-generated summaries now rewrite your content before it reaches the reader.

All of this happens after you’ve “acquired” the subscriber.

You may technically have the email address. You do not control whether the message is seen, how it’s framed, or whether the engagement signals you’re tracking represent real human intent.

That creates both risk and opportunity. The risk is obvious — your distribution channel is less reliable than you think. The opportunity is that publishers who recognize this early can build something more durable than inbox access.

AI summaries are the clearest stress test

AI inbox summaries expose the illusion most directly.

Apple Intelligence rewrites subject lines. Gemini summarizes newsletters inside Gmail before the reader ever opens them. When that happens, links disappear. Calls to action vanish. Nuance collapses into a few generated sentences. The platform decides what matters.

For many media businesses, the newsletter isn’t the product. It’s the front door. The value happens after the click — articles, events, research, subscriptions, leads, community. AI summaries interrupt that entire model.

Here’s the uncomfortable truth: if your content can be summarized by a machine without meaningful loss, the platform has more control over the relationship than you do. If the reader gets sufficient value from the summary and never clicks through, the balance of power has already shifted.

The operational question this raises is specific: how much of your newsletter’s value depends on the reader actually visiting your site? If the answer is “most of it,” then your distribution strategy has a single point of failure you may not be tracking.

A more useful definition of ownership

The old definition of audience ownership was transactional. I have the email address.

The modern definition is behavioral. This audience seeks me out, regardless of channel, format, or platform.

Those are very different things.

True ownership shows up in signals that are harder to fake. Readers notice when you don’t publish. They search for you by name, not just the topic. They follow you across platforms. They show up to events. They pay. They recommend you to peers without being asked.

A publisher with 10,000 subscribers who exhibit those behaviors has a more durable business than one with 200,000 subscribers who passively receive a newsletter they rarely open.

Ownership is not distribution control. It’s relationship depth and identity.

Why creator-led media keeps winning this fight

Creator-led media makes this visible — and the gap is widening.

Audiences don’t follow creators because of where they publish. They follow because they trust the taste, value the perspective, and feel a human connection. The best creator businesses have figured out something traditional publishers still struggle with: the voice is the brand, and the brand travels.

That portability is real ownership. Email to social. Social to podcasts. Podcasts to video. Video to events. The format changes. The loyalty doesn’t.

This isn’t a knock on traditional publishers. But it surfaces an important pattern: audiences bond with voices, not formats. With perspective, not infrastructure. Publishers that hide behind logos and channels tend to struggle here — not because they’re large, but because they’re interchangeable.

The practical implication is worth stating directly. If your newsletter could be written by anyone on your team and the reader wouldn’t notice, you have a distribution asset, not an owned audience.

Clicks can’t prove ownership anymore

The metrics most publishers rely on to validate audience ownership are the least reliable they’ve ever been.

Open rates have been compromised since Apple’s Mail Privacy Protection launched in 2021. But the problem has deepened. Click data is polluted by security bots, link pre-fetching, and privacy proxies. A meaningful percentage of “engagement” in most email programs does not represent a human making a decision.

If ownership can’t be validated in the inbox, it has to be measured further down the funnel. The signals that actually indicate an owned audience are harder to collect but harder to fake: repeat visitation patterns, event attendance, community participation, willingness to pay, brand recall, and preference under pressure.

Let’s be concrete. If a reader is choosing between your newsletter and a competitor’s, and they open yours first — that’s a signal. If they attend your webinar without a reminder email — that’s a signal. If they renew a paid subscription without a win-back campaign — that’s a signal.

You don’t own an audience because they clicked once. You own it when they come back on purpose.

What real ownership looks like in practice

Audiences that are genuinely owned tend to share a few characteristics.

They’re multi-format. Email is a touchpoint, not the entire strategy. They exhibit high trust — reflected in low churn and engagement that extends beyond the inbox. They have a clear identity: this is for people like me. They deliver original value that’s difficult to summarize or replace. And they operate as a mutual exchange — readers get insight and utility, publishers earn loyalty and permission.

These audiences are almost always smaller than vanity metrics suggest. They are also dramatically more resilient. They survive platform changes, algorithm shifts, and inbox filtering because the relationship doesn’t depend on any single channel.

The practical test: if Gmail moved your newsletter to a new tab tomorrow and open rates dropped 40%, would your readers come find you? If the answer is yes, you have something. If the answer is “probably not,” the ownership was always more fragile than it appeared.

What this means for media operators

This is not an argument that email is dead. Email remains one of the best tools available.

It’s a warning against treating newsletters as the destination instead of the connective tissue.

The operators who get this right tend to share a few traits. They build brands, not just lists. They invest in original insight and point of view that can’t be easily replicated by AI. They design for habit and trust, not reach alone. They educate advertisers on outcomes — repeat engagement, conversion, brand lift — rather than inbox metrics. And they treat email as one layer in a broader relationship architecture, not the whole thing.

Email is no longer a moat by itself. It’s a channel. The moat is the relationship.

Ownership is earned, not claimed

Platforms will keep changing the rules. AI will keep abstracting distribution. Metrics will keep getting noisier.

In that environment, the only audience you truly own is the one that would come find you if every platform disappeared tomorrow.

Everything else is just access.

Every major distribution channel in media has followed the same arc: open access, rapid adoption, growing dependency, then a rule change that shifts the economics in the platform’s favor. Substack is not exempt from this pattern.

That doesn’t mean abandoning it. I still use Substack as a distribution channel — the same way I post to LinkedIn, try to rank in search, or experiment with other platforms. But media operators who are building their business on Substack should understand what they’re signing up for, and what happens when the terms change.

Note from Tony, a couple weeks after the original post: I came out hot on this one. With some time, distance, and a few spirited debates with people I trust, I’ve softened my stance. I’d still exercise caution if paid subscriptions are your primary goal — now or in the future — but as a distribution platform, Substack still makes sense. The lack of APIs and Zapier integrations remains a real obstacle for a segment of the creator world.

We’ve seen this arc before

If you were running media on Facebook in the early 2010s, you remember what free distribution felt like. Publish a post, and your audience saw it — all of them. Until they didn’t. Organic reach collapsed. The algorithm prioritized friends and family. Publishers had to pay to reach the people they’d spent years attracting.

Search followed the same trajectory. Organic SEO used to be a reliable growth engine: publish quality content, rank, earn traffic, capture data. Today, AI-generated answers and zero-click search mean the clicks never make it to your site, and the behavioral signals that once powered audience strategy are hidden behind Google’s walls.

Substack’s app is moving in the same direction. Discovery, engagement, and billing are shifting inside their controlled environment. Your readers aren’t just yours anymore. They’re also theirs.

The pattern is consistent: platforms open up to attract creators, then close down to capture value. The timing varies. The outcome doesn’t.

The real cost of convenience

Substack’s iOS integration looks like progress on the surface: easier payments, simpler onboarding, and the chance to capture subscribers who never would have gone through your website.

Here’s what changes under the hood.

If a reader subscribes through the iOS app, Apple takes 30% of the revenue. That transaction isn’t tied to your Stripe account — it’s tied to the Substack-to-Apple billing chain, which means you don’t own the payment relationship. If you leave, that data doesn’t come with you.

To offset the Apple cut, Substack has said it will raise subscription prices for those readers so your net payout stays the same. The math works for the creator — until you realize your subscribers are paying more just so you can earn the same amount. That’s a pricing decision being made for you, not by you.

For a creator with 500 paid subscribers, the convenience may outweigh the tradeoff. For an operator with 5,000 or 50,000, you’re ceding both margin and strategic leverage. And once your billing relationship runs through someone else’s ecosystem, switching costs compound quickly.

The control illusion

The hardest truth for media operators is that no channel is fully under your control. Even the ones that feel like they are.

Email is the most durable channel we have because you can export and migrate your list. Unlike Facebook or Substack, you can take your data and leave. But deliverability is never guaranteed. Spam filters, inbox algorithms, privacy changes, and AI summaries all play gatekeeper. Opting in doesn’t mean your reader sees your newsletter every time.

That doesn’t make email less powerful. It means audience ownership isn’t binary — it’s a spectrum. You reduce risk by moving closer to full ownership, but you never eliminate it entirely.

The practical question isn’t “do I own this channel?” It’s “if this channel changed its rules tomorrow, how much of my business would break?” The answer to that question should drive your infrastructure decisions.

What to do with this information

If platform risk is unavoidable — and it is — the response isn’t retreat. It’s building a foundation that doesn’t depend on any single channel.

Start with your data. Build experiences that generate first-party signals you can actually use: preferences, behaviors, purchase intent, content engagement patterns. The more you know about your audience directly, the less you depend on platforms to tell you who your readers are and what they want. A publisher who knows that 12% of their audience is actively researching a specific topic has leverage that no platform can take away.

Then look at redundancy. If your entire audience relationship runs through one channel — whether that’s Substack, email, or anything else — you have a single point of failure. Pair email with your site, events, podcasts, or community. Each touchpoint reduces the damage when someone else’s algorithm shifts.

Treat platforms as distribution, not destination. Substack, LinkedIn, and every other platform can be useful for reach. Use them for what they are — amplification — but always create a path back to properties you control. The reader who finds you on Substack and ends up on your site, in your community, or at your event is worth more than the one who stays inside someone else’s app.

And assume the rules will change. They always do. Build optionality into your model so you can adapt when the incentives shift — because the incentives will shift.

This isn’t really about Substack

Platforms will always evolve in ways that maximize their own growth and revenue. That’s not malice. It’s business. For a long time, Substack and its creators sat side by side. Now the platform has gotten up and moved across the table.

The question for operators is whether you’re building with that reality in mind. Are you treating platforms as one layer in a broader strategy, or are you betting your model on rules you don’t write?

The audience you can reach through someone else’s platform is useful. The audience that would follow you if that platform disappeared is yours.

Those are different things, and the gap between them is where the risk lives.

Audience development roles are being cut, consolidated, or deprioritized across the media industry — and most of the people affected didn’t see it coming.

This isn’t just a budget cycle. It’s an identity crisis for the function. Executives are scrutinizing every team. Growth is flattening. First-party data strategies aren’t delivering fast enough. And somewhere along the way, audience development lost its seat at the strategy table — if it ever had one.

The metaphor that keeps coming to mind is nautical. Calm seas make it easy to forget how much skill it takes to steer a ship. You have time to test new sails, take scenic routes, and invest in long-term improvements. But when the weather turns, the captain’s job changes. Every decision has higher stakes. Priorities shift from exploration to survival.

For many audience professionals, the skies are already darkening.

Editor’s note: This post is not a silver bullet. I know many smart, capable audience professionals who are being impacted by decisions far beyond their control — roles eliminated not because of performance, but because of shifting priorities, restructures, or misaligned expectations. The goal here is to share a practical framework for those still in the fight: a way to reframe your work, articulate your value, and push for clarity in a moment of confusion. Every company is different. Every org chart has its own politics. But if this helps an audience lead reset their position and protect their role, it’s worth sharing.

The reset is real

A year ago, media operators were debating whether they needed a Chief Audience Officer. Now they’re questioning core audience job responsibilities entirely.

The signs are hard to miss. Audience leads replaced with junior CRM or lifecycle marketers. Merged audience-and-marketing orgs with no clear swim lanes. Leadership teams unable to explain what audience development actually does. Budgets shifting to demand gen or paid media because attribution is easier to point at.

The job hasn’t disappeared. But the definition of the job is collapsing under outdated expectations. In calm seas, audience development was about growing the list, improving deliverability, hitting engagement benchmarks, and supporting editorial or marketing goals. There was room to experiment. Time to build infrastructure. Cover to chase longer-term outcomes.

In stormy seas, none of that is enough. You need to prove that audience work drives monetization, reduces waste, accelerates conversion, and protects the brand’s future value. And you need to prove it fast.

This doesn’t mean abandoning the mission. It means communicating your value in terms the business understands right now.

Play 1: Redraw your charter and take ownership

Too many audience teams have been reduced to execution arms. “Can you build this segment?” “Can you resend the email?” “Can you launch a new newsletter?” You can’t earn a strategic seat if you’re treated like a support desk.

The reset starts with redrawing what you own. A simple model that works: three audience motions. Grow — identify and capture the right users, qualified, consented, and reachable. Activate — increase engagement and surface high-intent signals. Convert — turn audience action into revenue or strategic outcomes.

Every project, request, or initiative should fall into one of those three categories. If it doesn’t, ask why you’re doing it. This isn’t just an internal organizing principle — it’s a communication tool. When leadership asks what your team does, you should be able to answer in one sentence tied to one of those motions.

Play 2: Make revenue tie-backs non-negotiable

In calm seas, audience work could hide in aggregate metrics. “List grew by 12%.” “Open rates are up.” That doesn’t cut it anymore.

The move here is making revenue attribution part of your roadmap — even if it’s directional at first. What percentage of qualified leads came through known audience channels? Which audience segments convert at two or three times the baseline in campaigns? Do activated subscribers show higher lifetime value or lower churn than the broader base? Are sponsor campaigns powered by first-party segments outperforming site-wide traffic?

You don’t need perfect attribution to start. You need enough signal to draw a line between audience work and dollars. If you can show that a specific re-engagement campaign recovered $40,000 in at-risk subscription revenue, that’s a conversation leadership will pay attention to. If you can show that sponsor campaigns targeting first-party behavioral segments deliver 2.5x the click-through rate of run-of-site, that changes how the sales team thinks about your work.

If you’re not showing how audience drives revenue, you’re leaving the door open to be cut.

Play 3: Productize the signal you already have

Most audience teams are sitting on more data than they realize — clicks, replies, downloads, time-on-site, form fills, unsubscribe reasons, poll answers. The problem isn’t the data. It’s that nobody is turning it into something usable.

The play is straightforward: turn behavior into signal, turn signal into segments, turn segments into product.

Let’s be concrete. If you track topic-level engagement across newsletters and site content, you can build segments that map to advertiser verticals — “readers actively researching marketing automation” or “CFOs engaging with fintech content three or more times in 30 days.” Those segments are sellable. They’re also defensible in a way that “we have 50,000 email subscribers” is not.

Build “next action” journeys tied to content depth or recency. Develop internal dashboards that surface high-value audience activity for sales, editorial, and product teams. The goal is making your audience intelligence actionable across the organization — not just within your team.

Play 4: Build a real roadmap and defend it

A lot of audience teams are stuck in request mode. You get pulled into every campaign, launch, or fire drill. The real strategic work never gets done because the urgent work never stops.

The fix is a roadmap that maps to business outcomes, not just team activity. Phase one focuses on growth — adding net-new qualified audience, measured by the percentage of known users that match your ideal customer profile. Phase two focuses on activation — increasing engagement depth, measured by click-through rates, time-on-site, and repeat visits. Phase three focuses on conversion — driving monetization, measured by conversion rate and influenced revenue.

You can still say yes when other teams need help. But now you have a plan that justifies your team’s existence and a framework for evaluating whether ad hoc requests are worth the tradeoff. When someone asks you to drop everything for a one-off email blast, you can point to the roadmap and have an honest conversation about what gets delayed.

Play 5: Control your internal narrative

This might be the most important play, and it’s the one most audience leaders skip.

Audience development is often misunderstood because the people doing the work assume everyone else “gets it.” They don’t. If you’re not telling your own story internally, someone else is telling it for you — and they probably think you’re “the email person.”

Send a monthly internal update with KPIs and wins, framed in business terms. Present at go-to-market or executive meetings — not just data, but the story behind the data. Align your success metrics to other teams’ outcomes so your work shows up in their results, not just yours. Make your contribution impossible to ignore.

The audience leaders who survive stormy seas aren’t necessarily the ones doing the best work. They’re the ones whose leadership understands the work well enough to protect it.

Play 6: Rethink your team structure

The generalist “audience coordinator” role isn’t built for this environment. You need operators who can work cross-functionally, analyze signal and turn it into strategy, build infrastructure that scales, and speak both editorial and commercial fluently.

You may not get headcount. But you can cross-train, prioritize ruthlessly, and clarify ownership. Who owns re-engagement? Journey design? Segment prioritization? If the answer is “no one,” that’s where you start.

The team that can demonstrate it operates like a revenue function — not a service desk — is the team that keeps its budget.

This isn’t a crisis — it’s a crossroads

The audience development job hasn’t disappeared. It’s evolved. And the pressure isn’t unique to this function — every part of the media business is being scrutinized.

The difference is that audience teams are better positioned than most to close the gaps. You sit between product, sales, editorial, and marketing. You hold the keys to first-party data. You know what the audience actually wants — not what the algorithm says they want.

But that positioning only matters if you step up and lead. Redraw your charter. Tie your work to revenue. Productize your signal. Build a roadmap you can defend. Own your narrative. Structure your team for what the business needs right now.

The audience leaders who come through this intact won’t be the ones who waited for permission. They’ll be the ones who set the terms.

Most audience budget requests fail for the same reason: they sound like expenses, not investments. The team asks for tools, headcount, and infrastructure. Leadership hears cost. The request gets trimmed or shelved — not because the work doesn’t matter, but because the pitch didn’t connect it to the problems the business is trying to solve.

This is fixable. The gap isn’t in what audience teams need. It’s in how they frame the ask.

Why this budget cycle is different

Timing matters. The macro environment has shifted in ways that make audience infrastructure harder to deprioritize — if you frame it correctly.

Platform dependency is now a measurable liability. Google’s AI Overviews are changing search behavior. Social platforms favor short-form video that many publishers can’t produce profitably. The publishers who weather the next algorithm change will be those who own their audience relationships and don’t depend on a single channel for discovery.

First-party data has moved from “nice to have” to regulatory requirement. With privacy laws tightening globally and third-party cookies disappearing, your first-party data strategy determines whether you can run targeted ads, optimize content, or even understand who your audience is. Publishers without a coherent first-party data approach are already feeling the revenue impact.

Reader behavior is fragmenting faster than most teams can track. Your audience reads newsletters on mobile, listens to podcasts during commutes, attends virtual events, and engages on social platforms — often in the same week. The organizations that can connect those touchpoints and act on the patterns will outperform those still measuring channel by channel.

Frame your budget request around these business realities, not around what your team needs day to day. Your leadership is thinking about risk, revenue, and competitive position. Meet them there.

How to speak CFO language

The difference between a budget request that gets approved and one that gets shelved usually comes down to framing. A few examples of what that looks like in practice.

Instead of “we need more tools for our email list,” try: “Our current platform caps segmentation at a level that limits subscriber growth. Moving to a system that supports behavioral segmentation and triggered automation would let us grow from 50,000 to 150,000 active subscribers within 18 months. Based on our current revenue per subscriber of $9, that’s roughly $900,000 in incremental annual value.”

Instead of “we should improve audience analytics,” try: “We currently can’t attribute content performance to downstream revenue. That gap means we’re making editorial and marketing investment decisions without knowing what’s working. Closing that gap — even directionally — would let us reallocate spend toward what’s actually driving results.”

Instead of “we need more staff for audience development,” try: “Our current team can maintain existing programs but can’t build the re-engagement and lifecycle infrastructure that would reduce churn. Adding two roles focused on retention and activation would pay for themselves within a year if they move our churn rate from 8% to 6% — which represents roughly $200,000 in preserved annual revenue.”

The pattern: every request ties directly to revenue, efficiency, or risk reduction. You’re not asking for resources. You’re proposing a solution to a business problem, and audience development is the mechanism.

Use honest ranges where your numbers aren’t precise. “Between $150K and $250K in incremental value” is more credible than a single false-precision figure. Your CFO knows estimates are estimates. What they’re evaluating is whether your logic holds.

What to ask for and what to skip

Your budget request should feel essential, not aspirational. Be precise about priorities and honest about sequencing.

The foundation comes first: a customer data platform integration that connects your tech stack. Most audience data is scattered across tools that don’t talk to each other — your email platform, your CMS, your analytics, your ad server. Fixing this is foundational, not optional. Without it, every downstream investment delivers less than it should.

Next is email deliverability and engagement optimization. If newsletters don’t reach inboxes, nothing else matters. Budget for IP warming, reputation monitoring, and segmentation infrastructure. This isn’t glamorous work, but it’s the kind of investment that makes every other audience initiative more effective.

Then attribution and funnel tracking that connects content performance to business outcomes. You need to show that audience development drives revenue, and that requires the ability to follow a subscriber’s journey from first touch to conversion. Even rough attribution is better than none.

After the foundation is solid, you can make the case for audience research and testing capabilities — user research tools, A/B testing for emails and content, regular surveys. Cross-platform distribution and optimization tools. Retention programs like community platforms, events, and exclusive content that increase lifetime value and reduce churn.

What to skip — or at least defer: advanced personalization and AI-powered recommendation engines. These can improve engagement, but they require solid data infrastructure to work well. If the foundation isn’t in place, they’ll underperform and give leadership a reason to question the broader investment. Build the base first.

How to structure the pitch itself

Your budget presentation needs to accomplish five things in roughly fifteen minutes. That’s tight, and you’re not the only person in line for time and attention — so discipline matters.

Start with external pressure, not internal needs. Open with the two or three macro trends that are reshaping your competitive environment. Use specific examples — a competitor who invested in audience infrastructure and gained share, or one who didn’t and lost it. The goal is to make “standing still” feel like the riskiest option in the room.

Then be honest about your current gaps. Present your audience infrastructure as it actually is, not as you wish it were. Show where you can’t measure what you need to measure. Quantify the revenue or growth you’re leaving on the table because of resource constraints. Specificity matters here — “we can’t segment beyond basic demographics” is more persuasive than “we need better tools.”

Walk through your priorities in order, explaining how each investment builds on the one before it and connects to a specific business outcome. Use clear timelines and success metrics for each phase.

Present conservative financial estimates. Include a sensitivity analysis showing that even if your assumptions are off by 25%, the return is still positive. This builds credibility. Overpromsing is the fastest way to lose the room.

Close with your implementation plan — milestones, accountability measures, and how you’ll report progress. Then acknowledge the risks and how you’ll mitigate them. This is the part most people skip, and it’s the part that builds the most trust.

Share the slide deck and supporting materials ahead of time. Don’t burn your minutes reading slides. Focus the live conversation on the decisions that need to be made and the questions leadership actually has.

Handling the objections

Four objections come up in nearly every audience budget conversation. Anticipating them is half the battle.

“Can’t we do more with existing tools?” This is the most common pushback, and it usually means leadership doesn’t understand the limitations. Be specific: “Our current platform doesn’t support behavioral segmentation. We have 47,000 subscribers in a single list. We can’t personalize, we can’t trigger based on engagement, and our open rates have plateaued because we’re sending the same message to everyone. Growth without better infrastructure means diminishing returns.”

“Marketing can handle audience development.” This one comes from an org structure assumption — that audience and marketing are the same function. They’re not. Marketing typically focuses on acquisition. But if the majority of your revenue comes from existing subscribers — retention, renewals, upsells — you need expertise in engagement and lifecycle management. Those are different skill sets with different KPIs.

“What if AI makes this obsolete?” AI is exactly why you need better audience infrastructure. The publishers who adapt to AI-driven discovery changes will be those with direct relationships and first-party data that doesn’t depend on platform signals. This investment makes you more resilient to AI disruption, not more vulnerable to it.

“The ROI timeline seems too long.” Break it into phases. Email optimization and segmentation improvements can show returns within three to six months. Broader infrastructure investments pay back over twelve to eighteen months. Long-term value creation happens over years, but operational improvements start showing up in the numbers quickly.

After the approval: maintaining momentum

Getting the budget approved is the beginning, not the end. The most common failure mode is winning the pitch and then going quiet for six months.

In the first two months, focus on high-impact, low-risk improvements that demonstrate quick value. Email deliverability fixes and basic segmentation upgrades can show measurable results fast — and they build credibility for the larger investments.

Months three through six are infrastructure work. System integration, attribution setup, and data unification. This phase may not produce dramatic metrics, but it lays the foundation for everything else. Communicate progress in business terms, not technical ones. “We connected our email and CMS data” means nothing to leadership. “We can now see which content topics drive the most subscription conversions” does.

From month six forward, use the improved data and infrastructure to drive measurable improvements in the KPIs from your original pitch. This is where the investment starts delivering visible returns — and where you build the case for next year’s budget.

Report progress using the same metrics and language from your original pitch. Consistency builds trust. If you pitched on subscriber lifetime value and churn reduction, report on subscriber lifetime value and churn reduction. Don’t switch frameworks midstream.

The real pitch underneath the pitch

The budget conversation is ultimately about whether leadership sees audience development as a cost center or a growth engine. Your job isn’t just to get a number approved. It’s to change how the business thinks about the function.

Every data point, every scenario, every ROI estimate should quietly reinforce the same message: audience infrastructure is the foundation that makes everything else — content, advertising, subscriptions, events — work better.

Make that case clearly, back it with honest math, and follow through on what you promise. The budget will follow.