The distribution channels that don't depend on Google still sending you traffic

Moz published a piece in early 2025 called "How to Create a Defensive SEO Strategy for AI Search." It's worth reading. It's also written for someone managing a content portfolio across multiple domains, running technical audits, tracking structured data coverage, and briefing a team. If you're a creator with one site, a newsletter, and a schedule that doesn't include weekly crawl reports, the advice is real but the translation is missing.

This is what that translation looks like.

The Moz framework centers on a concept they call "defensive positioning" — protecting the organic traffic you already have against zero-click erosion. For agencies, that means schema markup, entity optimization, and doubling down on branded search queries where your audience is already looking for you specifically. The logic is sound. The tactics apply at a different scale than most creators operate at.

What the framework implies, without stating directly, is that the underlying shift is about channel dependency. A business that grew primarily through organic search discovery has a single point of failure. The defensive move isn't just technical — it's structural. You're trying to reduce your exposure to any one referral source.

For a creator, that reads more plainly: the channels that don't route traffic through a search results page are structurally safer right now than the ones that do.

Email lists are the clearest example. Not because email marketing is new advice — everyone has told you to build your list since 2015 — but because the structural reason changed. The original argument for email was that you owned the relationship and weren't subject to algorithm changes. That argument was correct then and it's still correct. The new version of the argument is more specific: email is one of the only channels where discovery doesn't require an intermediate step through a results page that might answer the question before the reader reaches you.

A subscriber who opens your email made an explicit decision. They signed up, they clicked, they're there. No zero-click equivalent exists for email because there's nothing between you and the inbox.

YouTube is more complicated. It has its own search function, its own recommendation algorithm, and its own version of zero-click risk in the form of on-platform features that answer questions without requiring a click. But YouTube search behavior skews toward discovery and demonstration rather than the short Q&A queries most at risk of zero-click erosion. People watching a twelve-minute video about setting up a specific workflow are not the same people who searched "what is a webhook" and got an inline answer. The intent is different, which means the substitution risk is different.

That doesn't make YouTube immune. It means video content built around depth, demonstration, and process — the kind of content that's hard to replace with a paragraph — holds up better than short explainers designed to rank for high-volume informational queries.

Podcasts are structurally resistant in a simple mechanical sense: there's no inline answer for an audio file. A query that surfaces a podcast episode in results still requires a click. That's not changing soon. The trade-off is discoverability — podcasts grow primarily through recommendations and cross-promotion, not search — which means they're better channels for retaining an audience you already have than for acquiring new readers who've never heard of you.

The more interesting channel dynamics are on the side that's newly complicated. Reddit is the clearest example. Google has dramatically increased its surfacing of forum content over the past two years. If your audience is asking questions on Reddit and getting answers there, that's not necessarily bad — you can participate in those threads, build presence in those communities. But if your strategy was to publish how-to content on your own domain assuming those readers would find you through Google, the current environment is routing some of that traffic to forum threads instead.

The defensive move for Reddit isn't to compete with it. It's to recognize that forum-format content and long-form editorial content serve different purposes, and optimize each accordingly.

The translation from SEMrush's framing to creator language goes like this: when they write that "search has changed and so have we," what they mean for their own business is that they're investing in brand search — queries like "SEMrush competitor analysis" where someone is looking for them specifically, not just for the category answer. They're not building a strategy around ranking for "what is keyword research" because that's a question a results page can now answer without them.

The creator equivalent is becoming someone your audience searches for by name. Not "best email marketing tips" — that's a category query that belongs to whoever the results page decides to surface. But "your name + topic" is a different kind of query. Someone who searches for you specifically is not going to be satisfied by a synthesized summary of your work.

That's the practical goal of brand building for this environment: shift the reader's behavior from searching for answers to searching for you.

A few concrete things worth doing before the end of the quarter.

Track where new subscribers actually came from over the last twelve months. Most email platforms show you the page where someone signed up, not the source that drove them there. Pull your referrer data from analytics. If a significant portion of your new subscribers in 2023 came through organic search and the proportion has dropped since, that number tells you how much of the problem is already here.

Audit your content by risk category. Short how-to posts, definitions, comparison articles, FAQ content — these are the formats most exposed to zero-click substitution. Long-form analysis, opinion, reported pieces, and content grounded in specific experience are harder to replace with an inline answer. If your content calendar is weighted toward the first type, you're carrying more exposure than you may realize.

Ask what your readers actually forward. The SEMrush and Moz frameworks both point, from different directions, at the same underlying shift: from content designed to get discovered by strangers to content that gets shared by people who already trust you. Forwarded content builds an audience without depending on any platform or results page. The person who forwarded the piece did the distribution work. The person who received it made a decision based on a trusted recommendation. Neither step required a search result.

The channels that are still working — email, well-executed video, strong community presence, genuine word-of-mouth — have always worked for the same reason: they depend on the quality of the thing rather than the mechanics of discoverability. The current environment is not creating a new standard. It's removing a shortcut that existed for a while.

That shortcut was real and useful. Building a subscriber base through organic search discovery was efficient, scalable, and cheap. That version of the playbook is less reliable than it was. The alternative isn't a replacement shortcut. It's the slower version that was always there.

Sources: Moz, "How To Create a Defensive SEO Strategy for AI Search"; SEMrush, "Search Has Changed. And So Have We."; MarketingProfs, "How Marketers Win Visibility in the Age of Zero-Click Search and AI Overviews"