The pattern everyone’s missing: channels are stable, surfaces are exploding
Look at the headlines you just skimmed:
- “What’s next for PPC: AI, visual creative and new ad surfaces”
- “Google Ads Surfaces PMax Search Partner Domains In Placement Report”
- “Are YouTube’s Latest Tools Ready for Businesses”
- “Bluesky growth: Is your brand ready for decentralized social media?”
- “How to integrate Agentic AI for social media into your workflows”
- “Dept Powers Up Its Global Content Studio With Adobe AI”
- “Can the creator economy stay afloat in a flood of AI slop?”
Same story, different angles: the channel list (search, social, video, email) is basically the same as 2016. But the surfaces inside those channels are multiplying and mutating fast:
- Search is now classic results, AI overviews, “summarize with AI” sidebars, shopping units, and PMax’s black box of partner sites.
- Social is feeds, stories, Reels/Shorts, TikTok search, DMs, agentic AI assistants, and soon decentralized feeds you do not fully control.
- Video is pre-roll, mid-roll, shoppable overlays, creator integrations, AI video units, and “tools” that blur content and ads.
- Owned is email, communities, AI-generated content hubs, and conversion flows that now depend on how AI systems interpret your pages.
Most teams are still organized, budgeted, and measured for channels. The market is increasingly traded on surfaces.
That gap is where a lot of wasted spend, weird results, and “why is performance so volatile?” lives right now.
Why this matters now: AI intermediaries are your new media buyers
A few headlines should make every CMO and media lead pause:
- Microsoft’s “summarize with AI” buttons being used to poison AI recommendations.
- Generative engine optimization guides showing up next to classic SEO content.
- Vectorization and transformers being explained to search marketers like they’re now part of the job (because they are).
- AI video tools and agentic AI for social media being pitched as workflow essentials.
Translation: your content and ads are increasingly mediated by AI systems you do not control:
- Search and recommendation engines that summarize, remix, and re-rank your content.
- Ad platforms that auto-generate creative, placements, and audiences.
- Scheduling tools and social assistants that decide when and where things appear.
These systems don’t care about your “channel strategy.” They care about:
- What’s on the page.
- How it’s structured.
- How users respond.
- How easily it can be turned into a snippet, a summary, or a recommendation.
In other words, they care about surfaces and signals.
From channel thinking to surface thinking
Channel thinking sounds like this:
- “We’re heavy on Meta and Google, testing TikTok, email is a retention channel.”
- “SEO is a long-term play, we’re doing conferences and content.”
- “We have a YouTube strategy and a paid search strategy.”
Surface thinking sounds more like:
- “We compete in search result pages, AI answer boxes, and shopping units for these 40 intents.”
- “We compete in short-form vertical video feeds across TikTok, Reels, Shorts, and YouTube recommendations.”
- “We compete in inboxes, notifications, and AI-generated summaries of our own site and docs.”
- “We compete in creator content, UGC, and AI-generated lookalike content in our category.”
Same budget, different mental model. The second one matches how attention and algorithms actually work in 2026.
Step 1: Map your actual surface graph
Before you optimize anything, you need a map. Not of channels. Of surfaces where your brand can appear.
Practically, this is a half-day working session with your performance, brand, content, and product teams. The output is a single-page “surface graph” that answers:
- Where can a prospect first encounter us?
- Search: classic organic, AI answers, shopping, local packs, PMax partner sites.
- Social: feed posts, Reels/Shorts, stories, search results, creator mentions, DMs, agentic AI suggestions.
- Marketplaces: category pages, sponsored listings, recommendation carousels.
- Owned: email previews, push notifications, AI summaries of our docs or support content.
- What does the unit actually look like?
- Character counts, aspect ratios, visible fields, truncation points.
- What gets shown by default vs on click or expand.
- Who is the real gatekeeper?
- Human (editor, creator, community mod).
- Algorithm (ranking, recommendation, auction).
- AI system (summarizer, assistant, agent).
Put this into a simple table or diagram. It is not a pretty slide exercise; it is a planning artifact. Then ask:
“Given this, where do we actually make money or meaningful progress?”
Step 2: Decide your “high-intent surface stack”
Not all surfaces are created equal. Some are nice-to-have impressions. Some are where money changes hands.
For most performance-oriented businesses, high-intent surfaces cluster around:
- Search intent surfaces
- Classic search ads and organic results.
- AI answer units that summarize options.
- Shopping and local units that show price, rating, and availability.
- Decision surfaces
- Product detail pages and comparison pages.
- Review sites, marketplaces, and creator “best of” lists.
- AI assistants answering “which X should I buy?”
- Retention and expansion surfaces
- Inbox previews, in-app messages, and account dashboards.
- Help centers and docs that AI systems crawl and summarize for customers.
Pick 3-5 surfaces where:
- Users are close to a decision, and
- You can realistically improve your presence in the next 90 days.
This is your “high-intent surface stack.” Treat it like a product roadmap, not an ad-hoc test list.
Step 3: Align creative and data around surfaces, not teams
The current org default:
- Paid social team makes social creative.
- Search team writes search ads and meta tags.
- SEO/content team writes blogs and landing pages.
- CRM team writes emails.
The algorithmic reality:
- A single product message might appear as:
- A PMax auto-generated headline.
- A snippet in an AI answer box.
- A caption in a TikTok video.
- A subject line in an email preview.
- Anchor text in a Wikipedia or knowledge-style page.
If those are not coordinated, you confuse both humans and machines.
A more effective setup for 2026:
- Create surface squads around your high-intent stack. For example:
- “Search and AI answers” squad: search, SEO, content, data science.
- “Short-form video feeds” squad: paid social, organic social, creative, influencer.
- “Decision pages and summaries” squad: CRO, product marketing, SEO, CRM.
- Give each squad:
- Clear commercial targets (revenue, qualified pipeline, LTV, not just CTR).
- Control over creative variants across formats for that surface.
- Shared reporting that cuts across channels but stays within surfaces.
You are not reorganizing the whole company. You are creating cross-functional pods that think in surfaces, not silos.
Step 4: Treat AI systems as a distribution channel you can influence
Generative engine optimization, AI video, agentic social tools, and AI-powered studios all point to the same reality:
You are now marketing to AI systems as much as you are marketing to humans.
That does not mean chasing every hacky prompt trick. It means doing three boring but powerful things:
- Make your content machine-readable and snippet-ready
- Clear headings, short paragraphs, and explicit answers to common questions.
- Structured data where relevant (product, FAQ, how-to, local business).
- Canonical, non-duplicative content to avoid cannibalization and confusion.
- Standardize your claims and proof
- Use consistent numbers, benefits, and positioning across pages and channels.
- Back up claims with sources AI systems can crawl (case studies, reviews, third-party coverage).
- Avoid spraying 20 different taglines for the same product; pick a spine and stick to it.
- Instrument how AI surfaces mention you
- Track when and how AI answer boxes, assistants, and summaries reference your brand.
- Monitor shifts after major content or site changes.
- Use that feedback loop like you would use search query reports or placement reports.
The goal is not to “game” AI. It is to make it easy and low-risk for these systems to recommend you accurately.
Step 5: Update your media buying playbook for surface volatility
A lot of media buyers are quietly struggling with:
- Performance swings that do not match obvious seasonality.
- Campaigns that work well in one placement but die in another.
- Platforms shifting inventory into new surfaces without clear controls.
This is what happens when surfaces change faster than your settings.
A more robust playbook:
- Budget by surface cluster, not by platform line item
- Example: “Search and AI answers” gets a combined budget across classic search, PMax, and any AI answer sponsorships, instead of separate fiefdoms.
- “Short-form vertical video” gets a shared budget across TikTok, Reels, and Shorts.
- Demand surface-level reporting wherever possible
- Use placement reports, creative breakdowns, and third-party tools to infer where your impressions actually show.
- Tag creatives and campaigns by intended surface (e.g., “short-feed,” “discovery-search,” “AI-answer-supportive”) so you can track intent vs reality.
- Set guardrails for black-box expansion
- Decide upfront how much budget you are comfortable letting platforms push into new surfaces without explicit approval.
- Use experiments: 80 percent in proven surfaces, 20 percent in “platform expansion” with tighter ROAS or CPA thresholds.
- Design creative for portability
- Assume your creative will be chopped, cropped, summarized, and remixed.
- Front-load the key message and proof in the first 2-3 seconds or first 80 characters.
- Use visual and verbal cues that survive without sound, captions, or full context.
Step 6: Build a “surface health” dashboard the C-suite actually uses
CMOs do not need another 40-page channel report. They need a simple way to see:
- Where we are winning attention that converts.
- Where we are underrepresented vs competitors.
- Where new surfaces are emerging that threaten our current edge.
A practical “surface health” dashboard has:
- 3-5 key surface clusters (e.g., “Search and AI answers,” “Short-form feeds,” “Decision pages,” “Inbox and notifications”).
- For each cluster:
- Share of visible presence vs top 3-5 competitors (rough, directional is fine).
- Revenue or pipeline contribution attributed to that cluster.
- Trend over the last 90 days.
- Notes on major platform or algorithm changes affecting that cluster.
- One line per quarter on what you are doing about it:
- “Q2: Expanding structured content to improve AI answer coverage.”
- “Q3: Consolidating creative for short-form feeds around 3 proven hooks.”
This is the bridge between media ops reality and boardroom expectations. It shifts the conversation from “Why is Meta ROAS down?” to “Our short-form feed cluster is underperforming; here is what we are doing across platforms to fix it.”
What to do in the next 30 days
If you want this to be more than a nice idea, here is a 30-day sprint you can actually run:
- Week 1: Map and choose
- Run the surface mapping workshop.
- Pick your 3-5 high-intent surface clusters.
- Week 2: Squad and instrument
- Form temporary surface squads with clear owners.
- Tag existing campaigns and content by intended surface.
- Set up a basic surface health view in your BI tool or even a spreadsheet.
- Week 3: Clean and standardize
- Align messaging and claims across your top 20-30 assets for each cluster.
- Fix obvious cannibalization or duplication issues that confuse both users and AI systems.
- Week 4: Test and reallocate
- Shift 10-20 percent of budget into surface-clustered experiments.
- Run at least one creative test designed explicitly for how that surface presents content.
- Review results as a surface squad, not as individual channels defending their turf.
Channels are the plumbing. Surfaces are where the water actually comes out. In a world of AI intermediaries, recommendation engines, and endless new formats, the brands that win will be the ones that buy, build, and measure against the surfaces where decisions are made.