Search Influence Weekly SEO/GEO/Online Ads Industry Update

Weekly Briefing — April 27, 2026: AI Overview Clicks, Demand Gen Delays, and AI as Ad Space

• Will Scott • Season 1 • Episode 4

Share episode

Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.

0:00 | 10:58

This week’s Search Influence briefing covers the latest shifts in AI search and AI-led advertising: AI Overview clicks may be stabilizing, citations still matter, low-quality AI content is contaminating answer engines, Google Demand Gen review delays are creating launch risk, and Microsoft is moving quickly into AI-first search advertising.

The larger takeaway: AI visibility is becoming an ecosystem problem. Brands need crawlable answers, proof assets, reviews, comparison content, structured pages, and paid media strategies that reflect how people now discover and choose businesses inside AI assistants.

In this episode

  • AI Overview clicks may be leveling off — Seer data shows AI Overview CTR rising from 1.3% in December to 2.4% in February, with cited pages still outperforming uncited pages.
  • The “AI slop loop” is a search risk — low-quality AI-generated SEO content can be retrieved and repeated by answer engines as if it were verified fact.
  • Local SEO still depends on satisfying the searcher — reviews, page speed, photos, local proof, and clear next steps all matter after the click.
  • Demand Gen review delays are creating launch risk — some campaigns are sitting in review for more than seven days, so time-sensitive pushes need backups.
  • Microsoft Ads is becoming a stronger AI test bed — AI Max is headed to open pilot, and Microsoft PMax now has better imports and landing-page reporting.
  • AI is becoming the next ad environment — brands will increasingly be discovered, compared, and selected inside assistants and answer engines.

Chapters

  • 00:00 — Opening note and weekly theme
  • 00:45 — TL;DR
  • 02:20 — AI Overview clicks may be stabilizing
  • 03:45 — The AI slop loop
  • 05:10 — Local SEO and searcher satisfaction
  • 07:05 — Demand Gen review delays
  • 08:10 — Microsoft’s AI ad stack
  • 09:20 — AI as the next ad space
  • 10:25 — Client intel and cheat sheet

References

Search Influence helps organizations earn visibility across search, AI answers, paid media, and the places customers make decisions. searchinfluence.com

SPEAKER_00

Hey, it's Will Scott's VoiceClone. The content is curated and reviewed by Will. The voice is AI. Welcome to the Search Influence weekly briefing for Monday, April 27, 2026. This week's theme is pretty simple. AI search is not replacing the fundamentals, but it is changing which fundamentals matter most. Clear answers, crawlable proof, better action paths, and tighter measurement are all becoming more important than raw volume. Let's get into it. TLDR, five things to know this week. One, AI overview, click loss may be leveling off. New SEER data says click-through rate on searches with AI overviews rose from 1.3% in December to 2.4% in February. That is still low, but the direction matters. Pages cited in AI overviews continue to earn more clicks than pages that are not cited. Two, bad AI content is now polluting AI search itself. Search Engine Journal called this the AI slop loop. AI generated SEO posts get published, retrieved by answer engines, and repeated as if they were verified facts. That makes original data, human review, and named expertise more valuable, not less. 3. Google demand gen reviews are lagging. Some advertisers are seeing campaigns sit in review for more than seven days. If a launch is tied to a promotion, enrollment deadline, or event, last-minute creative swaps are risky right now. Four, Microsoft is pushing hard into AI first advertising. AI Max is coming to open pilot in May, and Microsoft Performance Max now has better Google import support and landing page reporting. Five, AI is becoming an ad environment, not just a workflow tool. People are going to discover, compare, and choose brands inside AI assistance. That means SEO, paid media, content, and brand proof all have to work together. AI overview clicks, maybe stabilizing. Let's start with the SEER interactive data reported by Search Engine Land. For the last year, most of the conversation around AI overviews has been defensive. AI answers appear, organic clicks fall, and publishers panic. That is still the correct broad direction. AI overviews compress clicks, but the new data adds some nuance. Sear found that click-through rate on searches with AI overviews increased from 1.3% in December 2025 to 2.4% in February 2026. That is not a return to the old world, but it is a meaningful improvement from the bottom. The more important point is that cited pages still outperform uncited pages. If your page is included as a source in the AI overview, it has a better chance of earning the click than a page sitting on the same search results page without a citation. So the client-facing takeaway is not AI killed clicks. It is AI traffic is pickier. Visibility alone is not enough. The page has to be useful enough, structured enough, and trustworthy enough to be selected as a source. That means priority pages should have extractable answers, clear headings, first-party proof, concise definitions, and enough supporting detail that Google has a reason to cite them. The AI Slop Loop is a search quality problem search engine journal had a sharp warning this week about AI search eating itself. The issue is not just that low-quality AI content exists. We already knew that. The bigger issue is that answer engines can retrieve bad AI-generated SEO content, treat it like a source, and repeat it back into the ecosystem as if it were true. The example category here is fake or exaggerated industry news, things like invented Google update narratives, generic SEO claims, and unsupported tactical advice. Once that content is published, it can become part of the retrieval layer that AI systems use to answer questions. This is why the bar for content should be moving up, not down. AI can absolutely help with drafting, outlining, formatting, summarizing, and repurposing, but it should not be treated as the source of truth. For clients, the practical advice is straightforward. Do not publish AI written industry claims without verification. Add first-party experience. Quote actual subject matter experts. Use original data when possible, link to primary sources. And if the content is about a fast-moving topic, have a human check whether the claim is actually true before it goes live. In an AI search world, trust signals are not decorative. They are the difference between being a useful source and becoming part of the noise. Local SEO still comes back to satisfying the searcher. Near Media's latest episode with Cyrus Shepard brought click behavior back into the local SEO conversation. The tension is this: Google's public language around click data has historically been careful, but antitrust evidence, leaks, and field testing all point in the same practical direction. User engagement matters. If people ignore a result, bounce from it or immediately return to search. That is not a good sign. For local SEO, that means we cannot stop at categories, citations, and keywords. Those still matter, but they are only part of the system. The result has to earn the click, and the landing experience has to satisfy the intent. So the action items are very tangible. Improve titles so they match the searcher's need. Make reviews strong and recent. Use photos that prove the business is real. Add local proof on the landing page. Make the page fast, make the next step obvious. The easiest way to explain this to a client is Google does not just need to understand you. Searchers need to choose you and then feel like they made the right choice. Search volatility heated up again. Barry Schwartz reported another round of Google ranking volatility around April 23rd, after the March core update had already completed. There was no confirmed new core update. So this is not something to overlabel, but it is something to monitor. The risk with volatility is that teams react too quickly. A few noisy days can make everyone want to rewrite pages, roll back changes, or invent a story about why traffic moved. That can create more damage than the original ranking swing. The better approach is to flag meaningful changes than evaluate them against 14 to 28 day trends. Which page types moved, which query groups moved, did rankings change or only traffic? Did conversions move with traffic or did low-value traffic fluctuate? In other words, volatility is a reason to investigate. It is not automatically a reason to panic. Demand gen review delays are a real launch risk. On the paid media side, Search Engine Land reported that some Google Ads Demand Gen campaigns are sitting in review for more than seven days. Search and performance Macs appear to be moving normally, but Demand Gen is lagging for some advertisers. Google has acknowledged it as a known issue and says it is working on a fix. The practical implication is that Demand Gen needs more lead time right now. If a campaign supports a time-sensitive promotion, event, application deadline, or seasonal push, do not assume Creative can be swapped the day before launch. Build an extra review time, avoid unnecessary last-minute edits, and have backup coverage ready through Search Performance Max, Paid Social, or whatever channel can keep the launch moving if Demand Gen gets stuck. This is one of those updates where the strategy is not complicated, but the calendar risk is real. Microsoft is making its AI ad stack more interesting. Microsoft advertising had two updates worth watching. First, AI Max4search will enter OpenPilot in May. Microsoft is positioning it as AI-led query matching, asset personalization, and URL routing across Bing, copilot search, and copilot answers. The interesting part is the way Microsoft is framing control. They are emphasizing opt-in settings, brand inclusions and exclusions, term exclusions, messaging constraints, and more transparent reporting. That matters because a lot of advertisers are not opposed to automation. They are opposed to black box expansion they cannot inspect. Second, Microsoft Performance Max got easier to import and evaluate. The April update added support for importing Google PMX campaigns that use new customer acquisition goals. Microsoft also launched landing page reporting for PMAX, so advertisers can see performance by final URL, including spend, clicks, conversions, revenue, and return on ad spend. The takeaway is that Microsoft may be becoming a better, low-lift test environment. For accounts with healthy Google PMX performance, Microsoft import plus landing page reporting may give us enough transparency to run a controlled expansion. This is especially relevant for clients already getting qualified traffic from Bing or copilot adjacent searches. AI is becoming the next ad space ad exchanger's latest podcast, framed AI not only as a production tool, but as a new place where people discover and choose brands. That is the big strategic point. AI assistants are not just going to help marketers write faster, they are going to shape consumer decisions before the user ever reaches a traditional search result, social feed, or website. When an assistant narrows a choice set, the brand that appears in the answer has a huge advantage. The brand that does not appear may never get considered. This connects everything we do: SEO, paid media, content, reviews, videos, social, third-party citations, and comparison assets. The question becomes: how do we become the trusted answer when AI is helping the user decide? The answer is proof. Reviews, comparisons, outcomes, FAQs, videos with transcripts, third-party mentions, clear product and service pages, and content that explains who the business is for, what it does, and why someone should believe it. AI visibility is not one page or one platform. It is an ecosystem of evidence. A quick platform note Meta and AI Compute Meta also announced an agreement with AWS to bring tens of millions of Graviton cores into its compute portfolio for a genetic AI workloads. This is not an immediate client tactic, but it is a useful signal. The big platforms are still racing to support more AI heavy products. More compute usually turns into more automation, more recommendation changes, more AI-assisted ad products, and more shifts in measurement. So for agencies, the operating rhythm matters. Document what change, to test carefully, and explain plainly. Do not treat platform AI updates as one-off announcements and treat them as the new normal. Client intelligent three internal lessons. A few internal client conversations this week produced good lessons we can reuse without making them about any one account. First, for higher education program pages, AI visibility is often a page-level clarity problem. Strong reputation does not guarantee that AI systems understand which program is best for which prospect. The action item is to make each priority program page explicit, the program name, the audience, the format, the differentiator, the outcomes, employer data, credential details, and program-specific FAQs in crawlable HTML. For high-value comparison prompts, publish first-party comparison pages before third-party sites define the narrative. Second, interactive tools need SEO structure. A calculator or lead magnet should have a short entity-level URL, supporting copy, internal links, and a strong first-screen payoff. If the tool only becomes compelling after someone fills it out, pre-fill an example so the visitor immediately sees the value. Then build supporting content around the tool, setup guides, alternatives pages, and use case pages tied to known prompt gaps. Third, case studies need a simple story. Large accounts can produce endless performance detail, but a public case study should start with one business problem, one intervention, and one outcome. Listen for the narrative thread during reviews. What pressure existed, what changed, and what improved, then support it with one specific example and one broader proof point. Cheat sheet, six lines to use in client conversations this week. 1. AI overview traffic is not gone. It is pickier. Being cited gives you a better shot at the click. 2. Do not publish AI-written industry claims without verification. Bad AI content can get recycled into answer engines and damage trust. 3. Demand gen needs more lead time right now. Review delays mean time-sensitive campaigns need backups. 4. Microsoft Ads is becoming a better AI test bed. AI Max and PMAX reporting give us automation with more visible controls. 5. AI visibility is not one page or one platform. The strongest answers are supported by web content, reviews, video, social, and third party proof. 6. A case study is not a dashboard. Clients remember the problem, the action, and the outcome. That's the week. Go win something.