On Oct 22, 2026, Search Engine Roundtable’s write-up on three new AI search studies put a hard pin in something most teams feel but few budget like: AI search visibility is still not a meaningful direct-traffic channel compared to Google Search.
The studies differ, but the direction is the same: clicks and conversions attributed to AI search tools are a rounding error next to classic organic. If your 2026 plan quietly assumed “we’ll get discovered in ChatGPT/AI Overviews and traffic will follow,” this is the reality check.
What changed (and what didn’t)
What changed this week is not Google’s algorithm or some new product switch. What changed is the evidence stack: three separate studies landed close together, and they all point to the same outcome—AI answer engines are not sending much traffic out to sites.
What didn’t change: Google is still where demand capture happens for most businesses. People still search. They still compare. They still click through when they need pricing, availability, specifications, booking, or a vendor they can hold accountable.
Our take at BrandJunction: treat AI search visibility as brand presence and conversion assist. Do not treat it as a replacement for ranking and landing pages that convert.
Why this matters to a business owner (money, not hype)
AI visibility work tends to get pitched like an SEO sequel: “We’ll optimise for LLMs.” The problem is the same problem as early voice search hype: if the channel doesn’t refer traffic, it can’t pay for itself unless you measure it as influence.
In practice, most teams are doing two expensive things right now:
- Rewriting content to “sound good to AI” (often making it worse for humans).
- Chasing mentions/citations without a plan to connect that exposure to pipeline.
If you’re a marketing lead, that means budgets drift into work that looks modern but doesn’t move revenue. If you’re a business owner, it shows up as a quarter where “visibility improved” but enquiries didn’t.
One contrarian point everyone will miss: low AI-referral traffic does not mean AI doesn’t matter. It means the value is upstream. AI answers can compress consideration, steal informational clicks, and still shape who a buyer shortlists. So the question isn’t “should we invest in AI search visibility?” It’s “what’s the cheapest way to defend brand preference while we keep winning the clicks that still exist?”
Our POV: AI Overviews are an assist channel, not a growth channel
We usually recommend thinking in two lanes:
Lane 1: Demand capture (still Google Search). This is your pages that rank, earn clicks, and convert: service pages, location pages, comparison pages, pricing pages, category pages, and problem/solution pages with clear next steps. This is where the measurable ROI still lives, and it’s why our SEO services work is still built around technical clarity, intent mapping, and conversion paths.
Lane 2: Demand influence (AI answers + other “zero-click” layers). Here, you’re trying to be the brand that gets named, trusted, and remembered. You may get fewer clicks, but you can still win the deal if your brand is the one buyers feel safe choosing.
If you blend the lanes, you end up judging AI visibility by the wrong KPI (clicks), or judging SEO by the wrong KPI (mentions). Keep them separate, then connect them with measurement.
What we’re changing on client sites this month
This is the month to stop guessing and start instrumenting. While competitors are still hand-writing “AI-friendly FAQs,” you can build a measurement loop that tells you which topics and pages win both classic results and answer layers.
Three concrete shifts we’re making:
1) Build an “assist” view in GA4. Direct AI referrals can be tiny, but AI-influenced journeys often show up as brand searches later. Set up explorations that track: first touch = organic informational page, later touch = brand query landing on pricing/contact/demo. If you don’t have clean conversion events, fix that first.
2) Double down on pages that can win clicks. If AI summaries reduce informational clicks, your content mix has to move closer to decision. That means more “compare,” “pricing,” “alternatives,” “best for,” “requirements,” and “implementation timeline” pages—supported by real service delivery. That also means your web development services and SEO teams need to work together: speed, UX, internal linking, and conversion UX are not side quests anymore.
3) Stop “LLM optimisation” that breaks readability. We’re seeing teams flatten their copy into generic definitions because they think AI likes it. Humans hate it. Google also tends to reward clarity, specificity, and proof. Keep the content sharp: who it’s for, what it costs, what’s included, what’s not, time-to-value, and constraints.
One more operational note: if your tracking is messy, you can’t evaluate any of this. If you’re rebuilding flows or adding gated steps, budget it properly (our software development cost guide (US) is a good starting point for scoping).
What this means for your site
- Audit where revenue comes from today. In Google Search Console, export the last 28 days of queries/pages and mark the URLs that lead to enquiries or sales. Protect those rankings first.
- Create 2–4 “decision pages” in the next 30 days. Examples: “Pricing,” “X vs Y,” “Best [service] for [industry],” “Implementation timeline.” These are resilient even if AI reduces top-of-funnel clicks.
- Instrument brand lift. Track branded impressions/clicks in Search Console and correlate to content releases. AI influence usually shows up there before it shows up as referral traffic.
- Make sure your entity signals are consistent. Company name, services, locations, leadership, and credentials should match across site, schema, and profiles. Not because it guarantees AI citations—but because inconsistency kills trust everywhere.
Related moves from Google worth watching
Google keeps pushing education and community resources for site owners, which is a quiet signal: classic SEO fundamentals still matter, and Google expects teams to keep investing there. If you need to get your in-house team aligned quickly, point them to Google Search Central’s new learning-path playlists for SEO fundamentals and Technical SEO.
On the local side, Google is also iterating on business-facing surfaces. That matters because local intent queries are some of the most commercially valuable and most sensitive to UI changes. Keep an eye on this report on Google testing a local knowledge panel layout change, and if you have physical locations, sanity-check how your brand presents on desktop vs mobile.
And if you manage multiple locations or depend on GBP for calls, bookings, or footfall, note the beta launch of Google Business Home for Business Profiles. New interfaces often come with new defaults and missed fields—worth a 20-minute review.
If you do nothing for 90 days
You’ll likely end up in the worst middle ground: spending time “optimising for AI” while your money pages stagnate, your conversion tracking stays fuzzy, and competitors quietly ship decision pages that pull buyers out of the comparison phase.
AI answers may also keep absorbing informational clicks. If your strategy is still “publish more blog posts and hope,” you’ll feel the squeeze: impressions without traffic, traffic without intent, and reporting that can’t explain the gap.
Next step: pick one revenue-critical service line, map the 5–10 queries that lead to sales, and build (or fix) the pages that should rank for them. If you want us to pressure-test your plan, start with our SEO support for French markets or our SEO work for German-language growth, and we’ll tell you what to prioritise this month.
Primary source: Search Engine Roundtable
