Should SEO still be in your 2026 marketing strategy? The data says yes, but differently

In Radyant's client work, we keep seeing the same pattern. A lead tells sales they found the company by asking ChatGPT. GA4 files the same lead under Direct. A few weeks later the CEO forwards a LinkedIn post called "SEO is dead" and asks whether the budget should move.

Our answer is yes, keep SEO in your 2026 plan. Search is not collapsing: Graphite's traffic data puts SEO traffic down 2.5% year over year. What has changed is which pages still earn the click, and how badly click-based analytics see the leads that come through AI Search. So the work changes, and the measurement changes more.

This guide is for B2B marketing leads who have to defend, cut or rebuild an SEO budget this year. If your site already gets plenty of traffic and very little pipeline, the section on that case further down is the one for you. More SEO will not fix it.

Key takeaways

• Keep SEO in the plan, and change what you measure. Graphite's data shows SEO traffic down only 2.5% year over year. The bigger risk is judging the channel on traffic instead of pipeline.
• AI Search leads often show up as Direct in your analytics. A buyer gets your name in ChatGPT, types it into Google and books a demo. Radyant reads four signals together, starting with a free-text form field.
• Simple "what is" pages are losing clicks. Buying-journey pages are not. AI Overviews cut clicks by about 35% where they appear. Comparisons and implementation guides still get clicked.
• Organic gets cheaper over time, paid gets more expensive. WordStream's 2025 benchmarks put the average Google Ads click up 12.88% in a single year.
• You don't need a separate AEO program. At Planeco Building, one owned content strategy reached 5x organic leads in 10 months and #1 AI Search visibility in its competitor set.

Is search actually shrinking?

No. The "SEO is dead" story comes back every few years. It came after Panda, after mobile-first indexing, and now it comes with AI Overviews and ChatGPT. The data has not followed it.

Graphite's analysis of actual traffic data shows SEO traffic down 2.5% year over year. Traffic to search engines has been relatively flat since 2023 (-1.0%), and visitors to Google increased by 1.4% comparing Q4 2025 with Q4 2024.

Google said in March 2025 that it sees more than 5 trillion searches a year, about 13.7 billion a day, and that commercial query volume has gone up since AI Overviews launched. AI Search is growing fast, and the growth is real. For now, it adds to search rather than replacing it.

Where are clicks actually falling?

The drop is concentrated in a few places, and it is older than AI:

• The largest sites are growing. The top 10 sites grew organic traffic by about 1.6%. Declines are concentrated among mid-sized publishers ranked between the top 100 and 10,000, according to Search Engine Land's analysis.
• Zero-click searches are real, and they predate AI by years. 60% of searches end without a click, 77% on mobile. Google showed Featured Snippets in 80% of the cases where AI Overviews now appear, and those snippets already reduced CTR.
• AI Overviews reduce clicks by about 35% when present, but they only appear in roughly 30% of searches, per Ethan Smith's research at Graphite.
• Transactional, navigational and complex research queries still generate clicks. The queries losing clicks are simple informational ones, where a direct answer is enough.

Ethan Smith calls the "search is dying" story "based on a series of myths driven by speculation, exaggeration, flawed analysis, and false information." Something is changing. It is just narrower than the headlines.

Which parts of SEO are losing value, and which are gaining?

It would be nice to say the old playbook still works. Parts of it don't. Panstag puts it plainly: "If your site mainly publishes surface-level information, SEO in 2026 feels unrewarding because the platform no longer needs your page." Google and AI models can answer "What is CRM software?" without sending anyone to your glossary page. If that kind of page is the core of your SEO, you have a problem.

The fix is to change what you do inside SEO, and keep the channel.

What's losing value

• "What is" glossary pages. AI answers these directly. An 800-word definition page adds nothing ChatGPT can't put together from 50 sources.
• Traffic as the north star metric. HubSpot's estimated organic traffic dropped roughly 80% according to third-party tools, mostly in top-of-funnel informational content. In the same period HubSpot reported $3.13B in revenue for 2025, up 19% year over year. Traffic went down and revenue went up, so traffic tells you little about whether SEO is working.
• Keyword-first content planning. Starting from a keyword list and working backwards to topics produces generic pages that look like everything else on page 1.
• Thin informational content at scale. 500-word posts that exist to catch a keyword and add no original insight, data or perspective.
• Vanity ranking reports. Showing your CEO a #3 ranking for a keyword with 50 monthly searches and zero pipeline impact.

What's gaining value

• Deep buying-journey content. Comparison pages, evaluation guides and implementation guides, written for someone actively making a purchase decision. These queries still get clicks, because the buyer needs the detail and a short answer does not do the job.
• Pipeline and citation share as KPIs. How many leads came from organic? What percentage of AI Search answers in your category cite your content?
• Expert depth that can't be synthesized. Original data, real case studies and perspectives that don't exist anywhere else. Google and AI models both want to surface this, because it is what users find useful.
• Audience-first research. Start from what buyers need to know (sales calls, customer interviews, support tickets), not from what a keyword tool says has volume.
• Brand SEO. Someone who searches your name already knows what they want, and that searcher converts better than one who clicked a generic result. Building brand recognition through search is moving from a side effect to a plan of its own.

Why do your analytics hide the leads AI Search sends?

For anyone making budget decisions, this matters most. The biggest risk to your SEO investment is not AI Search taking your traffic. It is analytics that credit the wrong channel.

Here is how a lead travels in 2026. A buyer asks ChatGPT, "What's the best asset management software for mid-size facilities?" ChatGPT mentions your brand. The buyer types your brand name into Google, clicks your homepage and requests a demo. Your CRM records "Organic Search" or "Direct". ChatGPT gets no credit.

Across our client work at Radyant, we consistently see leads who say they found the brand through AI Search and show up as Direct or Organic in click-based analytics. At ToolSense, where we helped 10x organic leads in two years, the gap between what analytics reported and what prospects said widened as AI Search grew.

What happens when you trust the dashboard alone?

Marketing leads look at their attribution data, see that "AI Search" drives almost nothing, and conclude it isn't worth the money. Meanwhile the influence of being cited in AI answers never reaches the dashboard, so the next budget round cuts exactly the work that produced it.

The four signals we read together

No single attribution method gives you the full picture in 2026. We read four signals together.

Self-reported attribution. Add a "How did you hear about us?" field to your forms. Make it mandatory and free-text, because a dropdown forces people into categories that may not match their actual journey. LLMs make analyzing thousands of free-text answers trivial. Pair it with a custom CRM field sales fills in after the first call. "Oh, I asked ChatGPT for recommendations and your name kept coming up" usually surfaces there first. This is where you find out how many leads came from ChatGPT, Perplexity or a podcast episode. It is also the cheapest step in this guide: one form field and one CRM field.

AI-referral sessions. Your CRM and GA4 data. Keep it running, it is useful for direct-click channels like paid search. Just don't treat it as the only truth for discovery channels. It systematically undercounts AI Search, word of mouth, podcast mentions and any journey where the buyer doesn't click a trackable link.

Branded-search lift. Rising branded-query clicks and impressions in Search Console. Demand that shows up here a few weeks after your AI visibility climbs is the echo of that visibility, not a coincidence.

AI Search visibility. Citation share across engines. It is the leading indicator and moves before pipeline does.

Is SEO still cheaper than paid search for B2B SaaS?

Yes, and the gap widens over time. The paid side is easy to measure from published benchmarks. The organic side is our reading of how the channel compounds, not a number we have measured across our own accounts.

WordStream's 2025 benchmarks, built from 16,446 US search campaigns between April 2024 and March 2025, found the average Google Ads cost per click up 12.88% year over year, and search advertising costs rising every year for five years. As more advertisers compete for the same queries and Google adds more ad placements, it looks like paid acquisition costs will keep climbing.

Why does organic get cheaper while paid gets more expensive?

Paid search is rented attention. The moment you stop spending, the leads stop. SEO compounds: every page you publish and every bit of authority you build keeps working, so the cost per lead from organic falls as the content library grows.

We saw this with ToolSense: 10x organic leads in two years. The spend did not keep rising. The content and authority built in year one kept generating returns in year two, and each new piece added to that base.

Keep paid search. Just look at where each channel is heading. If paid clicks keep getting dearer at anything like that rate while your organic cost per lead falls as content compounds, the 24-month math is simple.

What should you do differently in SEO in 2026?

Keep investing, and invest in the version of SEO that still pays. For B2B SaaS marketing leads, three changes matter most.

Write the best answer, for every platform at once

A growing industry of "AEO experts" sells AI-specific tactics: prompt-style writing, citation optimization, content chunking. Most of it is good UX and content strategy under a new name. We have used question-based headings for years because users have specific questions. Key takeaways go up top because users don't have time to scroll. These tactics work in AI Search because they worked for users first.

What lasts is making your content the best, most complete and most trustworthy answer to the buyer's question. Google wants to serve that page, and ChatGPT and Perplexity want to cite it. They share one incentive: surface the most helpful content.

We ran this with Planeco Building: 5x organic leads in 10 months, a citation rate that climbed from 55% to over 110%, and #1 AI Search visibility in their competitor set. We made their owned content the reference in their space. No Reddit outreach, no backlink chasing, no separate "AI-optimized" versions. Just depth, expertise drawn out in regular interviews with the co-founders, and content that covered what their audience needed to know.

As Andy Muns, Director of AEO at Telnyx, discussed on our Masters of Search podcast, AEO might just be SEO. The overlap is large, and running them as separate disciplines creates extra complexity and duplicated work.

Treat brand search as a growth channel

When ChatGPT mentions your brand in a recommendation, what happens next? The buyer searches your name on Google, visits your homepage, and converts at a much higher rate than someone who clicked a generic informational result.

That is why brand SEO now matters so much. A branded search is often the step right after an AI recommendation, and the person making it has already narrowed the list. It also means your homepage matters more, because it is more and more often the entry point after a buyer discovers you in AI Search.

In practice, your SEO plan needs to build the content authority and brand recognition that get AI models to name you. That is SEO too, and the most valuable part of it.

AI Search visitors already behave this way. Semrush's 2025 study of AI Search traffic found the average AI Search visitor 4.4 times as valuable as the average organic search visitor, based on conversion rate. These visitors got a recommendation and are now checking whether to buy.

Expect buyers to search more, not less

"SEO is dead" assumes AI replaces search. The evidence points the other way: AI creates search behavior that didn't exist before.

Think about planning a trip to Paris. Before AI, you might run 5-10 Google searches about hotels, restaurants and activities. Now you have a 20-minute conversation with ChatGPT that raises dozens of follow-up questions you wouldn't have thought to ask. Some happen inside the AI tool. Some spill over to Google for checking, booking or deeper research.

Google's own numbers point the same way. In May 2025 Elizabeth Reid, Google's head of Search, wrote that in the US and India AI Overviews drive "over 10% increase in usage of Google for the types of queries that show AI Overviews." The total amount of searching grows, even as the click-through rate on any single query falls.

For B2B SaaS, that means buyers do more research across more platforms before they fill out a form. Companies that show up consistently across Google, ChatGPT, Perplexity and industry sources get a bigger share of that longer research journey.

As Ethan Smith discussed on Masters of Search, ChatGPT traffic converts six times better than Google search. The two figures (4.4x and six times) come from different data and don't match exactly, but they point the same way. The volume is smaller and the intent is much higher. Fewer clicks, better clicks, and you only see it if you measure the right things.

Should your company invest in SEO?

For most B2B companies, yes. How much, and in what, depends on where you start. Here is how we would call it for five common situations.

B2B SaaS with a long sales cycle

SEO is likely your highest-ROI channel. Your buyers research at length before they talk to sales. They compare solutions, read reviews and evaluate features, and being the best resource throughout that journey compounds into pipeline. Recommendation: invest heavily, and move the focus from informational content to buying-journey content (comparisons, implementation guides, ROI calculations). Measure pipeline contribution, not traffic.

Strong brand, declining organic traffic

Fix your measurement before you cut the budget. The traffic decline may sit in informational queries that AI now answers directly, while buying-intent traffic and AI-referred traffic grow. Read all four signals together, and check what self-reported attribution says. You may find AI Search driving real pipeline that your analytics can't see. Recommendation: fix measurement first, because it costs a form field and a CRM field, and then decide the budget on complete data.

Early-stage startup with no content

SEO compounds, so starting now builds an advantage that gets harder for competitors to close. "SEO takes too long" is less true than it used to be. With programmatic content engineering, results can come fast. When we launched a 247-page programmatic cluster for Planeco Building, 140 pages were ranking in the Top 3 within 72 hours, generating 60+ leads in under 6 months. Recommendation: start building now, use content engineering to speed it up, and pair it with paid for immediate demand while organic compounds. Budget paid for the months before organic starts to pay back.

Traffic-rich but pipeline-poor

More SEO won't help here. The content attracts the wrong audience, or fails to convert the right one. It usually means too much top-of-funnel informational content and not enough mid- and bottom-funnel content for buyers with purchase intent. Recommendation: audit your content against the buying journey, move production towards intent-dense content, and improve conversion paths on the pages that already get traffic. This is a doing more with less problem.

Highly competitive market with established players

Competing head-to-head on generic keywords is expensive and slow. Programmatic content aimed at long-tail, specific queries can win ground quickly. Combine it with owned content on your core differentiators and you can build visibility without outspending the incumbents. Recommendation: go for specificity over volume, and own the niche queries your competitors ignore. Winning against 100x budgets is possible when you compete on depth and relevance rather than domain authority alone.

What should you report instead of traffic?

Stop reporting traffic as your primary SEO metric. If traffic were a true north star, it would show up in investor reporting. It doesn't: HubSpot's full-year 2025 results report revenue and customer counts, not visits, because investors care about revenue and pipeline.

Track these instead:

• Pipeline from organic. How many qualified opportunities started in organic search, including AI Search captured through all four signals?
• Revenue attributed to organic. Closed-won revenue where organic played a part in the journey.
• Citation share. What percentage of AI Search answers in your category mention your brand? Tools like Peec AI track this. We compared Peec AI vs. Profound in detail.
• Conversion rate by channel. Compare organic, AI Search (self-reported), paid and direct. You'll likely find that AI Search converts at a premium.
• Self-reported attribution trends. Track the share of leads naming AI Search tools over time. It is your early warning for how fast the channel grows.
• Branded search volume. Growing branded search shows that your content and AI visibility are building recognition that converts later.

Traffic and rankings still have a place as operational metrics for the SEO team, for diagnosing problems and tracking execution. They are not the numbers to show leadership or to base budget decisions on.

If you take one thing from this: before anyone moves SEO budget, find out where your leads actually came from. Add the free-text field, then decide.

FAQ