Your traffic dropped. Your pipeline didn't have to.

This was early 2025, when AI Overviews were showing up everywhere, and one thing jumped out right away. Traffic was dropping. The alligator chart was opening up in the Search Console Performance report, with impressions climbing while clicks fell, and everyone internally started to panic. What does this mean for our clients? For our websites? For our business?

That panic was reasonable. Nobody knew yet what AI search would do.

But once we looked past the traffic line, the rest of the data told a different story. Impressions were up. Conversions held. Leads never dropped.

For one client, that was still a red flag. Organic traffic was the number one KPI our point of contact reported up to leadership. Our account manager brought me the problem in one line: "But the client wants traffic!" Traffic was dropping everywhere, and every study said so. I told him there was nothing more we could do about it. This was our new normal. What we could change was the report. So I changed what I showed them. I put leads and conversions right next to traffic, and showed their visibility against the competitor set. They were still number one, and it wasn't close.

The client didn't care about the headlines or the LinkedIn hot takes. They cared about results, and the results were fine. The scorecard was the problem.

That's the gap most teams are reporting into right now. A traffic decline and a pipeline decline are two separate problems, and the SEO KPIs you report decide which one leadership sees.

The SEO work itself hasn't changed as much as the headlines suggest. AI search still runs on crawling, indexing and ranking, which is why SEO is not going to die. What changed is where the buyer reads the answer, and what you need to measure to see it.

Why is organic traffic declining?

US organic search visits fell 13.2% from August 2024 to August 2026, according to Kevin Indig's Growth Memo. AI answers explain part of that drop.

In a Pew Research Center study of 900 US adults, Google users who saw an AI summary clicked a traditional result in 8% of visits. Without a summary, they clicked 15% of the time. Across all Google searches, SparkToro found 68% of US searches ended without a click in the first four months of 2026, up from 60% in 2024.

Paid search moved too. US paid search traffic grew 21% over the same two years Indig measured. He also points to heavier ad load, aggressive algorithm updates and people simply searching less. Meanwhile AI referrals make up just 2.85% of US SEO traffic and 0.5% of total visits.

Indig is careful here, and you should be too. His data doesn't establish whether AI or paid search caused the organic losses. If your board deck blames everything on ChatGPT, someone will eventually ask for the proof.

Whatever the mix of causes, zero click search isn't going away, so don't build next year's plan around traffic coming back.

Can you stop losing traffic to AI search results?

Not fully. In the same Pew study, users clicked a link inside Google's AI summary in just 1% of visits to pages that showed one. Of the 68,879 searches Pew tracked, 12,593 produced an AI summary.

Optimization won't change that. What you can control is whether your brand shows up inside the answer.

G2 surveyed 1,076 B2B software buyers in March 2026 and found 51% now start their research with an AI chatbot more often than with Google. A year earlier it was 29%. And 69% of those buyers chose a different vendor than they first planned because of what the chatbot recommended.

So a zero-click search still does work. It shapes the shortlist before the buyer ever lands on a website. HubSpot tracks two signals for this: mentions, when an AI answer names the brand, and citations, when it links to it. Both count toward AI visibility, even when neither shows up in GA4.

Google also says clicks from results pages with AI Overviews are higher quality, with visitors spending more time on site. It hasn't published data behind that claim, so treat it as a hypothesis to test in your own analytics.

HubSpot lost 80% of its blog traffic and grew revenue

HubSpot is the case every SEO has heard about. Its blog peaked at more than 20 million monthly organic visits and then lost about 80% of them. In 2025, HubSpot's revenue grew nearly 20%, as CMO Kipp Bodnar explained to Search Engine Journal.

The numbers underneath explain how both things can be true:

• Qualified leads from AI answers grew 1,850% from Q1 2025 to Q1 2026.
• AI answer visitors converted at about three times the rate of traditional search visitors.
• Deals from those visitors closed in about half the time.
• About 90% of HubSpot's content leads now come from channels that barely existed in its mix five years ago. The blog drives about 10%.

Two caveats before you paste this into a deck. SEJ notes the 1,850% likely starts from a small base, and the attribution is correlative. HubSpot is also a brand with enormous AI visibility to begin with.

HubSpot isn't a one-off. In Search Engine Journal's State of Search survey, published in September 2026, 39% of search professionals saw traffic fall or stay flat while leads and conversions held or improved. Only 15% lost both.

In October 2026, 97th Floor interviewed 12 in-house marketers at Ahrefs Evolve. U.S. Career Institute's traffic is down about 30%, while its leads and enrollments are up. ClearPoint Strategy has fewer MQLs and higher average contract values. Seven of the 12 said their leadership focuses on revenue, pipeline, leads or customers.

The pattern still holds outside software. Adobe found that in September 2026, AI traffic to US retail sites generated 83% more revenue per visit than non-AI traffic and converted 90% better.

Traffic still matters. It was just always a proxy. It measured attention, and we treated it like demand. When the answer moves to the results page, the low-intent visitors disappear first. The ones who still click are further along and more likely to buy.

Which SEO KPIs should replace sessions?

Five SEO KPIs replace sessions as the headline number: search visibility, AI citation share, AI share of voice, revenue per visit and search-influenced pipeline. HubSpot made a similar swap, trading traffic for revenue per visit, AI share of voice and content-influenced pipeline.

• Search visibility. Impressions and average position, AI Overviews and AI Mode included. Source: Google Search Console, Web search type.
• AI citation share. The share of tracked prompts where an AI answer links to your brand. Source: an AI visibility tracking platform.
• AI share of voice. How often AI answers name your brand compared with competitors. Source: an AI visibility tracking platform.
• Revenue per visit. The value of the visits that still arrive. Source: GA4 and your CRM.
• Search-influenced pipeline. Opportunities and revenue that touched organic or AI referrals. Source: your CRM, such as HubSpot.

One gotcha on the first one. Search Console counts AI Overviews and AI Mode inside the Web search type and doesn't break them out. You'll see the impressions. You won't see which ones came from an AI answer.

Sessions stay in the report, just lower on the page, next to the numbers that explain them. A CFO who sees traffic down next to pipeline up asks a very different question than one who only sees the traffic line.

How should you report AI citations?

Report AI citations as averages over several weeks, never as a daily snapshot. Ahrefs found that 50% to 83% of AI citations disappeared by the next day, depending on the platform.

Only 9% of respondents in the SEJ survey said they feel very confident measuring and interpreting AI visibility.

The study tracked about 37,500 citations across five AI tools every day for 20 days. Next-day loss was 50% on AI Overviews, 58% on AI Mode, 59% on ChatGPT, 81% on Gemini and 83% on Copilot. About 75% of the citations that vanished came back within the 20 days.

If you report a single day, you're reporting noise. A citation that dropped on Tuesday may be back by Friday. Monthly averages give leadership a trend line instead of a coin flip.

Pages that ranked in Google were about seven times more likely to keep an AI Overviews citation for a full week, 29% compared with 4%. AI Mode showed the same gap, 28% compared with 4%. Ahrefs calls this a correlation, not a proven cause.

Still, the direction is clear enough to act on. Rankings and durable AI visibility travel together. That's the case for funding SEO and AI search as one program instead of two competing line items, which I've made before in AI Overviews: A straightforward approach.

How to build the scorecard

Most teams already have the tools. The work is five measurements, set up once and reported the same way every time.

1. Search visibility from Search Console. Use the Web search type and split branded from non-branded queries with a regex filter. Track impressions and average position month over month, and remember that AI Overviews and AI Mode are blended in.
2. AI referrals in GA4. Create a custom channel group for AI assistants that matches sources like chatgpt.com, perplexity.ai, gemini.google.com and copilot.microsoft.com. Place it above Referral so those visits don’t get lumped in. Then compare revenue per visit, or conversion rate if there’s no revenue, against organic.
3. AI citation share and share of voice. Build a set of 50 to 100 prompts from real buying questions, the ones sales hears on calls. I track them in Ahrefs, Profound and Writesonic. Report the monthly average, never a single day.
4. Search-influenced pipeline in your CRM. Whether the site runs on WordPress, Drupal or Webflow, every lead form should capture UTMs and the original referrer in hidden fields, plus a “How did you hear about us?” field, all passed into the CRM. I typically use HubSpot, where those land on the contact record and roll up to the deal, so you can report on which opportunities touched organic search or an AI referral. I set this up for every client, and when it’s missing, it’s the first thing I recommend. That free-text field is where “ChatGPT recommended you” shows up, even when the referral data doesn’t.
5. Report on a set rhythm. Monthly reports show the trend. QBRs are where you tie it to pipeline and budget. I don’t recommend weekly reporting. AI citations swing too much, and weekly numbers turn every dip into a fire drill.

What to tell your CFO

Next month's report is coming either way. Lead it with sessions and you'll spend the meeting explaining a decline. Lead it with visibility, AI citations and pipeline, and you'll be explaining a win.

If your agency still leads its monthly report with sessions, it's protecting itself, not you. If your SEO still leads with clicks, they're already behind. And if that's who you're paying, look elsewhere.

That's what happened with my client. The SEO work stayed the same. The scorecard changed, and so did the conversation. A program measured that way is one a CFO keeps funding.

If your current scorecard still leads with sessions, book a search measurement review. You'll walk away with a scorecard your CFO can follow.