The invisible funnel: What the data says about how logistics buyers actually find you in 2026

A logistics marketing report can now be completely accurate and still tell you almost nothing.

Sessions flat. Form fills down a little. Meanwhile the buyer who will sign a contract with you in March is already reading about your category and already asking an AI assistant which providers are worth considering. None of that activity shows up in the report.

The measurement system most logistics companies run was built for a web where discovery produced a visit. That link between discovery and visit has been coming apart for 3 years, and the 2026 data shows how far it has gone.

What follows is the evidence, source by source, and what we think it means for how a logistics company should spend and measure marketing money this year.

1. Two out of three Google searches now end without a click

The clearest number comes from SparkToro's June 2026 study, built on Similarweb clickstream data. Across January through April 2026, 68.01% of US Google searches ended without a click to the open web.

The trajectory matters more than the number. In 2016, roughly 45% of US Google searches were zero-click. By 2019 it was about 49%. In 2024 it reached 60.45%. In 2026 it stands at 68.01%.

That is 12.5% growth in clickless queries in 2 years, which Rand Fishkin calls the fastest acceleration of the phenomenon in a decade.

The mechanism is AI Overviews. Ahrefs research puts AI Overviews on more than 20% of all searches, and their follow-up study found that when one appears, clicks fall by nearly 60%. Seer Interactive's 2026 analysis puts AI Overviews on roughly 1 in 3 informational queries.

Publishers feel it first. The Ahrefs traffic tracker, which follows more than 75,000 opted-in domains, recorded an 8 percentage point decline from June 2025 to May 2026, a 22% drop in the share of traffic Google sends to those sites in 12 months. Those are sites with professional marketers actively working to grow traffic.

Google has said AI Mode passed 1 billion monthly users, with queries more than doubling every quarter. The pressure on the click is not easing.

2. The machines are reading your site more than some of your prospects are

While human visits fall, a different kind of reader arrives.

TollBit's State of the Bots research, drawn from its publisher network, found that by Q4 2025 there was roughly 1 AI bot visit for every 31 human visits, up from 1 per 200 at the start of that year. Human visits across the network declined 9.4% over the same period. The Register's coverage framed it plainly: bot traffic is closing in on human web traffic.

Two details in that research matter for logistics companies specifically.

Retrieval bots overtook training bots. RAG bots, the crawlers that fetch live pages to answer a question being asked right now, now scrape more than the crawlers that collect training data. This is the difference between a machine studying the web last year and a machine reading your site this afternoon because a shipper asked a question about your category.

B2B and professional content is among the most heavily scraped categories in the entire index. Technology and consumer electronics content saw scraping grow 107% in the period TollBit measured. The competition for machine-readable authority in commercial categories is already underway.

The practical consequence: your website has a second audience now, and its entire job is deciding whether your pages are worth citing when a buyer asks a question.

3. Your buyers ask a machine before they ask you

Research from LinkedIn's B2B Institute found that 94% of B2B buying groups consult a large language model before they speak with sales.

Mimi Turner, who leads that work, describes these systems as retrieval rather than discovery. They assemble an answer out of what already exists: customer stories, published expertise, expert commentary, and mentions across the web. A company with a thin public record gives the machine very little to work with, and the machine answers anyway, using whatever it can find about your competitors.

The consultation happens silently, and it happens early.

Seer Interactive measured what it is worth to be part of that answer. Across 53 brands, 5.47 million queries, and 2.43 billion organic impressions, brands cited inside an AI Overview earned 120% more organic clicks per impression than brands the answer left out. Being in the answer is now a measurable traffic asset, and being absent is a measurable tax.

4. Even the buyers who raise their hands are not ready to talk

Nothing is faster on the other side of the form, either.

NetLine's 2026 State of B2B Content Consumption report found that the average B2B buyer takes 47.7 hours to open content they requested themselves, a gap that has widened 43% since 2021. Two full days pass between the form fill and the first look, which makes that form closer to a bookmark than a hand raise.

NetLine's intent data points the same direction. Buyers expecting to purchase within 12 months rose 17.7% year over year, while immediate 0 to 3 month intent fell 15.7%. The pipeline is lengthening, not compressing.

Dreamdata's 2026 analysis of B2B journeys puts the full path at roughly 211 days and 76 touches before purchase. In a 211-day journey, the touch that gets attribution credit is simply the last one that happened to be measurable.

5. The paradox in our own client data

Here is where the picture turns from discouraging to actionable.

Across client accounts we manage, Google Analytics began classifying AI-assistant referral traffic this year, which let us see a channel that used to hide inside direct traffic. The volumes are small. The behavior is not.

In one logistics client account, 3 weeks of tracking produced 19 visits from AI assistants in July, and those 19 visits produced 2 form submissions. In a separate account in a different sector, we measured AI-assistant traffic converting at roughly 10% to form fill against roughly 2% for traditional organic search.

Small samples, and we treat them as directional rather than definitive. But the direction is consistent with everything above and with what TollBit measured on referral behavior: AI referral click-through rates have collapsed to a fraction of a percent, so the few people who do click through have already been pre-qualified by the answer they read.

The visitor arrives having been told, by a system they trust, that your company is a credible answer to their question. That is a warmer introduction than any ad has ever purchased.

This is the shape of the invisible funnel. Enormous influence, tiny measurable traffic, and conversion rates that look like a data error until you understand what happened before the visit.

6. What this changes about where the money goes

Four implications follow from the data, and none of them require abandoning what already works.

Published expertise now works like infrastructure. Retrieval systems can only cite what exists. Every substantive piece your company publishes, and every trade article, podcast appearance, and industry panel that carries your name, becomes retrievable material. The 120% citation advantage Seer measured accrues to companies with a public record deep enough to be worth quoting.

Volume stopped being the lever. With most new web pages now carrying AI-generated content, undifferentiated publishing adds noise to an index that is already saturated. What retrieval systems and human buyers both reward is specific, verifiable, first-hand expertise: real operating numbers and real customer situations, attached to a named expert who can defend them. Your operators know things no model can generate, because those things have never been written down.

Brand search becomes the honest scoreboard. When discovery stops producing clicks, the reliable signal that marketing is working is whether more people are looking for you by name. Google Search Console and Google Trends both show this at no cost. A branded search is a memory, typed.

Attribution deserves a demotion. In a 211-day journey where 94% of buying groups consult an AI system privately and 68% of searches produce no click, no attribution model will capture the causal chain. Correlation over time between marketing investment, branded search, and qualified pipeline is a more honest instrument than a last-touch report that credits whichever channel happened to be visible at the end.

7. A measurement set that matches the market

For logistics companies rebuilding their marketing scorecard, we recommend tracking these alongside traditional metrics rather than in place of them.

• Branded search volume, quarter over quarter. Lives in Search Console and Google Trends. Tells you whether market memory is growing.
• AI-assistant referral traffic and its conversion rate. Lives in GA4 channel classification. Tells you whether the answers include you.
• Citation presence in AI answers for your core buying questions. Test manually across assistants and log it quarterly. Tells you whether you are in the consideration set.
• Direct traffic trend. Lives in GA4. Acts as a proxy for recall and dark-social influence.
• Share of first sales calls that open with prior awareness. Lives with your sales team, ideally as a CRM field. This is the signal closest to revenue.
• Web mentions and earned placements. Lives in your media monitoring. Tells you what retrievable record you are building.

None of these are vanity metrics. Each one measures a step in a buying process that now happens mostly out of view.

The short version

Discovery kept growing. The click stopped following it.

Two of every 3 Google searches end without a visit. AI systems read commercial websites at a rate approaching 1 visit for every 31 humans, and 94% of B2B buying groups consult one before they ever contact a seller. When those systems name your company, you earn more than twice the clicks of companies they skip, and the handful of people who do arrive convert at rates that traditional channels rarely produce.

The buying is still happening. Logistics companies still switch providers and still choose the name they trust when a situation forces the decision. What changed is that almost all of the influencing now happens somewhere your analytics cannot see.

The companies that grow through this period will be the ones that invest in being known and being citable, and that measure memory instead of sessions.