The Infrastructure of B2B Discovery Changed. Most Companies Are Still Marketing to the Old One.

The way B2B buyers discover, evaluate, and choose vendors has been structurally rebuilt over the past eighteen months. The platforms changed. The mechanics changed. The economics of attention changed. And the evidence is no longer speculative. It's sitting in a stack of research from independent sources that all point in the same direction.

Most logistics companies are still marketing as if the old infrastructure is intact. Consider what that looks like in practice. A mid-market 3PL allocates $180K to trade show booths at MODEX and Manifest. Their blog has fourteen posts, the most recent from October 2023, each one a 600-word summary of a topic their competitors have also covered. Their website services page says "end-to-end supply chain solutions," a term interchangeable with that of the next 40 companies in their peer group. Their sales team runs outbound sequences triggered by an intent platform, calling accounts that "surged" on warehousing topics, reaching voicemail 90% of the time. Their CMO reports pipeline metrics that haven't meaningfully moved in three quarters and is being asked why the $45K intent data contract isn't producing results.

That company isn't doing anything unusual. They're doing what most logistics companies do. The problem is that every one of those activities was designed for a discovery infrastructure that no longer exists in the form they assume. And for logistics, where buying decisions are high-stakes, relationship-driven, and happen maybe twice a decade per account, the shift hits harder than in categories with shorter purchase cycles and lower switching costs. When a logistics buyer makes a choice, they're committing to a partner that will touch their customers' experience for years. The bar for making that shortlist is higher. The cost of being invisible is steeper. And the infrastructure that determines visibility just changed.

This piece lays out what changed, what the data says about it, and what it means for any logistics company that plans to grow over the next five years.

Search Fragmented. The Map Got Bigger.

SparkToro and Datos analyzed millions of US desktop searches in Q4 2025 and found that search now happens across 41 or more platforms. Traditional search engines (Google, Bing, Yahoo) still account for 80.76% of search volume. Still, the remaining 19.24% is distributed across commerce platforms like Amazon and Walmart (9.76%), social networks like YouTube, LinkedIn, Reddit, and TikTok (5.37%), and AI tools like ChatGPT, Claude, and Deepseek (3.19%).

That 3.19% for AI tools will get dismissed by anyone who doesn't understand compounding. But the important thing about that slice is not its size. It's its influence on everything else in the pie.

When a VP of Supply Chain asks ChatGPT for 3PL recommendations and receives a list of 5 companies, that list doesn't stay within ChatGPT. It travels with the buyer to Google, LinkedIn, peer conversations, and every subsequent search they conduct for the next seven months. The AI answer doesn't capture the majority of the buyer's research behavior. It SHAPES it. It sets the initial shortlist that every other interaction either confirms or fails to displace.

Meanwhile, Zyppy's analysis of the Datos data shows that roughly 60% of Google searches now produce zero clicks to third-party websites. Google is answering more queries directly, through AI Overviews, featured snippets, and knowledge panels. The traffic that does flow to websites is increasingly concentrated among sources that Google's systems trust enough to cite.

For a logistics company, this means the old assumption, "rank on Google and the traffic will come," is operating on a map that no longer matches the territory. The territory now spans 41 platforms, 60% of Google searches don't send anyone anywhere, and a growing share of initial discovery is happening on AI platforms that most logistics companies have never thought about optimizing for.

The Machines Became the Gatekeepers

TollBit's State of the Bots report documented what's happening beneath the surface. At the start of 2025, there was 1 AI bot visit for every 200 human visits to a website. By the end of the year, that ratio was 1 to 31. AI bot traffic to websites is growing at an average quarterly rate of 26%. Human traffic is declining at a rate of 2% per quarter.

The bots doing the heaviest reading aren't training crawlers. They're RAG bots, the retrieval-augmented generation agents that power real-time AI answers on ChatGPT, Perplexity, and Gemini. TollBit found that RAG bots scrape each page roughly 10 times as often as training crawlers. They can't store the entire internet, so they return every time a new query is issued. Your content isn't read once and filed. It's continuously evaluated whenever a buyer asks a question that touches your category.

And B2B professional content is the single most scraped category on the internet, up 62% in six months. More than news. More than entertainment. More than lifestyle. The machines are hungriest for exactly the kind of content logistics companies produce, or should be producing. For an industry that has historically underinvested in content relative to SaaS and fintech, this is both the problem and the opportunity. The 3PL with deep operational expertise that has never published any of it is sitting on a gold mine that the machines can't access. The logtech company that has published extensively about the problems its platform solves is being scraped, cited, and recommended, while larger competitors remain invisible.

AirOps and Kevin Indig studied what determines whether those machines actually cite you. Their analysis of 353,799 pages and 6.8 million headings found something that should change how every B2B company writes content: the H2 heading is doing more work than the body copy. Pages with H2S that closely match a user's query earn a 41% citation rate. Pages with loosely related headings drop to 29%. The AI treats your H2 as the query and the paragraph below it as the answer. If the heading matches how buyers actually search, you get cited. If it reads like a generic section label, you get skipped.

Their structural study of 12,000+ URLs went further. Pages with sequential heading hierarchy (H1 to H2 to H3, properly nested) are 2.8 times more likely to be cited by ChatGPT. Pages with rich schema (three or more types) showed up in 61% of ChatGPT citations versus 25% in Google-only results. Pages with well-organized list structures appeared in cited content at 17 times the rate of uncited pages.

The machines aren't just reading your content. They're grading it on structure, specificity, and whether it actually answers the question a buyer asked. Most logistics websites fail that test, not because their services are bad, but because their content was never built to pass it.

The Citation Economy Replaced the Ranking Economy

Seer Interactive's 2026 AIO study across 53 brands and 5.47 million queries put hard numbers on what citation means for traffic. Being cited in Google's AI Overview delivers 120% more organic clicks per impression than not being cited. Per million impressions on informational queries: 33,500 clicks if no AIO is present, 20,743 if you're cited in the AIO, and 9,445 if the AIO appears and you're not in it.

The gap between cited and not-cited is now the primary dividing line in organic search. Ranking on page one still matters, but if an AI Overview appears and you're not in it, you're competing for less than half the clicks the cited sources receive.

Seer's team, led by Tracy McDonald, framed it precisely: "The goal is not to answer the question. The goal is to be the source Google points to when it answers the question."

This finding maps directly to what we measure with the Logistics AI Search Visibility Index. When we tested 127 logistics companies across 500+ prompts on ChatGPT, Gemini, and Perplexity, we found that 38% of companies scored below 5% visibility. The AI platforms knew they existed. The RAG bots had visited their sites. But the content wasn't authoritative enough to cite.

The companies at the top of LASVI, above 35%, the ones that appear consistently as default answers in their category, shared a pattern that had nothing to do with company size or ad spend. Pillar content architecture increased visibility scores by 3.1x. Executive thought leadership with a named human behind it produced 2.8x. Original industry reports with proprietary data produced 2.4x. Content averaging over 1,500 words with genuine substance produced 1.9x.

Every one of those multipliers requires first-party expertise. You can't build a pillar page on temperature-controlled LTL consolidation without operational experience in temperature-controlled LTL. You can't publish an industry report with proprietary benchmarks unless you have proprietary benchmarks. The AI isn't looking for the best-written page. It's looking for the page that comes from somewhere real.

Cyrus Shepard's analysis at Zyppy independently validated this by ranking 17 content types by their survivability in the zero-click era. The types most likely to endure: original research, owned audience channels, expert perspective content, and in-depth reviews with firsthand evidence. The types most likely to be displaced: generic guides, FAQ pages, listicles, commodity blog posts. Most logistics companies are producing almost exclusively from the bottom of that list and wondering why nothing is working.

The Content Flood Made Expertise More Visible

Ahrefs analyzed 900,000 newly published web pages in April 2025 and found that 74% contained AI-generated content. Three out of four new pages on the internet are produced with AI assistance, drawing from the same training data, producing the same default outputs, flooding the same channels with content that reads fine and says nothing original.

This flood is making genuine expertise more visible, not less. When 74% of new content is a variation of the same AI-synthesized average, the 26% that comes from first-party knowledge stands out like a signal fire in fog.

Harvard Business School and BCG studied this dynamic directly. Their experiment with 758 consultants found that AI made knowledge workers 25% faster and boosted output quality by 40%, but only on tasks inside AI's capability frontier. The consultants who deferred to the machine, who let it do the thinking rather than using it to accelerate their own judgment, saw quality decline. The tool made them faster at being wrong.

A Stanford, MIT, and NBER study tracking 5,179 workers confirmed the asymmetry. AI boosted novice productivity by 34%, making a two-month employee perform like one with six months of experience. But for experienced, highly skilled workers, the impact was minimal. The researchers' explanation: "High-skilled workers may have less to gain from AI assistance precisely because AI recommendations capture the knowledge embodied in their own behaviors." The AI was trained on what experts already know. It gave novices access to the average of expert behavior. For the experts themselves, it was a mirror, not a lever.

Max Bernstein synthesized this into three emerging classes: domain experts using AI (pulling ahead), domain experts not using AI (good work, falling behind on velocity), and AI users without domain expertise (getting exposed). His observation of the third group carries the most weight for logistics marketing: "The tool that was supposed to help them is now the thing that gives them away."

In a market where buyers compare what they read against twenty years of operational experience, content without genuine expertise behind it collapses the moment someone pressure-tests it. A shipper reading a blog post about warehousing optimization can feel the absence of first-party knowledge the way a musician hears an instrument slightly out of tune. AI made it possible for anyone to produce content that sounds authoritative. It also widened the gap between sounding authoritative and being authoritative, making it more consequential than ever.

The Buyer Changed Too

The infrastructure shift isn't just about platforms and algorithms. The buyer's behavior changed alongside it.

Dreamdata's 2026 research found that B2B customer journeys now average 211 days and 76 touches before a purchase. Not calls. Not meetings. Touches. LinkedIn posts read and never engaged with. AI answers that named a company. Case studies are half-finished and closed, leaving no trace. Peer recommendations in Slack channels that no analytics platform will ever see.

NetLine's 2026 Content Consumption Report added another dimension: the average B2B buyer takes 47.7 hours to open content they actively requested. That's the Consumption Gap, and it widened 43.2% since 2021. A buyer downloads your case study and doesn't open it for two days. The signal that fires on your intent platform has nothing to do with whether they've engaged with what you sent them.

And 6sense's 2025 Buyer Experience Report found that this should reframe every conversation about pipeline: buyers shortlist an average of 4 vendors on Day One, and 95% of the time they purchase from someone on that initial list. The vendor that contacts first wins roughly 80% of deals.

Combine these three data points, and the picture becomes uncomfortably clear. The buyer builds their shortlist during the invisible months, shaped by AI answers, content encounters, and peer conversations that produce no trackable signal. By the time the intent platform flags a "surge," the shortlist is already set. By the time the buyer opens your content, two days have passed. By the time your SDR calls, the decision is already made in the right direction.

TollBit's referral data confirms the invisibility. Click-through rates from AI applications collapsed from 0.8% in Q2 2025 to 0.27% by year's end. The buyer sees your brand name in a ChatGPT answer and doesn't click through. They don't visit your website. They don't fill out a form. They add you to a mental list and move on. The influence is real. The attribution is impossible. That's 211 days of invisible brand building that no dashboard will ever capture.

The Constraint Nobody Talks About

Georgson & Co published research this year that reframes the growth conversation entirely. Their thesis: growth is not just about demand. It is constrained by availability. No firm, no brand, can outgrow its mental and physical availability.

Physical availability means the buyer can find you. Mental availability means the buyer recalls you when a need arises. Growth happens when a company expands the set of situations in which it is easy to buy from them and easy to recall them. Many companies have demand. Few expand availability. That is why growth is rare, uneven, and hard to sustain even in competent organizations.

Bill Chidley, author of *The Brand Vortex* and former VP of Strategy at Interbrand, adds the financial language: brand is a store of value. A reservoir of accumulated belief, the business either fills or drains. Every piece of content that builds authority, every AI citation that names you, every LinkedIn post that makes a buyer pause for three seconds is a deposit. Every generic SDR sequence, every undifferentiated outreach, every premature cold call to a "surging" account is a withdrawal from a reservoir that may not have enough in it.

Todd Kaplan, CMO of Kraft Heinz, offered perhaps the most useful frame for how availability actually builds: "Brand building is a bit like pointillism. Every time you activate your brand, you're placing a dot in somebody's brain." Those dots cluster over time to form richer perceptions. Placed with intent, they become a coherent picture. Scattered at random, they become noise.

For logistics companies, the availability constraint is the real reason growth stalls. The service may be excellent. The pricing may be competitive. The operations team may be world-class. But if a buyer can't find you on the platforms they search, can't recall you when a demand point fires, and never encounters your name in an AI answer, the growth ceiling has nothing to do with your service quality. It has everything to do with whether you exist in the places where decisions get made.

What LASVI Actually Measures

This is the context that produced the Logistics AI Search Visibility Index. When we built LASVI, we weren't building an SEO tool. We were building a diagnostic for the availability constraint in the AI age.

LASVI measures the percentage of relevant AI prompts where a logistics company is cited, recommended, or referenced across ChatGPT, Gemini, and Perplexity. It answers a question no other metric in logistics marketing currently addresses: when a buyer asks an AI for help in your category, does your brand appear in the response?

The 127 companies we tested showed a distribution that closely matches Georgson's availability thesis. 38% scored below 5%. Invisible. They had demand for their category, but zero availability on the platforms where a growing share of buyer discovery begins. The industry median was 14%. Visible on niche queries, absent from the broader category conversations. The top quartile averaged 41%. These are the companies AI treats as default answers. They show up consistently across platforms and query types because they built the kind of published authority that AI systems recognize as definitive.

The gap between 5% and 41% is not a marketing budget gap. We found companies with $6.82 billion in revenue scoring invisibly on Gemini, and companies a fraction of that size appearing on every platform. The gap is an availability gap. The low-scoring companies had websites full of press releases, spec sheets, and generic service descriptions. The high-scoring companies had deep pillar content, including executive perspectives, original research, and case studies with real metrics. The content types that Shepard ranked as Strong in his zero-click survivability framework. The structural patterns that AirOps identified as citation drivers: query-matching headings, sequential hierarchy, and rich schema.

We've watched this play out across our own client work. A last-mile delivery platform came to us as a startup with no marketing infrastructure. Over nearly three years of pillar content development, executive thought leadership, and systematic SEO, they accumulated 185 AI mentions across ChatGPT, Gemini, and AI Overviews with 433 citations. 230 page-one keywords. When they raised a Series C, the co-founder requested our impact data for investor due diligence, because the AI visibility and organic authority we built had become part of the company's valuation story.

Every piece of the new infrastructure, the search fragmentation, the machine gatekeepers, the citation economy, the content flood, the invisible buyer journey, the availability constraint, connects to a single measurable outcome: are you in the answer when the buyer asks the question?

What This Means Going Forward

The logistics companies that will grow over the next five years are the ones that recognize the infrastructure shift and build for it. The ones operating on the old blueprint will spend more every year to achieve less, and they'll blame the market, the sales team, or the tools, because the real cause, an availability constraint they can't see in any dashboard, will remain invisible to them until it's too late.

The migration is from content that gets displaced to content that gets cited. From generic guides and FAQ pages that AI summarizes and replaces, to original research, executive thought leadership, and case studies built on real outcomes that AI systems recognize as authoritative and buyers recognize as genuine. From commodity blog posts written for keywords nobody searches anymore, to pillar content with query-matching headings that earn a 41% citation rate because they answer the exact question the buyer typed. From hoping Google sends traffic, to building the kind of published authority that 41 platforms, traditional search engines, AI tools, and social networks all reward.

The measurement has to change with the strategy. LASVI tracks whether you exist in AI answers. Seer's citation framework tracks whether Google's AIO trusts you enough to cite. Traditional organic metrics still matter, but a company with flat traffic and rising LASVI scores is gaining invisible influence that will surface in the pipeline within 6 to 12 months. A company with strong traffic and zero AI citations is living on reserves that are quietly being depleted.

Todd Kaplan, the CMO of Kraft Heinz, described brand building as pointillism: every activation places a dot in somebody's brain, and those dots cluster over time into richer perceptions of who you are. For logistics companies, the canvas just got bigger. The dots now include AI citations the buyer never clicks, LinkedIn posts the buyer never likes, case studies the buyer reads halfway and closes, and peer conversations the buyer has in rooms you'll never enter. Every one of those moments is a dot. Placed with intent, over months and years, they form the picture that determines whether you make the shortlist when a demand point fires.

The infrastructure changed. The evidence is not ambiguous. Seventeen independent sources, from Harvard to TollBit to Seer to Georgson, all arrived at the same conclusion from different angles: visibility in B2B is no longer a function of ranking, budget, or company size. It's a function of whether you built something the machines trust enough to cite and the buyers trust enough to remember.

Every logistics company is either placing dots or producing noise. The ones who know the difference are already five years ahead.

‍