Happy Triple Threat Thursday.
Here’s one Signal to notice, one thing to Spark growth and one Shift to consider.
This week's theme: Bots crossed 50% of all web traffic last year. That's a startling shift in who your website is actually for.
Google announced background information agents on May 19 that scan the web around the clock on a buyer's behalf, without a search query required. They're arriving at your website whether you know it or not. That's not new behavior. It's now permanent infrastructure. Most operators are optimizing their website for a visitor type that has become secondary.
📡 Signal — What’s Changing
Why are AI agents now the primary readers of most business websites, and what does that mean for how buyers find you?
For most of the internet's existence, a website visitor was a person. Someone who found a link, clicked it, read what was there, and decided what to do next. That model is no longer accurate.
According to HUMAN Security's 2026 State of AI Traffic report, bots now account for 51 to 52% of all global web traffic, crossing the 50% threshold for the first time in 2024. AI bot traffic specifically grew 187% from January to December 2025. Human traffic grew 3.1% over the same period. The fastest-growing category isn't training crawlers scraping content to build language models. It's agentic bots, real-time systems performing tasks on behalf of users, which grew 7,851% in 2025. Cloudflare's CEO put the projection plainly at SXSW in March 2026: AI bots will exceed human web usage in the near term, based on traffic data from roughly 20% of all global websites.
Google formalized this on May 19. Background information agents now scan the web continuously, delivering synthesized updates to users without a search query required. A buyer researching vendors in your category may have an agent monitoring that space right now. That agent is reading your website. It is synthesizing your positioning. It's doing in seconds what a human researcher would spend an hour doing, and it's doing it before any human has taken an action you can see.
At the same time, 60% of Google searches end without a click to any external website. When an AI Overview answers the question, the buyer doesn't need to visit. Your content was read. A verdict was rendered. No session appeared in your analytics.
Why it matters now:
A website built to convert a human visitor is a different object than a website that earns citation authority with AI agents. The human-optimized version has a hero image, social proof, testimonials, and a contact form. The agent-readable version has specific plain-language answers to the questions buyers actually ask, consistent factual signals across every external source, and enough structured specificity that an AI tool can summarize you accurately in two sentences. Most websites are built for the first version. The primary reader is increasingly the second kind of visitor.
What to do this week:
Ask a simple question: if an AI agent landed on your homepage right now, what would it conclude? Not from the design, not from the photography, not from the testimonial slider. From the text. What problem do you solve? For what kind of company? What results follow? If the text doesn't answer those questions clearly in the first 200 words, the agent moves on.
Fewer than 1 in 20 business websites answer all three questions plainly anywhere above the fold. That's not a guess. That's what hundreds of website audits for operators at this revenue stage actually show. The information exists somewhere on the site. It just isn't where the first reader, human or machine, is likely to find it.
⚡ Spark — What to Try This Week
How can operators check whether their website communicates clearly enough for AI agents to cite them accurately?
This three-part check takes about 30 minutes. The goal is to read your company the way an AI research agent reads it, not the way your designer intended it.
Open an incognito browser and search your company name. Read the AI knowledge panel or summary before any organic links. This is what AI tools synthesize when a buyer asks about you. Look for three things: what problem you solve, who you work with, and what results are mentioned. If the result says "a company that provides services to businesses" or returns nothing coherent, the agent cannot cite you usefully. That's the citation gap in plain view.
Search one of your primary service categories as a question a buyer would actually type: "What are the best logistics partners for food distribution companies in California?" or "Which HR consulting firms specialize in manufacturers with 50 to 200 employees?" Look at what appears in AI Overviews. Note which companies are cited and what those companies have written clearly that you haven't. This is your competitive gap in the research phase.
Check your Google Business Profile. Verify the category is specific, the description leads with the problem you solve rather than a mission statement, and the most recent reviews have a response from your team. Google's Personal Intelligence feature, now live in nearly 200 countries, incorporates this data into personalized AI results. A consistent, responded-to Business Profile is a citation signal, not just a local SEO tactic.
Why it works:
AI tools cite specificity. "We provide managed IT services to companies in the Sacramento region with 25 to 150 employees who need compliance-ready infrastructure" can be cited precisely. "We help businesses succeed with technology" cannot. The content on your website was written to persuade a skeptical human. That isn't what an agent needs. An agent needs factual density, plain language, and external verification. The check takes 30 minutes. What you find tells you whether you exist in the research phase or not.
🔄 Shift — How to Rethink It
Is your website still serving the function you designed it for, or has its primary reader already changed?
Default belief: Our website is for our customers.
Flip: Your website's primary reader is no longer a human deciding whether to contact you.
A professional services firm spent $45K on a website redesign two years ago. Photography, brand feel, mobile experience, conversion rate optimization. Human visitors converted meaningfully better after launch. The team felt good about it. Then their traffic reporting started showing something odd: overall sessions were flat, but inquiries from new prospects, people who had never heard of them before, were declining. Not referrals. New discovery. An audit showed that 60% of searches in their category were returning AI Overviews with competitor names cited. The new website hadn't touched that problem at all.
The redesign optimized for the visitor who arrives. It did nothing for the buyer who never had to.
Why it matters:
This isn't a website design problem. It's a mental model problem. Operators built websites for humans, measured them by human behavior, and invested in them to improve human conversion. That model was accurate for 25 years. It is less accurate now, and it will be less accurate next year than it is today. The investment calculus for what a website is supposed to do is changing, and most marketing budgets haven't reflected that yet.
Read your homepage the way a machine would. Open the page source or copy the text into a plain document. Strip everything except the words. Does the remaining text answer what you do, for whom, and with what result? If not, that's where to start.
Search your category the way your best-fit buyer would phrase it to an AI tool. See who gets cited. If you aren't among them, ask what those companies have written that you haven't.
Ask your marketing agency or ops lead one specific question: "What specifically are we doing to improve how AI agents summarize our company?" If there's no concrete answer, the research phase has no strategy.
The visitor who never clicked is still forming an opinion about you.
📚 Worth A Look
What should you be reading about AI content and audience trust this week?
TechCrunch's May 19 coverage of Google I/O explains background agents that gather information continuously on users' behalf, the most consequential change to how buyers encounter vendors since the search engine itself.
CNBC's March 26 summary of HUMAN Security's 2026 AI traffic report, covering the data behind bots crossing 50% of web traffic and agentic AI growing nearly 8,000% in a single year.
McKinsey's 2026 marketing data: 72% of CMOs are increasing brand budgets while the research channel buyers actually use most sits 17th on the priority list.
📈 TL;DR
Bots now account for the majority of global web traffic. The fastest-growing category is agentic AI doing real-time research on buyers' behalf. Your website was built for the 49%. The 51% has different requirements.
📈 One Question
If an agent read your website homepage right now and had to summarize your company in two sentences, what would it say?
Thanks for reading Triple Threat. See you next Thursday with another Signal, Spark, and Shift.
— Alexandria Ohlinger
p.s. If this helped you think sharper or move faster, share it with someone who builds the way you do. And if you want more practical insight between issues, connect with me on LinkedIn or schedule a strategy session.