Good morning.
Uplifting news to start the week: A 25,000-pound electric airplane just took its first flight in upstate New York. It stayed up for 27 minutes and burned about $5 worth of electricity doing it.
Five dollars. That’s less than the coffee you’d buy at the gate while waiting to board it.
ChatGPT may ignore robots.txt and it sets the pace on AI traffic; but Google is not giving up
It feels like a new trend that the AI-related news footnotes carry more weight than the headline.
Here’s what we mean.
Small pond, big splash: Demandbase says monthly ChatGPT referrals to B2B sites climbed from 645,000 to 2.6M year over year.
Impressive, until you notice that’s roughly a third of one percent of the 11B visits it measured.
Perplexity referrals fell. Gemini and Claude stayed flat. So “AI traffic” currently means one company’s product decisions. Mostly.
Ask nicely, get ignored: OpenAI’s documentation says robots.txt may not apply to ChatGPT-User, because a person triggered the fetch.
TollBit found it reached disallowed pages on more sites than any other bot of its kind. Your server logs show what arrived. Your robots.txt only shows what you asked for.
On Google’s end, the sparkle is now optional: It’s adding a toggle to remove visible watermarks from Nano Banana, Omni, and Lyria output.
Invisible SynthID and C2PA metadata stay on, and the EU keeps its mandatory labels. Everyone else gets cleaner creative.
Workhorse upgrade: Gemini 3.7 Flash is now in AI Mode, globally, for Pro and Ultra subscribers searching in English.
The quality of output between AI Mode and the Gemini app still has us sceptical about Google’s way of implementing AI, so this is a welcome upgrade.
Enough about AI though. Let’s change gears and look at a real challenge humans in marketing have…
Your next job may be a content play on social media
Turns out the hardest brief you’ll get this year is the one where the product is you.
The job funnel is broken, and the numbers say so: The average corporate posting pulls 242 applications, 75% of resumes get filtered by AI in under a second, and only 2-3% reach an interview.
So people stopped submitting and started publishing.
Creative strategist Josh Katz posted a 60-second video resume that pulled hundreds of thousands of views, several interviews, and freelance work for months after.
The lesson? Post where the search happens. TikTok now beats LinkedIn among Gen Z at 22.1% versus 20.8%, and 46% say they landed a job or internship through it.
Better yet, get sourced instead of applying. 91% of US companies use social in hiring, and 82% use it to find people who aren’t applying anywhere.
Those candidates are eight times likelier to get hired than job board applicants.
Meanwhile, everyone is rewriting their past: Nearly 20% of LinkedIn users edit old roles years later, usually bolting AI keywords onto jobs they’ve long left.
You don’t need to invent anything. But if you ran AI-assisted creative or automated your reporting, say so, because recruiters are searching for exactly that.
Treat the job hunt like a campaign, and the creative like it’s going to a client.
Retail marketers: this BFCM, don’t waste budget on one-off Promo Hunters.
This BFCM, don’t chase Promo Hunters.
Elevar by Audiense connects your site to ad platforms with session enrichment, server-side event delivery, and profit-aware signals so you target the shoppers who were already going to buy — and keep buying.
Elevar is the go-to solution for retail brands as it preserves browsing and cart history, surfaces predictive and calculated value events to Meta and Google, and separates one-time discount buyers from real subscribers and repeat purchasers at checkout.
The result: ad spend that optimizes for long-term profit, fewer wasted returns, and campaigns that actually improve customer lifetime value.
Their exclusive 2026 BFCM Playbook reveals six behavioral buyer patterns and the tracking blind spots that make last-click attribution lie.
Read now and see how Elevar maps signals to smarter BFCM audiences.
Don’t treat your Google AI search report like a scoreboard
Google finally split generative AI visibility out of the rest of Search Console.
But there’s still debate over how to read it. And should you even read it anyway. Well, Carolyn Shelby broke down what the new report tells and what it won’t.
- What you get: Impressions from AI Overviews and AI Mode, plus the URLs that appeared, segmented by page, country, device, and date. It’s rolling out to a subset of site owners first.
- What you don’t get: Queries. Clicks. CTR. Average position. Citation placement. Conversion data. So yes, all impressions, no answers.
Some key points to keep in mind knowing this…
An AI impression isn’t a search impression: In classic search, an impression means a listing was shown for consideration.
In AI Overviews, your link might be prominent, buried among citations, or hidden until someone expands. Links must be scrolled or expanded into view to count.
In AI Mode, follow-up questions count as new queries, generating fresh impressions.
The lesson: Don’t blend generative and organic impressions into one “total visibility” number, and definitely don’t calculate a blended CTR. You can produce the number. It won’t mean anything.
The math isn’t broken, but the dimensions are: Chart data aggregates at property level, so two URLs from your site in one response counts as one impression.
In the page table, each URL can get its own. Page-level totals won’t add up to property-level. That’s aggregation, not a bug.
Where you can get real value: Compare AI visibility against organic performance. Pages that rank well but rarely get retrieved may bury the answer in tabs, images, or JavaScript.
Pages with modest rankings but heavy AI impressions usually have a clean definition, stat, or table near the top of a section. Study those first.
The lesson: One successful URL is a clue, not a playbook. Change one heading, watch the graph rise three days later, and you haven’t reverse-engineered AI Mode, you’ve found a coincidence with good timing.
There’s also an opt-out toggle for AI features. Establish a baseline before anyone lunges for it.
When everyone’s using the same marketing playbook, psychology gives you an edge
Most marketers pull ideas from the same formulas, frameworks, and templates.
There’s no harm in using what works—but it also means your offer risks blending in.
But marketers who understand how people actually make decisions know how to package powerful offers and stand out—even when they’re selling similar things as their competitors.
Want to receive one psychological principle each week so you sharpen your instincts for positioning and persuasion?
Branding might be reclaiming the crown as performance hype fades
Looks like short-term performance chasing finally ran out of juice. Or money. Probably both.
European CMOs have spoken, and long-term brand building is officially back at the top of the board.
- Branding takes #1: McKinsey’s survey places Branding at the very top of CMO priorities, with Authenticity closely following at #4.
- Hype takes a back seat: Trendy levers like Gen AI (#17) and Social media and influencers (#20) sank to the bottom, showing a shift away from quick-fix tactics.
The Crew’s Take: Don’t take this at face value, but this could signal that performance marketing grew overcrowded and expensive.
CMOs may realize performance captures existing demand, but brand equity creates it. When acquisition costs spike, brand resonance is your ultimate margin armor.
Our takeaway: Rebalance your budget.
Invest in distinct brand assets, authentic messaging, and top-of-funnel positioning that build mental availability long before a prospect reaches your sales funnel.
AI MARKETING: This daily newsletter condenses the latest and greatest AI developments into a 5-minute read. They read the noise, pull out the signal, and explain what actually matters for work, strategy, and the tools your team is already using. Trusted by 700k+ readers. Subscribe right here.*
META: Australia’s under-16 ban now has hard numbers with 750,000+ accounts removed since December 2025. AI reads bios and captions for age clues, plus a re-registration crackdown. The platform wants age verified once at the OS or app store level. A model other regulators will likely copy.
META: Ads for illegal offshore casinos, some spoofing Crown and Star, are running across Facebook and Instagram in Australia, and the regulator can’t fine anyone thanks to a jurisdiction loophole. Reform is before parliament so it’s only a matter of time until the feed gets a cleanup.
BLUESKY: Videos up to 10 minutes are now in beta testing, alongside new Protocol Services meant to simplify building on AT Protocol. Active users have reportedly been shrinking, so the question here is if content brings them back…
X: On the other side of the web, text overlays just landed in the iOS video editor for X, with font and positioning controls built in. One less reason to bounce out to a third-party app before posting and one more nudge toward native uploads.
SOCIAL MEDIA: Harvard researchers analyzed 2 billion Reddit comments and found threads sour as they grow, not because of algorithms, but because the positive takes get used up first and standing out means going negative.
*This is a sponsored post.
A man rode out of town on Sunday, stayed a whole night at a hotel, and rode back to town the next day on Sunday. How is this possible?
You can find the answer here.
Interested in advertising in one of our newsletters?
Connect with over 100,000 of the world’s best marketers who read a newsletter by Stacked Marketer.
