Google is losing ground to Meta, facing legal action, and breaking its own ads

Google’s been sitting at the head of the table for too long.

Time to play musical chairs: Meta is on track to claim more global ad spend than Google in 2026.

In numbers, that’s 26.8% versus 26.4%, driven by AI automation that’s faster to deploy, easier to measure, and simpler to justify to a CFO.

If you have a Google-heavy media mix, take time to pressure-test whether that budget is actually pulling its weight.

And news doesn’t get better for Google: Advertisers are bundling legal claims against Google to claw back money they say an illegal monopoly cost them.

Pooling 25 or more claims shifts the leverage significantly. It’s a legal workaround that turns a process designed to favour Google into one that actually pressures it.

Resolution is likely 12 to 24 months out. If you’ve spent consistently on Google Ads, it’s worth finding out now whether you have grounds to join a future claim.

Google is also fighting itself: Automated crawlers pulled ads this week due to server errors that had nothing to do with advertisers’ own sites.

The crawlers don’t wait for a human to confirm the problem is real. If they hit an error, your ads go down regardless of whether your site is actually broken.

Check your Policy Manager now and appeal to anything that looks off.

Phew, we hope you had a better start of the week than Google.

You can now post faster on TikTok, scale Threads without the admin, and reach Reddit’s untapped event audience

TikTok’s own AI has doomscrolled enough to understand what makes a good visual.

Less fixing, more posting: TikTok’s Symphony Creative Studio now runs on Seedance 2.0, ByteDance’s latest AI video model.

This keeps product visuals consistent across segments and motion smoother.

Why we care: Fewer revision cycles means you’ll have more time to scale campaigns and burn less budget on back-and-forth. This should lower the cost of producing TikTok content at scale.

Threads grew a nervous system overnight: The Threads API now lets you schedule ghost posts, approve replies, and embed public posts on external sites without App Review.

It also supports GIFs and real-time webhooks. The new oEmbed API makes pulling Threads content into external sites faster and friction-free.

If Threads is on your radar, this update removes most of the manual overhead that made it awkward to manage.

Reddit wants to remind you something big is coming: Reddit’s Reminder Ads—which let you send timed push notifications to opted-in users—are now available globally.

One tap on the Remind Me button will nudge 24 hours out and then once again at launch.

36% of US Reddit users aren’t on Instagram, and Premier Boxing Champions saw a 3x higher CTR versus Meta for event-based campaigns.

That’s a measurable intent signal from an audience most campaigns miss entirely. Nice.

Why saying “AI-designed” can hurt your sales

Imagine you find out that a major competitor used AI to develop their new product. You try it out yourself and see that it’s pretty good.

So, you slap “AI-created” on the label, hoping for the same result. But it flops. Why?

Well, Science Says discovered the uncomfortable truth behind the answer.

Apparently, researchers ran 9 experiments across physical products, such as perfumes and snacks, and the findings are pretty damning for anyone leaning too hard into the AI branding.

• When a product was labeled “AI-designed” versus human-designed, purchase intention dropped by up to around 29%.
• Shoppers also assumed 75% less human involvement in the process and that perception alone was enough to kill desire. Humans are fans of other humans.

We buy products that feel like an extension of ourselves. A biker doesn’t just buy a Harley; they buy into a community, an identity, a designer they can relate to.

When AI is the sole creator, that human connection disappears. There’s no one to identify with. The product becomes anonymous, which can feel like a lack of care went into it.

How to fix it: Human-AI collaboration framing actually outperforms fully human-designed products. Purchase intent runs 3.5% higher than human-only and 13% higher than AI-only.

The framing: “Our designer Alex used AI to develop this feature.” That’s it. A human name and a human story. AI as the tool, not the author.

One exception worth noting: Rental products flip this dynamic.

People were 14% more willing to rent AI-designed camping gear than human-designed alternatives. When utility beats identity, people won’t care about the origin story.

The bottom line: AI in your process is fine. It could even be a selling point, if you frame it correctly. What can kill your conversions is erasing yourself from the story entirely.

Lead with your people and let AI stay behind the scenes.

How to win the AI summary war?

As AI becomes the new “search,” buyers are using it to do the heavy lifting.

To show up in those results, you need to provide what the robots are looking for.

B2B buyers are leveraging AI to accelerate their decision-making in several key ways:

• Summarize and compare options (61%): They want the “CliffNotes” version of the market.
• Analyze proposals or pricing (56%): AI is the new budget auditor.
• Overview potential vendors (50%): A quick way to build a shortlist.

Only a tiny 6% say they don’t use AI for buying at all, making AI-readiness a mandatory requirement for modern content.

AI is empowering the buyer. If your data is buried in unsearchable PDFs or vague marketing speak, you’ll be invisible to the bots summarizing the market.

What can you do: Don’t try to game AI crawlers. Think about how you can be clear.

If your positioning relies on vague language like “end-to-end solutions” or “best-in-class performance,” an AI won’t know how to represent you and neither will your buyers.

Audit your homepage and product pages with one question: could an AI accurately summarize what you do, who you serve, and why you’re different in three sentences?

If not, that’s your rewrite brief.

Also, if AI can’t summarize what you do, neither can your buyers.

Pro subscribers get breakdowns of how LLMs are reshaping B2B research, plus tips and frameworks to show up in AI-generated shortlists. $349/year for up to 10 teammates.