Bing 1B MAU, AI Max Adds Shopping/Travel, GPT-5 x3

Estimated reading time: 19 minutes

This week brought three pieces of platform news that any agency advising on search needs to absorb. Microsoft confirmed Bing has reached 1 billion monthly active users, the first time the search engine has hit that milestone. Google’s earnings showed search revenue up 19% YoY with the company explicitly attributing the growth to AI experiences, while AI Max expanded beyond search into Shopping and Travel campaigns with a new natural-language interface called AI Brief. And Botify data shows OpenAI’s crawl activity has roughly tripled since GPT-5 launched, with OAI-SearchBot now generating more events than GPTBot. Underneath the headlines, the week also produced fresh data on AI citations, brand visibility, and a properly funny experiment where a researcher convinced AI search to recommend a brand that doesn’t exist.

Bing Hits 1 Billion Monthly Active Users

On Microsoft’s Q3 FY2026 earnings call, Satya Nadella confirmed something that would’ve sounded fanciful three years ago: Bing has 1 billion monthly active users. That’s a tenfold jump from the 100 million daily active users Microsoft reported in 2023 when it first wired AI into the product. Whether Copilot interactions count toward that number isn’t fully clear, but the headline figure is now in the public record.

Search ad revenue (excluding traffic acquisition costs) grew 12% year-on-year, the third consecutive quarter of double-digit growth. The trajectory tells its own story: 21% growth two quarters ago, then 16%, then 10%, now 12%. Microsoft has guided Q4 to “high single digits”, which suggests the bounce is cooling.

The market-share number is where it gets awkward. StatCounter still puts Bing at around 5% global search share as of March 2026. So you have a search engine where the user count looks like Google’s and the share number looks like, well, Bing. The most plausible reading: most of those 1 billion MAUs touch Bing through a default integration (Edge, Windows, Copilot) rather than choosing it actively. CFO Amy Hood credits “higher volume and revenue per search” plus 20 consecutive quarters of Edge market-share gains. For agencies, the practical question is whether Microsoft Ads warrants a bigger slice of client budgets given the audience reach. Probably yes, in B2B and finance verticals where Bing has always over-indexed; less obviously in pure consumer plays.

Google Q1: Search Revenue Up 19%, Ad Revenue Up 15.5%

Alphabet’s Q1 2026 numbers landed strong. Google Search revenue clocked $60.4 billion, up 19% year-on-year. Total Google ad revenue hit $77.3 billion, up 15.5%. Overall Alphabet revenue was up 22% to $109.9 billion. Sundar Pichai explicitly credited AI experiences for the lift in search usage.

What’s interesting in context: this lands the same week the EU is finalising whether to force Google to share search data with rivals (binding decision still expected by 27 July, per last week’s coverage). The data here strengthens both sides of that argument. Google is genuinely growing, not stagnating, but the growth has the regulatory shape of a company widening its lead. Worth watching how the EU case proceeds against this backdrop.

Google AI Max Expands to Shopping, Travel, and Adds an AI Brief Interface

A meaty AI Max update from Google Ads. The headline: AI Max is no longer search-only.

The new pieces:

AI Max for Shopping uses Merchant Center data to generate adaptive ads against long-tail and exploratory queries. Positioned as a one-click upgrade for existing Shopping campaigns.
AI Max for Travel consolidates the previously fragmented “Search Campaigns for Travel” formats into a single interface with unified reporting.
AI Brief is a Gemini-powered natural-language interface that lets advertisers steer AI: define messaging rules, specify queries to prioritise or avoid, and shape audience targeting. Crucially, it generates previews before campaigns launch.
Text disclaimers and URL automation are paired together so AI can pick the most relevant landing page per query while keeping required legal messaging intact.

Google’s framing is that AI Max is becoming “a foundational layer across Google Ads” rather than a campaign type. That’s a meaningful shift. For agency teams, the practical implications are real: regulated-industry clients now have a workable compliance route into AI Max via the disclaimer feature; Shopping and Travel teams should test AI Max against their current setups; and the AI Brief interface is the most direct natural-language steering tool Google has shipped to date. Worth getting hands-on early.

OpenAI Crawl Activity Tripled Since GPT-5 Launch

Botify analysed roughly 7 billion OpenAI-bot log events from its enterprise client dataset, spanning November 2024 through March 2026. Chris Long of Nectiv wrote up the findings as a guest post. The headline numbers are dramatic.

“OAI-SearchBot now generates more activity than GPTBot.”
– Chris Long, Nectiv

Specifically, OAI-SearchBot recorded about 3.5x more events after GPT-5’s August 2025 launch (around 2.2 billion additional events), while GPTBot showed about 2.9x more events (1.8 billion extras). Pre-GPT-5 the search-to-training ratio was 0.95; it’s now 1.14. Translation: ChatGPT is doing more live web fetching for user queries than it is bulk-training, and the gap is widening.

ChatGPT-User (the bot that fetches a single page when a user asks for it) actually declined 28% from December 2025 to March 2026, suggesting OpenAI is moving live-query work onto the more efficient OAI-SearchBot infrastructure.

The vertical breakdown is striking. Healthcare saw approximately 740% more OAI-SearchBot activity post-GPT-5. Media/Publishing came in at 702%. Marketplaces, software and retail clustered at 190-216%. Travel sat at just 30%. If you’ve got clients in healthcare or publishing, AI search optimisation isn’t a “next year” priority any more.

One sobering bit of context: even tripled, OpenAI crawl volume is roughly 4% of Google’s. In Botify’s most recent 30-day window, Googlebot logged 18.2 billion events versus 887 million combined OpenAI bot events. The story isn’t AI displacing Google. It’s AI growing fast inside a still-Google-dominated ecosystem.

Back Button Hijacking: 15 June Deadline, and an AdSense Setting That’ll Trip You Up

Last week’s new spam policy now has teeth. Site owners have until 15 June 2026 to remove back-button-hijacking functionality, after which Google will issue manual actions and algorithmic adjustments. Google has started sending email warnings via Search Console.

The unhappy twist: a default setting inside Google AdSense vignette ads can itself trigger the penalty. So sites running stock AdSense configurations could find themselves in violation through no deliberate choice. If you run AdSense for clients, this is the audit you want to do this week, not next month.

Bing Webmaster Tools Demos New AI Reporting Features

Krishna Madhavan from Microsoft demoed forthcoming AI reporting features for Bing Webmaster Tools. The list is genuinely interesting: citation share, grounding query intent, semantic topic labelling, and GEO-focused recommendations. If this lands as advertised, Bing will become the first major search platform to give site owners proper AI-visibility telemetry. No release date yet, but worth keeping an eye on. Of all the things Microsoft could ship to differentiate itself this year, AI-visibility reporting in Webmaster Tools is genuinely useful in a way most platform features aren’t.

Comparison of AI Citation Patterns Across Five Engines

A study of citation behaviour across five AI search engines finds that the specific sources each engine prefers vary widely, but they converge on one thing: brand citations carry disproportionate weight for AI visibility. Different engines pull from different parts of the web (some lean heavily on Reddit, others on news, others on Wikipedia), but consistent brand mentions across that ecosystem matter regardless of which engine you’re optimising for.

For agencies, this reinforces what last week’s “bland tax” piece argued: distinctiveness and brand presence beat volume of generic content. The new wrinkle is that the citation calculus is engine-specific, so single-engine visibility checks miss most of the picture.

AI Search Clicks Often Go to Local Domains: Similarweb Across 10 Markets

Similarweb’s analysis of 10 markets shows AI search traffic frequently routes to local-language domains rather than global ones. The distribution varies by sector, but the broad pattern holds: when an AI engine has a credible local source available, it’ll pick that over a translated international page. This is quietly important for any UK agency advising on international expansion. The “hub and spoke” model where one global domain serves everything translated is actively losing to genuinely localised properties.

ChatGPT Triggers Web Search 78.3% of the Time on Commercial Prompts

A small-but-pointed study tested 90 prompts across beauty, legaltech and IT to see when ChatGPT actually goes to the web for fresh sources. The finding: commercial prompts trigger a web search 78.3% of the time, while informational prompts trigger one only 3.1% of the time. Of the 20 prompts that produced fan-out queries, 18 were commercial. Of the 42 fan-out queries those generated, 39 were commercial too.

The strategic takeaway flips the conventional GEO playbook. If you want to be cited in ChatGPT, the content type that earns it isn’t long-form educational guides; it’s “best-of” lists, comparison pages, and evaluation-focused commercial content. Educational pieces without commercial framing rarely make it into the fan-out path.

That’s not the whole picture (this is one prompt set across three industries; results need replicating across other models and verticals), but it does explain why some clients with extensive informational content libraries see almost no ChatGPT visibility. The match between content type and prompt intent is doing more work than people realise.

AI Sees Your Brand as Math, Not Messaging

Three articles about how AI “sees” brands landed on Search Engine Land in the same week. Not a coincidence. The collective argument: AI evaluates brands as mathematical vectors in semantic space, not as human-readable messaging. Consistency across content strengthens that vector; drift dilutes it. And, importantly, retrieval (whether the AI even fetches your page) precedes ranking (what position it places you in).

The practical implication for content teams: incoherent positioning across pages doesn’t just confuse human readers, it actively weakens your numerical signal in the AI’s encoding. One on-message page among ten different-message ones gets averaged into noise. The way to fix this is what good brand strategy always demanded: tight, consistent positioning. The new urgency is that the cost of drift is now machine-detectable and instantaneous.

How AI Models “Understand” Your Brand

Companion piece to the “brand math” article. Argues AI doesn’t comprehend brands in any meaningful sense; it pattern-matches them. Success requires consistent positioning across three layers: training data, retrieval, and generation. Misrepresentation creeps in when these three diverge: training has one version of your brand, retrieval surfaces a different one, generation produces a third. Worth a read alongside the brand-math piece. They’re really one argument in two voices.

From Links to Brand Signals: The New SEO Authority Model

The third in this week’s brand-and-AI thematic cluster argues that authority in 2026 derives from a multi-signal credibility model: mentions, citations, brand strength, consistent entity presence. Backlinks alone (the long-standing PageRank descendant) no longer cut it. Digital PR and earned media now carry weight comparable to and sometimes ahead of traditional link-building. If your link strategy still measures DA score and total inbound count without tracking unlinked mentions or citation breadth, it’s measuring the wrong thing.

Can a Fake Brand Win in AI Search? An Experiment Says Yes

A researcher spent a month seeing whether they could get a completely fictional brand cited by AI search systems through structured claims and content repetition. They could. The experiment doesn’t prove AI is uniquely gullible; it proves AI visibility follows predictable patterns rather than truth verification.

This sits awkwardly alongside last week’s “AI search is eating itself” piece. Both argue the same uncomfortable thing from opposite angles: AI systems treat consistent, structured information as authoritative regardless of whether it’s accurate. For brands, that’s a feature when you’re optimising your own visibility and a problem when bad actors do the same to misrepresent you.

Your AI Visibility Tracker Is Quietly Breaking Your Analytics

A pointed argument that the current generation of AI visibility tracking tools introduces measurement noise that obscures real visibility. The mechanism: trackers themselves trigger AI fetches, inflating what looks like organic AI activity. Combined with last week’s “ghost citation” piece, the picture is that AI visibility measurement is genuinely hard right now, and the tools selling certainty are arguably the problem. If you’re paying for an AI visibility tool and basing strategy on its dashboards, audit whether the measurement itself is sound before doubling down.

AI Gives You the Vocabulary. It Doesn’t Give You the Expertise

“The real divide isn’t human vs. AI, but retrieval vs. judgment, where long-term value is built through experience, not automation.”
– Search Engine Journal

Honestly, this is the rare AI think-piece that says something genuinely useful. The argument is that AI raises the floor on technical vocabulary (anyone can sound competent in a brief), but the ceiling on judgment (what to actually recommend, given the messy specifics of a client) is unchanged. For agency teams, this matters when you’re hiring or when you’re explaining why a senior strategist still costs more than a Claude subscription.

The Protection Paradox: Block AI Crawlers, Then Pay to Get Seen

A useful look at the contradiction many publishers have walked themselves into: aggressively blocking AI crawlers via robots.txt last year, then watching their visibility in AI answers crater, then quietly considering paid licensing deals or content partnerships to get back in. The lesson is that AI access policy needs to align with where your audience actually finds you. Hard “block everything” stances made sense in 2024 when nobody knew what AI training would do to traffic; in 2026 they’re often costing more in lost visibility than they gain in negotiating room with AI vendors.

APAC Search Strategy Goes Beyond Google and Baidu

APAC search has fragmented in ways most Western agencies still under-appreciate. Beyond the obvious Google-Baidu binary, regional engines (Naver, Yandex, Yahoo Japan) and bundled AI tools have created multi-system visibility requirements. The piece argues multi-system visibility is the new APAC benchmark, replacing the old “rank in Google plus Baidu and you’re done” rule. Particularly relevant for B2B SaaS clients with APAC expansion plans on the roadmap.

Google Tests “Enter AI Mode” Button and More Citations Inside AI Mode

Two related Google AI Mode tests this week. First, the call-to-action button taking users from AI Overviews into AI Mode is being tested as “Enter AI Mode” rather than the previous “see more” or similar phrasings. Second, Google appears to be showing more links and citations inside AI Mode results, mirroring what AI Overviews did roughly a year ago. Both tests are small but indicative: Google is steadily upgrading AI Mode to feel less like a sidebar feature and more like a proper search surface in its own right.

Can AI Mode Ads Drive Conversions, or Is It Just Awareness?

PPC practitioners are starting to ask the right question about AI Mode ads: do they actually drive purchase intent, or are they brand-awareness inventory dressed up as performance? The piece offers a measured framework for evaluating incremental conversion lift specifically attributable to AI Mode placement, distinct from generic search performance. The honest answer in 2026 is “we don’t quite know yet”. Treat early AI Mode ad spend as awareness that may also convert, not as straight performance, and budget accordingly.

Microsoft Ads PMax Website Publisher URL Report Adds Conversion and Spend Metrics

Microsoft Advertising has added conversion and spend data to its PMax Website Publisher URL report. That’s a meaningful transparency upgrade for a campaign type where it’s historically been hard to see exactly where budget went. If you run Microsoft PMax, this is a useful new audit lens.

The AI Skills Salary Premium for SEOs

A pragmatic piece on how AI skills now affect SEO hiring and salaries. The headline is that AI literacy isn’t a “specialism” any more; it’s embedded in most senior SEO roles. Job postings with AI skill requirements pay measurably more than equivalent roles without. For agency hiring, this means two things: junior pipelines need AI training built in from day one, and salary bands for senior roles probably need a meaningful refresh against the current market.

Strategic Direction: Where Search Marketing Is Heading

Three patterns crystallised this week. The first is platform consolidation. Bing’s 1 billion MAU, Google’s 19% search-revenue growth, and OpenAI’s tripled crawl activity are all signs of a market where the three meaningful AI-search players are getting stronger, not weaker. The “AI is eating Google” framing of 2024 isn’t matching 2026 data. AI Mode, ChatGPT, and Bing+Copilot are all growing inside an environment where Google still dominates by an order of magnitude on crawl volume and search share. Multi-platform visibility planning isn’t optional, but Google-first thinking is also not yet wrong.

The second pattern is the brand-as-vector argument hardening into a full SEO model. Three SEL pieces in one week, plus the citation-comparison study and the fake-brand experiment, all push the same point: distinctiveness, consistency and brand-signal density now drive AI visibility more than backlink counts or content volume. If your authority strategy still measures Domain Authority as the primary KPI and treats unlinked mentions as a soft signal, you’re optimising last decade’s algorithm. The shift is real and it’s mostly a content and PR shift, not a technical one.

The third pattern is enforcement closing in. Google’s back-button-hijacking deadline (15 June), the EU data-share decision (27 July), and the warning notices through Search Console all turn last quarter’s policy headlines into operational deadlines. The agencies that come out cleanly are the ones running compliance audits now rather than reacting after a manual action lands.

Key Takeaways

Bing has crossed 1 billion MAU. Even with 5% global market share, that’s a meaningful audience for B2B and finance budgets to consider testing.
Google AI Max is no longer search-only. Shopping, Travel, AI Brief and text disclaimers all shipped this week. Worth running parallel tests on existing Shopping clients now.
OpenAI crawl activity tripled since GPT-5 launched. Healthcare and publishing verticals saw 700%+ jumps. AI search optimisation is no longer a 2027 problem for those sectors.
Back-button-hijacking deadline is 15 June. Audit AdSense vignette settings now; default configurations can themselves trigger penalties.
Three independent SEL pieces this week argue AI evaluates brands as numerical vectors. Inconsistent positioning across pages directly weakens AI visibility.
Commercial prompts trigger ChatGPT web search 25x more often than informational ones (78.3% vs 3.1%). “Best-of” and comparison content earns AI citations; pure educational content rarely does.
Bing Webmaster Tools is set to ship AI-visibility reporting (citation share, GEO recommendations). First major platform to do so. Worth tracking for clients reliant on Bing traffic.

Frequently Asked Questions

Should I be reallocating client budget to Bing now that it’s hit 1 billion MAU?

Modestly, yes, especially for B2B, finance and software clients where Bing has historically over-indexed. But remember the audience-vs-share gap: most of those MAUs touch Bing through defaults rather than active choice, so behaviour can differ from Google. Run an A/B before committing to a permanent reallocation.

What’s the actual deadline I need to remember for the back-button hijacking penalty?

15 June 2026. After that date, Google starts issuing manual actions and algorithmic penalties for sites that haven’t removed back-button-hijacking behaviour. Audit AdSense vignette settings as part of that check; defaults can trigger the penalty without you doing anything wrong intentionally.

Do I need to test AI Max for Shopping immediately?

If you have ecommerce clients running Google Shopping at scale, yes. Run a parallel AI Max campaign alongside the existing setup and compare for 4-6 weeks. The Merchant Center integration is the meaningful piece here; it lets AI Max cover long-tail and exploratory queries that traditional Shopping campaigns miss.

If commercial prompts dominate ChatGPT citations, should I stop producing informational content?

No. Informational content still does work for traditional search and brand authority. But if AI citations specifically are a goal for a client, the content mix needs to skew toward “best-of”, comparison and evaluation pieces rather than purely educational guides. Treat ChatGPT visibility as a content-format question, not a content-quality one.

How do I actually check AI visibility now if the trackers are unreliable?

Triangulate. Use multiple AI-visibility tools rather than one, run manual brand-name queries across ChatGPT, Perplexity, Claude, and Gemini at fixed intervals, and watch your server logs for AI-crawler patterns. Single-tool dashboards are too noisy to trust right now. The Bing Webmaster Tools AI reporting (when it ships) will be the first piece of native platform telemetry, but until then it’s manual triangulation.

Conclusion

This week’s news made one thing clear: the AI-search story isn’t a single platform displacing Google. It’s a multi-engine ecosystem stabilising into something agencies can actually plan around. Bing’s user count, Google’s earnings strength, and OpenAI’s crawl growth all point the same direction. So does the brand-vector cluster of articles, which make a coherent case that AI visibility is largely a brand-strength problem dressed up as a technical one. Meanwhile, real enforcement deadlines (back-button hijacking, EU data-share) turn last quarter’s policy headlines into operational must-dos. Plenty to act on this week.

Need help adapting your search strategy for the AI era? Contact the Anicca team for expert SEO and PPC guidance.

Table of Contents

Platform & Earnings

SEO & Compliance

AI Search, GEO & AEO

PPC & Paid Media

Industry & Workforce

Wrap-Up