May saw Google Marketing Live 2026 take place, with many announcements off the back of it. Elsewhere, TikTok are following in the footsteps of Meta by offering a paid, ad-free option, and two of Switzerland’s finest watch brands stirred up a storm on the streets. And Google’s second core algorithm update of the year is here. Here’s the news affecting performance marketing in May.
SEO
1. Buckle up, the second core algorithm update of the year is rolling out
Google’s latest broad core update landed on the 21st of May, and it’s already causing a stir. By the weekend, the SEO community was reporting everything from 30% surges to ‘total collapse.’
This seems to be an update with teeth, but it’s important not to panic and make knee jerk changes to your site whilst it rolls out. It’s predicted that the rollout will take two weeks, but we know from experience that it could be longer until it finishes, and even longer still until the dust settles.
What to do when it ends
- Isolate the data – If you stay still, it’s much easier to see if a traffic drop is a direct result of the update or just your own technical tinkering.
- Wait for the dust – Don’t start any kind of recovery plan until at least a week after the rollout is officially confirmed as complete.
If you’re lost in the data, send a signal to our SEO experts.
Source: Search Engine Roundtable
2. GA4 finally adds an ‘AI Assistant’ channel
Previously, traffic from the likes of ChatGPT, Gemini, and Claude was unceremoniously dumped into the ‘Referral’ bucket. Now, Google does the legwork for you, automatically tagging this traffic as the ‘ai-assistant’ medium.
Google is playing its cards close to its chest. They haven’t published the full referrer list yet – only name-dropping the big three (ChatGPT, Claude, and Gemini) as examples.
This plug-and-play version might not catch as much as the custom Regex setups we’ve been building to track the wider AI ecosystem. If you’ve already got a robust custom channel in place that tracks niche bots, don’t hit delete just yet. We’ll be keeping a close eye on the data to see if Google’s net is wide enough to catch the smaller players.
If you can’t see AI Assistant in your GA4 channels yet, be patient – it’s coming.
Source: Search Engine Journal
3. Schema isn’t the silver bullet to AI citations that LinkedIn gurus want it to be
Ahrefs just tracked 1.8K pages that added JSON-LD schema, and the results are a cold shower for anyone betting on it as an AI visibility lever. According to their study, adding schema produced zero major uplift in AI citations across any platform.
With the amount of noise out there on GEO hacks, this data is a much-needed reality check. Machines were built to read the mess of the web; you can’t just code your way to the top.
What does this mean for you?
- Don’t bin your schema – this isn’t a license to delete your structured data. Schema remains critical for communicating your site’s entities (who you are, what you sell, and your credentials) to search engines and voice assistants.
- Fix the foundations first – whacking schema into the mix won’t fix a site that’s slacking on content quality or link equity. AI models prioritise visible HTML content over hidden code. If your expertise, authority, and trust (E-E-A-T) aren’t visible to a human, they’re invisible to the bot.
- Test, don’t guess – never blindly trust a LinkedIn carousel. Ahrefs is a data powerhouse, but your site is its own ecosystem. Run your own controlled test: add schema to 10 pages, leave 10 without, and compare your AI citations after 30 days. If the lines don’t diverge, you have your answer.
Use schema as the ‘infrastructure’ it was meant to be, not a shortcut to a citation.
Source: Ahrefs
4. Google introduces new features to AI results
Google has just announced five upgrades to how it displays links in its generative AI search features (AI Mode and AI Overviews). After months of being accused of killing clicks by summarising everything into a neat little box, Google is trying to prove it still cares about sending traffic to actual websites.
New features include
- Explore More Links – Google is sticking suggested deep-dives at the end of AI answers to keep you clicking.
- Subscription Badges – If you pay for a news site, Google will flag their links as subscribed within the AI summary.
- Human Quotes – more previews from forums and social media (with names and handles) to prove there’s a real person behind the advice.
- Inline Citations – links are now being peppered directly next to the relevant sentences, not just hidden at the bottom.
- Hover Previews – on desktop, you can now peek at a site’s title and URL by hovering over a link before you commit to the click.
It seems like Google is doubling down on the core principles of E-E-A-T – now is not the time to slack on authority and brand building.
Source: Google – The Keyword
Digital PR
1. Synapse rolls out WhatsApp integration and restricts free accounts
Journalist requests and PR pitching platform Synapse has announced major changes to its user model, moving full platform access to premium, paid tiers from mid-May. Users remaining on free tiers are being transitioned to restricted plans, which will significantly limit their access to the core stories marketplace and live journalist request feeds.
Synapse has also launched a brand-new WhatsApp integration designed to deliver a live feed of media opportunities straight to a user’s phone. This feature works both ways; while PRs receive real-time journalist requests, journalists also receive a direct feed of PR pitches on their mobile devices.
The restriction of free accounts means the barrier to entry for free-tier media monitoring is rising, making budget allocation for premium PR tools more necessary than ever to avoid missing breaking media opportunities. This might be another win for agencies that prioritise access to paid tools that marketeers may not have in-house.
This WhatsApp integration is closing the gap between PRs and journalists, opening hand-to-hand communication. It makes it easier to respond to requests faster, but means competition will be pitching at double speed too, increasing the pressure on DPRs to be ever ready to pitch.
Source: Synapse
2. Swatch x Audemars Piguet collab causes an IRL ruckus that leads to online buzz
Swatch sparked nationwide news coverage following the release of its highly anticipated, limited-edition “Royal Pop” collaboration with luxury watchmaker Audemars Piguet. The launch drew huge crowds to brick-and-mortar stores, forcing nationwide store shutdowns due to safety concerns and overwhelming demand that far outstripped available stock.
It made national news, dominating both social media feeds and major publication headlines, and thousands of disappointed shoppers took to their own socials to complain about the scarcity and share the crowds in real time.
In a digital marketing landscape often dominated by algorithmic reach and online ads, this campaign is a reminder that there is no PR quite like manufactured exclusivity and scarcity. It proves that experiential, physical activations (what you might call ‘real life’) can still generate an enormous wave of organic, authoritative digital coverage.
Creating a genuine sense of FOMO forces top-tier news outlets to cover the retail launch as a cultural event rather than a standard product release. It’s the latest in a long list of examples to add to the “is all PR good PR?” debate. This mainstream press attention delivers earned media and high-authority backlink profiles, but is it great for all your recent coverage to be about how you messed up and left lots of customers disappointed? The debate is sure to continue.
Source: BBC
3. Vibe coding: time for tools to fix your DPR activity?
Coding agents like Claude Code and ChatGPT codex are now extremely competent at helping non-coders build sites, apps and tools. Tools have long been a favourite of DPR campaigns and of journalists, e.g. “enter your postcode to discover the crime rates in your area”, but they have usually required extensive web development expertise to build and maintain. This cost barrier meant they were only an opportunity for brands and outlets with big enough budgets – now a tool like Google AI Studio means the only thing you need is a Google account to build a site purely through natural language prompts.
What might have previously been a data-driven DPR campaign could also become a tool that lives on the client site, with minimal extra work required. If you can create something that is genuinely useful for the public, it’s potentially another way to gain links, even after campaign outreach has stopped. Recent examples of this include https://wherescool.uk/ from the DPR world, using public data to list everywhere in the UK that has air conditioning, or the game Bottleneck from a journalist, which puts players in the chair of a crisis maritime coordinator during the Strait of Hormuz closure.
It’s important to remember that just because you can, doesn’t mean you should. If your campaign isn’t landing with journalists, it’s unlikely to suddenly start working just because you vibe coded a tool for it. And once you’ve built a tool, you’ll likely need to dedicate time to keeping it working, so if there are bugs or issues that your AI assistant can’t work out how to fix, you could end up sinking a lot of time into something that was supposed to be a quick win.
The democratisation of coding could be a good thing for DPR, especially smaller agencies, but after the thrill of seeing an idea come to life on a webpage, there’s still the daily slog of maintaining an asset.
PPC
1. Ads in AI Mode
This month, Google introduced native, conversational ad formats directly inside its interactive AI search interface. New units like Conversational Discovery and Highlighted Answers use AI to dynamically tailor ad copy and write custom product explainers based on the ongoing user chat.
These placements bypass traditional keyword matching entirely, requiring brands to use fully automated AI Max or Performance Max campaigns. Because Gemini auto-generates ad copy directly from your assets and Merchant Center, flawless data hygiene and rich feed details are now critical for conversion.
Source: Google
2. Display merges into Demand Gen
Google is moving Google Display Network (GDN) placements directly into Demand Gen campaigns as part of a push toward unified, AI-driven setups. Advertisers can now manage Display inventory alongside YouTube, Discover, Gmail, and Maps in a single environment, though running ads exclusively on GDN remains an option for now.
Standalone Display campaign management is becoming less central, forcing teams to rethink how they separate upper-funnel discovery and performance-focused media buying. Integrating GDN into Demand Gen unlocks newer AI-powered features and optimisation, with Google reporting an average 9.5% increase in ROI for advertisers who make the shift.
Source: Search Engine Land
3. Direct Offers
Google has expanded Direct Offers, letting retail and travel advertisers deliver real-time promotions (discounts, giveaways, hotel rates) directly within AI Mode answers and Search results. The update introduces Gemini-powered deal bundling to combine multiple assets into targeted custom packages, alongside a native checkout integration for Universal Commerce Protocol (UCP) merchants. Advertisers using AI Max and Performance Max campaigns can now build these offers natively inside the Google Ads UI and enforce automated cost caps to protect margins.
These placements capture high-intent users exactly when they are making buying or booking decisions, giving brands a low-friction tool to incentivize instant conversions. The addition of native checkout turns conversational research into completed sales instantly, removing traditional web-funnel drop-offs. Because management relies on data from Merchant Center and Hotel Center, clear promotional asset organisation is essential for Google’s AI to build optimal bundles.
Source: Google
4. Business Agent for Leads
Google has launched Business Agent for Leads, an AI-driven ad format that embeds a conversational Gemini agent directly inside Search and AI Mode ads. Instead of filling out static forms, users click a “Chat” button to ask specific questions, which the agent instantly answers by reading and crawling the advertiser’s website. Once the user’s questions are answered, Google pre-fills a native lead form with their verified contact details so they can submit their inquiry without ever leaving the Google interface.
This shifts the lead qualification process ahead of the form submission, changing the entry point from low-intent casual clicks to deeply qualified prospects who have already interacted with your data. Because Gemini answers questions dynamically without a fixed script, thin or shallow website content will instantly result in poor conversations and lost leads, making detailed, on-brand site copy highly critical. Leads from this format arrive in your CRM packed with full conversational history, requiring agencies and sales teams to adapt their outreach workflows to leverage that captured context.
Source: Google
5. AI-powered Shopping ads
Google also announced AI-powered Shopping ads this month, a new ad format that uses Gemini to surface the most relevant products dynamically within conversational search results. The format replaces static product listings by automatically translating technical product specifications into simplified, human-centric feature summaries. These ads use AI to generate real-time, custom text explainers that justify to the user exactly “why” a recommended product matches their specific needs.
This feature shifts e-commerce advertising from a numbers game of keyword match and product imagery to a conversational framework built on consumer education and transparency. Gemini dynamically extracts and translates technical product features to pitch the consumer, so your underlying Merchant Center feed data must be accurate, rich, and highly detailed. Providing immediate, tailored product justifications speeds up the research phase for complex items, helping to drive higher conversion rates and fewer abandoned carts.
Source: Google
6. Improved click-to-call ads
Last but by no means least, Google has updated its Click-to-Call Ads, utilising Gemini AI to automatically transcribe and analyse the actual content of phone calls generated by the ads.
The system moves past tracking simple call metrics, like call duration or button clicks, to identify the real intent, topics discussed, and specific consumer queries spoken during the conversation. This technology instantly flags high-value customer inquiries, filters out unqualified interactions, and automatically feeds conversation-driven performance data back into Google Ads.
This eliminates call length as a flawed proxy for lead quality, shifting optimisation toward actual conversion signals like booked appointments or verbal quote requests. Because Gemini uses automated call intelligence to score leads, lead-generation campaigns can dynamically adjust bidding models to target prospects who demonstrate explicit purchasing signals over the phone. Put simply: you’re getting even more information.
Source: Google
Paid Social
1. Meta expands advanced ad placement controls to Threads
Meta has expanded third-party block list support to Threads ad placements, bringing it in line with controls already available across Facebook and Instagram. These block lists (independently verified by brand safety partners DoubleVerify, IAS, Scope3, and Zefr) ensure ads aren’t shown alongside content that’s been flagged as unsafe, offensive, or brand-inappropriate – without the manual effort of excluding individual profiles or categories.
Meta claims over 99% of content next to ads across its platforms is brand safe – but treat that number with caution. Every major platform reports the same figure, and third-party verifiers don’t have direct access to Meta’s data, meaning Meta has some influence over what gets measured and reported.
Meta is also bringing two more verification partners (Channel Factory and Protected by Mediaocean) to Threads in the coming months, giving advertisers additional options for how they manage and verify brand suitability. All this means Threads is becoming a more viable paid channel. Brand safety controls are a strong signal that Meta is actively maturing Threads as an advertising platform.
Like we said, don’t take the 99% brand safety claim at face value. Platform-reported safety metrics are not fully independent. Marketers should consider using third-party block lists and verification tools actively rather than relying on Meta’s headline figures – especially for sensitive or premium brands.
It also means Threads now has stronger brand safety infrastructure than X, which makes it a more credible option for brands that want a text-based, conversation-driven ad environment. Marketers who have been sitting on unallocated budget since deprioritising X should be seriously considering whether Threads is now the right home for it.
Source: Social Media Today
2. TikTok launches an ad-free subscription plan in the UK
TikTok has launched an ad-free subscription tier in the UK, priced at £3.99 per month, available to users aged 18 and over. Subscribers won’t see ads, and their data won’t be used for advertising purposes. The move was likely prompted by UK GDPR rules, which require platforms to obtain user consent before collecting personal data for advertising – making a paid, ad-free option a way of meeting that legal obligation.
TikTok first began testing the ad-free plan back in 2023, and it’s currently unclear whether the subscription will be extended to the US market.
Any user who subscribes to the ad-free tier is completely removed from targetable audiences – and their data can’t be used for advertising at all. If the subscription gains traction (particularly among older, higher-income users who can afford it), paid social campaigns on TikTok could see reduced reach and a shift in the audience profile that remains.
It’s the continuation of a broader industry trend – Meta, YouTube, and now TikTok have all introduced ad-free paid tiers. For marketers, this signals a long-term structural change, with the most engaged, privacy-conscious users increasingly opting out of the ad ecosystem entirely.
Sponsored content from creators sits outside the traditional ad system and isn’t blocked by ad-free subscriptions. As paid placements reach fewer people, partnerships with TikTok creators offer a way to maintain visibility with audiences who are actively choosing to avoid ads.
Source: TechCrunch
3. UMG and TikTok renew agreement to combat unauthorised AI music
TikTok and Universal Music Group have renewed their global licensing agreement, with a shared commitment to remove unauthorised AI-generated music from the platform and improve how artists and songwriters are credited. This follows a very public falling out in 2024, when UMG temporarily pulled its entire music catalogue from TikTok over concerns about AI-generated content and copyright. The dispute was part of a broader industry anxiety around AI tools capable of mimicking artists’ voices or generating counterfeit songs, highlighted by viral AI tracks imitating artists like Drake and The Weeknd that racked up millions of streams before being removed.
The deal is being seen as a potential template for how the wider tech industry should handle the intersection of AI, intellectual property, and platform accountability, particularly as regulators in the EU and US states tighten their approach to AI-generated content.
With UMG back on solid footing with TikTok (and the platform actively cracking down on unauthorised AI music) there’s less risk of licensed tracks being pulled mid-campaign. For brands using UMG-licensed music in TikTok ads or branded content, this agreement provides greater stability. The platform is now formally committed to removing unauthorised AI-generated tracks. Brands or agencies experimenting with AI music tools for ad creative should ensure any music used is properly licensed – using unlicensed AI-generated audio could result in content being removed and campaigns rejected.
As this deal sets a precedent and regulators increase pressure, expect other platforms to follow with similar frameworks. Marketers should start reviewing their broader music licensing practices now, particularly where AI-generated or AI-assisted audio is involved in creative production,
Source: TechCrunch
4. Meta launches new group-focused Forum app
Meta has launched a new standalone app called Forum, which pulls all of a user’s Facebook Groups into a separate Reddit-style experience, with a focus on asking questions and getting answers from real people within those communities.
The real driver behind Forum is AI data, not user experience. Reddit has become one of the most cited sources for AI chatbot answers because it provides real, human responses that are vetted through upvotes and downvotes, giving AI tools reliable reference points. Meta is trying to replicate this with Forum, encouraging more question-and-answer activity that can be used to improve its own AI models.
Most people already access their groups through Facebook and have established habits there. Meta may need to push activity more deliberately toward the app, or place greater emphasis on the question-and-answer features, in order to make it stick.
Forum doesn’t carry ads yet, but Meta has a track record of monetising new surfaces once they reach scale (see: Threads). If Forum gains traction, it could become a new placement option, potentially offering highly contextual targeting based on group topics and interests. Meta is directly going after what makes Reddit valuable to both users and AI. Marketers currently investing in Reddit ads should monitor whether Forum pulls audience attention away, and factor that into how they think about community-led advertising more broadly.
Source: Social Media Today
When the channels you rely on are changing so much and so often, you need people behind you who not only know what’s going on, but know how to get advantage out of it for your brand. If you don’t have that, get in touch.