William Marks

ChatGPT apps just launched in the EU and LinkedIn is drowning in hot takes. Everyone’s suddenly an AI distribution expert. Every CEO is asking their team the same question: where’s ours? Wrong question. Or at least, incomplete. Because this isn’t just a new feature - it’s ChatGPT shifting from a chat product to a distribution platform. The real question is where this conversation with your customers should happen. Build inside the ChatGPT ecosystem and you get access to over 200M weekly active users. More importantly, your app surfaces at the moment of intent, when someone’s already in a conversation looking for a solution. The distribution is built-in. But there’s something bigger: if your app becomes the default answer for a specific task or category, you own that intent. That’s distribution no brand could buy five years ago. The challenge: you’re one app amongst thousands in a store you don’t control. Or you could build your own instead. Embed conversational AI into your website, app, or customer channels. The places where you already own the traffic and the relationship. You’re not competing for discovery. You’re enhancing the experience for people who already chose you. There’s another benefit worth mentioning. The structured knowledge you build for your conversational AI doubles as fuel for Generative Engine Optimisation (GEO). By organising your brand’s intelligence for your own agents, you ensure you’re the one citing the facts when public AI engines like Google and SearchGPT synthesise answers for your customers. Some brands will benefit from being where users are actively searching. Others from deepening engagement on their own channels. Most will eventually need both. A ChatGPT app for discovery, custom AI for owned touchpoints. It depends on where your customers are, where you have traffic today, and where you’re building for tomorrow. Whichever path you take, the constraint is similar. These platforms control the visual framework. Your brand gets a conversational thread and the ability to call your APIs. No colour palette. No navigation. Just conversation. Your differentiation lives in how well you know your customers and your domain. Responses that feel specific rather than templated. Decisions based on actual understanding, not just the product specs. Your domain expertise, customer insights, product catalog. This becomes your brand voice when the visual layer disappears. At Cyphr, we’ve spent years building conversational experiences for brands. Not just the technical build, but helping answer the strategic question first: where should this conversation actually happen? Exploring what makes sense for your brand? ChatGPT app, custom build, or both? Let’s talk it through. Happy to share what we’re seeing and help you figure out the right path for your specific situation. #ChatGPTApps #ConversationalAI #BrandStrategy #AIProduct #SEO #GEO #DigitalStrategy​​​​​​​​​​​​​​​​

18

Really excited to announce this one: Leaf Signal's 2026 UK State of Tracking report is now live 💥 I've lost count of how many times over the last 8 years a brand's told me their tracking is "fine" and "we have it in hand". Basically brushing it under the carpet. I don't blame them it's not sexy and it's hardly ever got a designated stakeholder in or around the business. So we decided to check using Watcher (our agentic monitoring auditing tool) the conversion tracking across 500 of the UK's biggest Shopify stores. What did you find Mark I hear you heckle? Well I'll tell you... 🥁🥁 🥁 🥁 🥁 𝟵𝟴% had broken tracking on Meta, Google Ads and/or GA4... 🤯🤯🤯🤯🤯 If you're brand side, or a performance marketing / CRM / SEO / Design & Dev agency - this is definitely worth a read. Full findings and report 👇👇👇👇👇 https://lnkd.in/ewDQqPKt  Massive shout out to our wonderful head of marketing Charlie Semmence for producing this report with her team, and to all our partners. #performancemarketing #dtcbrands #performancemarketingagency #SEOagency #ecommerce #tracking #GA4

20

1 Kommentar

Three CMOs in the last month have told me some version of the same thing: nobody can tell them who owns what AI says about their company. The search, brand, and PR teams all lay claim to it, while at the same time pointing fingers at somebody else for the hard, tricky parts like attribution analysis and journey strategy. One CMO said it seems like every team wants the credit when it works, and every team wants to point fingers when there's a question. This was survivable when search was ten blue links. It isn't now, because the AI search team is doing something the old SEO team never did. They aren't chasing rankings. They're teaching a model what your company is. Here's the pattern I keep running into. The search team publishes a claim tuned for intent. "Best." "Used by 99% of the Fortune 500." A competitor comparison that's technically true and aggressive. Nobody is lying. Search has always rewarded claims that match what people type, not claims that match the company narrative. PR would have added a qualifier. Brand would have said that's not how we describe ourselves. Neither was in the room. Then AI picks it up, and the claim stops being yours. A study of more than 25 million links found 84% of AI citations come from earned and third-party sources, not your own site. Across 75,000 brands, off-site brand mentions correlated with AI Overview visibility about three times more strongly than backlinks (0.664 vs 0.218). So the thing your AI search work moves is the thing PR and brand were hired to protect. Same asset. Different room. And the models are careless with it. In the BBC and European Broadcasting Union study of 3,000 AI answers, 45% contained at least one significant issue and 31% had serious sourcing problems, including missing or misleading attribution. Now an analyst preps for your earnings call with AI and gets your search team's framing back, attributed to somebody else. The CEO texts the CMO. The CMO calls comms. Comms calls SEO. You can't delete your way out. You can pull the pages you own. The citations you don't own are already in training data. The fix is one room, and clear ownership inside it. Brand owns the claim set, what we are allowed to say about ourselves. PR owns how those claims travel and who repeats them. SEO and AI search own where they show up and what the model learns. All three work from the same buyer research, because the buyer doesn't experience your org chart. They ask a model a question and get one answer. Run AI search as a standalone SEO motion and you're funding a brand problem out of a keyword budget. At Camino5 we run a claims audit as part of AI search work: every claim your content is making, mapped against what the models are already saying about you. DM me "AI search" and I'll show you what we look at.

2

2 Kommentare

Car News or Fake AI click Bait While there is no evidence of car manufacturers explicitly hiring "AI marketers" for the sole purpose of creating clickbait, the industry is embroiled in a complex relationship with AI-driven content and programmatic advertising that often leads to this outcome. 1. Indirect Funding via Programmatic Advertising Major car brands inadvertently fund AI-generated clickbait through programmatic advertising. MIT Technology Review The "MFA" Problem: Over 140 major brands, including automotive companies, have been found to have ads appearing on "Made For Advertising" (MFA) sites. These sites use AI to churn out low-quality, high-volume content designed only to trigger ad impressions. Lack of Oversight: Roughly 90% of ads on these junk sites are served through platforms like Google Ads, meaning car manufacturers often pay for these "clicks" without knowing their ads are appearing next to AI-generated spam. ACM: Association for Computing Machinery +1 2. AI-Generated Clickbait and Misinformation The automotive sector has become a primary target for AI-generated clickbait that spreads false information: Fake YouTube News: Channels like "Motorcycle Central" have used AI to create videos claiming major CEOs are "saying goodbye to dealerships" to move to direct-to-consumer sales. Brand Refutation: These AI campaigns are so prevalent that manufacturers like Indian Motorcycles have had to issue official statements to debunk misinformation generated by these AI channels. Facebook 3. Deliberate AI Use in Legitimate Marketing Manufacturers are increasingly using AI for data-driven "performance marketing," which focuses on high click-through rates (CTR) but aims for higher quality than traditional clickbait: Affiliate Models: Platforms like CarExpert use AI and data to replace traditional "linkbombing" with targeted prompts (e.g., booking a test drive), achieving 10% CTRs compared to industry averages. AI Assistants: Companies like Impel provide "agentic AI" solutions that manage everything from the "first marketing click" to final service, automating interactions to maximize engagement. In-Car Advertising: Research into AI-driven in-car ads suggests future systems could use driver data to serve highly targeted—and potentially intrusive—advertisements. Mi-3.com.au 4. Recent AI Marketing Mishaps The "$1 Chevy" Incident: A Chevrolet dealership's AI chatbot was famously tricked by a user into "selling" a $76,000 truck for $1. This highlighted the risks of deploying AI-driven sales and marketing tools without sufficient guardrails. AI in Listings: Users on platforms like Carsales have noted a surge in "copy-and-paste" AI-written vehicle descriptions, which can feel like clickbait due to their generic, overly-enthusiastic tone. Reddit +4

2

2 Kommentare

Despite the narrative, the data shows that Berlin is clearly Germany’s number one startup hub. The recent report from Startup-Verband is used to suggest that Munich is the #1 German startup hotspot, based on startups per capita. No one would call Cambridge the UK’s number one startup hub just because it beats London per capita. Scale, momentum, talent, capital, and global relevance still matter. Last year, Berlin created more than twice as many startups as Munich and showed the strongest year-over-year recovery among Germany’s major ecosystems. We should celebrate Munich’s momentum and we should invest more across Europe and strengthen collaboration. But we should also be precise about what the data actually says. A strong innovation hub in Berlin that pulls in talent and capital is critically important for a sovereign European. And fuelling a false narrative is damaging. Berlin is not just number one in Germany. There is such great energy and momentum building up: I believe it is on its way to becoming Europe’s leading startup hub - and that is good for Germany, Europe and ultimately the world. The Delta Campus UNITE #berlinaufdie1

208

41 Kommentare

Had a tough conversation this week where I had to deliver some hard truths about a product launch. The metrics looked okay on the surface. But when we dug deeper, the conversion story wasn't adding up. It would have been easy to spin it positively. Talk about the engagement. Focus on the potential. Instead, we talked about the real challenges. Being the person who points out these issues isn't fun. But here's what I've learned: honest assessment, even when disappointing, builds credibility. Clients don't need cheerleaders. They need someone who will help them see what's really happening so they can make informed decisions about where to invest their time and money. Sometimes the most valuable thing you can do is help someone pivot before they've burned through their runway. #ProductStrategy #HonestFeedback #StartupReality

8

3 Kommentare

After the Bitpanda fine hit the news, a CCO messaged me: "We have the same gap. Our marketing team publishes without checking with compliance. How do I fix this without slowing them down?" This is the real question. Not "do we need compliance checks?" — obviously yes. But: how do you add them without becoming the department that blocks everything? The answer isn't a new approval process. It's an automated pre-check embedded in the publishing workflow. Before any marketing goes live: is there a published white paper for this asset? Has the 20-day notification window passed? Are required disclaimers present in the copy? Are contact details included? Four checks. Automated. If all pass — publish without waiting for compliance sign-off. If any fail — route to compliance with the specific gap flagged. Marketing stays fast. Compliance stays informed. The audit trail exists regardless. The fix isn't more process. It's a smarter system. What compliance check at your firm still depends on someone remembering to do it? #MiCA #CASP #Compliance #Marketing #Operations

1

1 Kommentar

Somewhere along the way, brands and agencies lost the art of strategy.   We traded an ironclad plan for: - Putting out fires - Constant reactiveness - Chaotic ops - Duct-taped process   I’m not anti-speed. Bias for action is one of our values.   But action without strategy is like flooring it with no map. Lots of motion. Very little progress.   Beyond retainers, we do one-month audits for brands.   The output: a crystal-clear 3–6 month roadmap. - What to do - What not to do - And in what order   No fluff. No “everything matters.”   They’re not cheap--$12.5K–$15K. I’ve tried hard to productize them. To lower the price, go for scale.   But here’s what I’ve learned: You can’t shortcut a world-class plan.   The kind of plan that helps a team do less. So they can do better work. The work of their life.   AI helps. Diagnostics help. Process helps. But the hard part is still by hand. And there’s always a struggle to make it great.   But great is the bar. And great things take time.

15

9 Kommentare

Good morning LI Community. Happy Monday 😀 Here is some very interesting research about AI. Thanks for sharing Alexander Knoll I recall reading that use of AI synthetic data often produces results around the mid-point, and as mentioned below, make people sound similar/more alike than they really are. What’s your take on this? Study links are in the comments. #mrx #marketresearch

3

1 Kommentar