State of Brand

According to Boston Consulting Group (BCG), 72% of CEOs now say they’re the main decision makers on AI at their companies, double the share from last year. Agencies have picked up on it. On Publicis Groupe's Q3 call last week, Arthur Sadoun said CEOs are getting more involved in the work Publicis does. The company has won Microsoft, PepsiCo and other big accounts this year without a competitive review, and PepsiCo’s CEO talked up the agency on his own earnings call. Normally the CMO would be the one doing that. We think B2B marketing leaders should pay close attention. Sales leaders expect bigger budgets and more headcount in 2027, while marketing budgets have been flat at 7.8% of revenue. Gartner found only 32% of executives think their CMO makes strong strategy recommendations based on data. And while almost half of marketing teams have handed tasks to AI, only 6.5% track whether it’s paying off. There’s still a lot marketing can own. Buyers want business cases and ROI evidence earlier in the process now, and most sales leaders say AI isn’t helping them produce it. Marketing already does a lot of this work, so it’s well placed to lead here. If CMOs want a say in how AI money gets spent, they need to bring the CEO a plan with real numbers before an outside vendor gets there first. We wrote more about it at State of Brand here: https://lnkd.in/guD6ueAr

New data out this morning: nearly half of marketing teams (47.8%) replaced human tasks with AI in the past year. Only 6.5% measure whether it’s paying off. That’s from Basis' annual AI and the Future of Marketing survey, which was published today. Another 65% either don’t track AI ROI at all or don’t know whether anyone at their company does. In 2023 the number replacing tasks with AI was 10.4%. Last year it was 27.9%. Half of marketing orgs now run AI agents, mostly on reporting, briefs, content and media planning. That’s a lot of what junior marketers used to do..... The contradiction in the data: 70% of marketers say their own job is safe from AI, and half expect big job losses across the industry. We built attribution, MMM and incrementality testing to defend every media dollar. AI is getting a pass on all of it. If marketing can’t show what AI returns, someone else will decide what marketing is worth. Full story on State of Brand here: https://lnkd.in/geM6tguj

Criteria Corp surveyed 1,027 US hiring professionals this summer and 70% of them said a candidate had ghosted them at some point. When they were asked if they'd ever ghosted a candidate, 19% said yes. I'll give them some benefit of the doubt. A lot of what candidates call ghosting is an applicant tracking system that never sent the rejection, or a hiring manager who went quiet and left the recruiter with nothing to say. Most recruiters wouldn't count that against themselves. The person waiting on an answer still does, though. Honestly, the bigger story is the list of hiring priorities for next year.Improving employer brand came in last of the ten options, picked by 12% of respondents. Candidate experience got 31%. I get why. If you run hiring, "employer brand" probably sounds like something the marketing team owns. But most candidates are never going to watch your culture video. They will find out whether you emailed them after the final round. It also seems like a bad year to put this at the bottom of the list. About half of the people surveyed said hiring has gotten harder since last year, and 64% expect to be hiring more in 2027. The candidates everyone wants have other offers, and they remember how the last company treated them. One more number from the report: 73% of hiring professionals said work ethic is the top thing they look for. Candidates would like to see a bit of that from employers too. A two-line rejection email is a decent place to start. Read the full piece on State of Brand: https://lnkd.in/gcXKAJ4j

Eight people on LinkedIn say they interviewed the same single mother. Her husband died eight months ago. She has a 2-year-old, she’s working on her master’s, and she needs a job more for her kid than for herself. Since July that story has gone up on at least eight accounts, nearly word for word, and every one of those posters says they were the person who interviewed her. Joe Meier pulled several of the posts together last week, and once you see them side by side it’s hard to read any of them the same way. Most of the accounts sell resume services. One copy got 4 reactions and another got 516, with the exact same text. We wrote about what this says about LinkedIn right now. The platform has spent the year going after AI slop and engagement pods, but this story got through because it reads like something that actually happened. That’s also why it does more damage than a post someone ran through ChatGPT. Anyone sharing a true story about a candidate now has to write for readers who’ve learned to doubt anything that opens with “I interviewed.” Full piece on State of Brand: https://lnkd.in/gchFE_VZ

When I ran marketing at ClickUp, we did basically everything. Billboards, a big SEO program, every performance channel you can name. So I understand why marketers are treating AI search like the next thing to optimize. Buy the tool, learn the tricks, get the ranking. I don’t think that’s how it works, though. The more I look at what ChatGPT, Perplexity and Google actually cite, the more it looks like journalism. The G2 numbers from this spring stuck with me. About half of B2B software buyers now start their research in a chatbot instead of Google. A year ago it was 29%. And a third of them ended up buying from a vendor they’d never heard of before. If you’re a company nobody’s heard of yet, that’s pretty good news. It also means you can lose a deal you never knew you were in. The other thing people miss is that most of what AI says about you comes from somewhere else. Reddit, Inc., LinkedIn, G2 reviews, trade publications. And ranking on Google doesn’t carry over the way you’d hope. Ahrefs found only 12% of AI-cited URLs were also in Google’s top 10. What does get picked up is stuff that doesn’t exist anywhere else. An interview with a named expert. Your own data, with a date on it. That’s been the whole idea behind Outlever from the start, and it turns out to line up really well with how these tools choose sources. I wrote up what I’d tell any B2B team to do, including the boring plumbing stuff most companies skip. Read the full piece on State of Brand here: https://lnkd.in/gWxqhfZv

Microsoft wants thousands more employees speaking for it in public. In the same memo, it moved communications out of marketing and put it under Brad Smith, whose team handles antitrust, litigation and government relations. Most coverage treated this as a win for employee advocacy. We think the reporting line matters more here.... Someone has to write the guidelines those engineers and sellers will post under. At Microsoft that will now happen next to the people who read every public statement for how it might play in a hearing. It's a familiar pattern. Two weeks ago we covered Vanta's AI writing policy, which came from its general counsel, not marketing. And LinkedIn is testing a feed where Company Pages reach their followers at most, so brand reach is moving toward people whether companies plan for it or not. For CMOs, the question is whether your brand team has a seat when those rules get written. Read the full piece on State of Brand here: https://lnkd.in/gCEAkGhG

Business Insider ran a story this week about “loud working.” Basically, making sure everyone knows what you’ve accomplished so you’re less likely to lose your job in the next round of layoffs. Under LinkedIn's write-up, one man said he’d been let go because his manager didn’t know what he had contributed. Imagine doing the work, then finding out the person deciding whether to keep you had no idea you’d done it. Anyone in marketing will recognize the problem. The channel closest to the sale always gets the credit. The work that got the customer interested in the first place gets harder to defend when budgets tighten. It’s easy to see how that happens with people, too. Everyone remembers who presented the launch results. Fewer remember who spent the previous three months fixing the data pipeline that made those results possible. Marketing leaders are in a tough spot here. They’re already defending their teams when budgets get cut. Now they could be fielding a flood of “here’s what I did this week” updates from people worried their work isn’t being seen. I understand why someone would send that email. But knowing whose work matters shouldn’t depend on who’s best at reminding you... We wrote about what it would look like to measure people with the same care good marketers put into figuring out what actually drove a sale. Read the full piece on State of Brand here: https://lnkd.in/gnkxE8vs

LinkedIn is testing a feed that only shows posts from people you've connected with or followed. It's a small test, and LinkedIn has tried versions of this before without shipping them. But it sent me back to one number from Metricool's 2026 study: only about 7% of Company Pages grew their following last year. Right now the algorithm gives pages roughly the same impressions per post that people get. In a follow-only feed, that goes away. A page reaches its followers at most, and LinkedIn hasn't said yet whether pages appear in the tab at all. So here's a useful exercise for any brand team: if your LinkedIn reach were cut down to the people who actually chose to follow you, how big would it be? We wrote up what's being tested, LinkedIn's history with these experiments, and what brands can do before it decides. Article on State of Brand here: https://lnkd.in/gcasT9KM

Profound pulled the instruction file from Muse, Meta's shopping agent, and shared it with us at State of Brand. Among other things, it tells Muse three times to put well-known merchants at the top of its recommendations. It also has Muse drop listings priced well above or well below the usual range, look at no more than three stores on the open web, and keep in-chat checkout for products in Meta's own catalog. Jennifer Zou, an economist at Profound, ran 50 shopping prompts through Muse. Meta's catalog supplied about 95% of the products Muse considered, but most of what it showed shoppers came from its web search. "Much more weight is placed on candidates returned by the web search," she told us. Her advice depends on who you are. Big retailers can probably rely on being found through browser search. Smaller brands get more out of connecting to Shopify and Meta's catalog. If you run a newer brand that has grown by getting discovered and pricing under the big names, the full piece is worth your time... Read it here on State of Brand: https://lnkd.in/gKR4abpm

One Muck Rack number has been everywhere in PR circles this year: 84% of the links ChatGPT, Claude and Gemini cite point to earned media. It usually shows up as an argument for more PR spend. The same research had a less popular finding. The overlap between the journalists PR teams pitch most and the journalists AI cites for their brands was about 2%. Part of the reason is that "earned" covers a lot of Reddit, Inc. threads, review sites and Wikipedia pages. Nobody gets those through a pitch. So what does get cited? Mostly original research. One set of benchmarks puts its citation rate somewhere between 38% and 65%. Standard blog posts land at 6 to 15%. The usual fallback doesn't hold up much better. SparkToro found 68% of US Google searches ended without a click in the first four months of this year, and company-page posts on LinkedIn reach around 1.6% of followers. Showing up in AI answers turns out to be a publishing job: original data, interviews with people buyers respect, and opinions on the news while it's still news. It has to be kept up for years. Most PR teams aren't set up for that, and they were never meant to be. Our latest piece looks at where the next marketing dollar should go, and why the order of PR and owned media is flipping. Read the full piece on State of Brand here: https://lnkd.in/gS2tDQT5