Weekly B2B Marketing Digest / Week 45

• Tas Bober – Landing pages fail when the ask doesn’t match the intent

Tas Bober talks about how many landing pages jump straight to high-friction CTAs like demos and sales calls, even when buyers aren’t ready. She breaks down three levels of offers and shows why matching the ask to the buyer’s stage makes pages convert better. Tas makes the case for slowing down and giving people the right next step instead of pushing too hard too early.

 

 

 

• Josh Blyskal – ChatGPT’s Entity Update is shaking up brand visibility

Josh Blyskal explains that ChatGPT is now mentioning fewer brands per response and visibility has dropped across the board after a big structural update. He believes the shift toward entity recognition is laying the groundwork for ChatGPT Shopping, where brands need to be validated across sources like Wikipedia, Reddit and review platforms to show up. Josh says the gap between top and lower-ranked brands is shrinking, which creates more room for challengers.

 

 

 

• Jean Bonnenfant – Why attribution doesn’t reflect how people actually buy

Jean Bonnenfant shares Elena Verna’s take that attribution, especially last-click, misses most of what influences a buyer, since real decisions come from a mix of touchpoints spread out over time. She highlights how Google often gets credit for demand it didn’t create and why marketers need to look beyond dashboards to things like surveys, MMM and creative experiments. Jean brings it back to the idea that marketing is about understanding people, not just numbers.

• Bram Van Der Hallen – Why overcomplicating Meta campaigns hurts performance

Bram Van der Hallen calls out the obsession with fancy frameworks and complex Meta setups, reminding marketers that performance comes from basics like a strong offer, clear goals and good creative. He explains that over-splitting campaigns only limits learning, weakens results and wastes time. Bram argues that Meta works best when you stop micromanaging and let the algorithm do its job.

• Morgan J Ingram – How to build a social selling program that actually works

Morgan J Ingram says most companies fail at social selling because they force everyone to “do LinkedIn” without a real plan. He breaks down a practical approach: start with volunteers, test with SDRs, document what works and build momentum from the team up. Morgan makes the point that social selling is about relevance, personal connection and culture, not forced posting and generic DMs.

• Anthony Blatner – LinkedIn’s flexible ad creation still misses the mark

Anthony Blatner is excited that LinkedIn is finally adding flexible ad creation but says the current version isn’t what advertisers really need. Instead of letting users pair specific creatives and copy, LinkedIn mixes everything automatically, which doesn’t work for B2B campaigns. Anthony urges LinkedIn to build the feature so marketers can manually match the right assets together.

 

 

 

• Julia Kinner – Why most segmentation frameworks fall apart

Julia Kinner explains why many companies’ segmentation models don’t actually help them prioritize or make better decisions. She points out the most common issues: no sense of value or right to win, segments that can’t be acted on and a lack of clear path-to-purchase data. Julia says if your segmentation hits too many of those problems, it’s probably time to rebuild it properly.

 

 

 

• Brendan Hufford – Why he turned down 3 clients last month

Brendan Hufford explains that he walked away from three clients because they couldn’t answer three basic questions: who their ICP really is, what pain actually drives their decisions, and which single problem they’re willing to own. He shows how vague answers lead to checkbox marketing, while clarity around one deeply-owned problem leads to real authority and impactful content.

• Peep Laja – What his 2025 research studies taught him

Peep Laja shares what he learned after running multiple original research studies: the topic matters more than anything, repeating strong studies turns them into long-term assets, and distribution partnerships make a huge difference. He argues that original research is still one of the smartest plays in B2B SaaS – especially as AI makes generic content easy and forgettable.

• Curtis Howland – How to tell if creative or media buying is the real problem

Curtis Howland breaks down how to diagnose ad performance by separating creative metrics from media buying metrics. He explains that grouped data hides real issues and that both teams impact CPA and ROAS differently. His point: measure each side independently so you fix the right thing instead of blowing up a system that’s actually working.