I spent 80 hours on planes last week. Laptop closed. Wi-Fi off. Phone away. Just thinking.
The business management framework we use at Level, EOS, calls this a clarity break. I'd encourage everyone to build them into your routine. It doesn't need to be 80 hours. Two hours here, two hours there, where you're not checking email or solving the next client issue. You're just thinking about yourself, your team, and how everything fits together.
Here's what came out of it.
In any organization, there are three competing forces: innovation, growth, and structure. All three matter. But they're naturally in conflict with each other at all times. When you push hard toward growth and innovation, structure breaks down. When you invest in structure and efficiency, the other two can stall. The instinct is to push all three at once, and then we get frustrated when something gives.
That frustration is the wrong response. The real strategy isn't what you say yes to. It's what you say no to. If we're a little messy on the structure side while we drive growth and innovation forward, that's okay. Saying it out loud matters. It means we're honest about where we are instead of pretending we can be everything all at once.
If you haven't taken a clarity break in a while, put one on the calendar. You don't need an airplane. You just need two hours without a screen.
Patrick Patterson
CEO at Level Agency
Reddit for Marketing Leaders: Replay + recap
Last week we partnered with Reddit for a virtual insider session on AI visibility, community trust, and performance marketing.
The recap covers how Reddit is co-cited alongside brand websites in nine out of ten AI-generated answers, why AI models pattern-match for consensus instead of popularity, how brands like Philadelphia Cream Cheese and Mayo Clinic are showing up on the platform without getting rejected by the community, and what paid formats actually work when your audience came to learn.
If you missed it, the full replay and key takeaways are live now.
Level is going to LeadsCon 2026, April 22-24 at the MGM Grand in Las Vegas. Here's what we're bringing:
Booth #463: Our full team will be on the floor all three days. Come talk lead gen, measurement, AI, whatever's on your plate.
Speaking session: "From Attribution to Incremental Value: What It Really Takes to Be Data-Driven in Modern Marketing" with Nicole Jennings (Level), Ryan Staniak (COUNTRY Financial), and Brad Stephenson (Level) moderating. Thursday at 3:15 PM.
Live demos: We're showcasing the Level Intelligence Suite at the booth, our proprietary AI-powered toolkit built for performance marketers. Come see it in person.
Nintendo Switch 2 giveaway: We could have gone with a branded tote bag. We didn't. Booth #463 to enter.
If you're going to be in Vegas, we'd love to connect.
OpenAI COO says ads in ChatGPT will be "an iterative process" OpenAI's Brad Lightcap confirmed the company is committed to getting ads right inside ChatGPT, emphasizing user trust and privacy as top priorities. The interesting framing: ads "done right can be additive to a product experience." If you've been tracking the answer-layer monetization shift we've covered in past issues, this is the next chapter. Give it a few months, Lightcap says. Growth leaders should be watching closely. Read more
Gemini task automation is slow, clunky, and super impressive The Verge went hands-on with Gemini's new task automation for Uber, DoorDash, and more. It's early. It's rough. But it's a real preview of where agentic AI is heading, and the speed of iteration should make every growth leader pay attention. Read more
"You are an expert" prompts can damage factual accuracy New research shows that assigning an expert persona to an LLM can actually make it less accurate. The model becomes more confident but not more correct. Worth keeping in mind as teams build AI workflows and assume that more specificity always equals better output. Read more
Why CMOs struggle and how aligning with CFOs changes everything The CMO-CFO relationship is the most underleveraged dynamic in the C-suite. MarTech makes the case that getting it right unlocks budget, credibility, and strategic influence. If your marketing team struggles to justify spend, the fix might be relational, not analytical. Read more
Where is email marketing headed in 2026? Litmus Live panelists got into how AI inbox sorting is making relevance the new deliverability. Opens and clicks are increasingly unreliable metrics thanks to bots and Apple's Mail Privacy Protection. The practical shift: engagement signals like conversions, site visits, and downstream behavior matter more than top-level vanity numbers. Read more
Google Ads for sensitive categories: What you need to know If you're running Google Ads in regulated verticals like healthcare, finance, or education, policy restrictions shape what's possible before strategy even enters the conversation. WordStream breaks down what's changed and how to navigate it. Read more
Deep dive: AI
A rogue AI agent caused a serious security incident at Meta
An internal AI agent at Meta posted a response to an employee's question on an internal forum without authorization. A second employee acted on that advice, which turned out to be inaccurate. For nearly two hours, sensitive company and user data was accessible to engineers who weren't cleared to see it. Meta classified it as a Sev 1 incident, their second-highest severity level.
The agent wasn't some experimental side project. It was likened to OpenClaw, the open-source agentic model that's gained serious traction in enterprise environments. It held valid credentials. It passed every identity check. It just acted outside its intended scope.
This follows a separate incident where Meta's own AI safety director described an OpenClaw agent deleting her entire email inbox while ignoring her commands to stop.
For growth leaders deploying AI agents internally, the lesson isn't to slow down. It's to invest in guardrails before the agents need them. The question isn't whether your AI agent will act outside its boundaries. It's whether your organization will notice when it does.
The Trade Desk headlines are a symptom. The disease is older than you think.
Our own Anjlee Majmudar, VP of Programmatic Media, wrote this one, and it's the piece I'd send to any client asking what the Trade Desk controversy actually means.
The short answer: it's not a Trade Desk story. It's a programmatic story. Fee opacity, hidden markups, and auto-enrolled tools aren't new. They're structural. Dollars flow through multiple layers before reaching a publisher, and at every layer someone takes a cut. Some of those cuts are disclosed. Many aren't.
Anjlee traces it back to the ANA's 2016 transparency report and the ISBA/PwC study that found 15% of advertiser spend was completely unattributable. The current headlines aren't a revelation. They're a reminder that the incentives haven't changed.
The piece also lays out what a different model looks like: radical fee transparency, multi-DSP flexibility, and supply path optimization that puts the advertiser in control. If programmatic is a line item in your budget, this is required reading.
Descript is an AI-powered video and podcast editor built for people who don't want to learn a traditional editing timeline. The core concept: you edit video by editing a transcript. Delete a sentence from the text, and the video cuts with it. Rearrange paragraphs, and the footage follows.
The AI co-editor, Underlord, handles cuts, pacing, and filler word removal from plain-language direction. Studio Sound cleans up audio to near-studio quality. They've also added lip-synced translation across 39 languages.
If your team produces webinars, podcasts, or social clips, this is where the "we don't have a video editor" excuse stops working.
What does ChatGPT actually say about you?
When a potential buyer asks an AI agent for a Top 5 list in your category, do you make the cut? More importantly: Why or why not?
Don't leave your brand’s reputation to an algorithm you haven't audited. Our scorecard pulls back the curtain on how AI platforms frame your brand sentiment and which sources they use to vet your authority.
Inside your custom report:
Sentiment analysis: Is the AI's summary of your brand accurate or outdated?
Citation audit: Discover which third-party sites are feeding the AI your info.
AEO roadmap: Immediate steps to improve your "AI discoverability."
