Founders are calling their content marketing "new media" now, and I think the rename is mostly there so nobody has to say the word content out loud.
I get why. Spor made the honest case for it last night, that "content" sounds like TikTok slop and tech people don't want to be lumped in with that. Fair. But the work underneath is the same, it's still founders making videos about their own category and hoping strangers pick them over everything else on their phone.
People are paying a lot more for it though. There are Head of Media job posts out right now at $150k to $220k, with editors, a studio and on-camera talent attached, and the founders behind one of them say they want most of their leads to come from organic video. That's a newsroom's fixed costs with a marketing team's distribution.
The most useful number here comes from someone who actually runs a media company. Alex Konrad went through his own Upstarts numbers for 2025, and about 10% of the traffic to his biggest story came from X, and under 5% from LinkedIn for each of his top five. His own read was that social is good for awareness and vibes and not for getting people to read the thing. And that's a real media business with scoops people actually want.
So before the studio budget I'd ask one question, would anyone watch this if your logo wasn't on it? If yes, you're building media and good for you. If not, it's content marketing with better lighting, which is fine, just budget it like content marketing.
ืืงืื ืฉืื ืืื ืืืืขืื ืฉืืืื ืืื ืืืืชื ืืืืฆืื ืืช ืืืืืกื ืืืืืืจ ืฉืืฆืืืจื ืฉื ืืืจื ืฉืขืฉืืชื ืืืื
Every marketing team is supposedly behind on GEO right now. Most of them are going to burn a quarter chasing traffic they can't even find in their own analytics.
Marie Haynes pulled 12 months of GA4 data from 100 sites, about 145M sessions. ChatGPT sent 0.62% of what Google organic sent. All the LLMs together got to 0.73%.
ChatGPT visitors do convert better, 13.8% vs about 3%, and that's the one number every GEO agency puts on slide one. Google organic still brought roughly 27x more conversions than all of them combined.
And 45 of those 100 sites weren't tracking a single key event. They have no idea what converts on their own site, but sure, let's buy a dashboard of ChatGPT mentions.
I'll care about GEO when it's a number I'd notice if it disappeared.
Idan Benishu ๐ retweeted
It blows my mind how many PRDs die the day the feature ships. Everyone fights over the doc for two weeks, it gets approved, the thing ships, and nobody opens it again. The success metric on page one sits there like a gym membership in February. Aakash Gupta said it this week: the PRD is now โa decision record more than a permission document.โ Fine, but a record nobody reads after launch is just a nicer-looking permission slip. Write these 4 things into it, or donโt bother writing it:
1. Problem = a number + a quote
Most PMs write โUsers struggle with reporting.โ Thatโs an opinion with a font. Top PMs write how many accounts raised it in the last 90 days and paste two exact lines from sales calls and support tickets under it. Pull the count and the quotes before you write a line of scope.
2. Shipped โ Success
Most PMs write โLaunch in Q1โ under success. Cool, thatโs a date. Top PMs name the one metric the feature has to move (activation, NPS, CSAT, expansion, whatever your business actually runs on) and the number it has to hit to stay. Write that number before the first prototype, while nobodyโs in love with it yet.
3. The kill switch has a NAME
Most PMs skip rollout or write โgradual.โ Gradual to what? Top PMs write how many users get it, for how long, and who flips the switch. If nobody owns the kill switch, nobody flips it, and youโll babysit a dead feature for two years because killing it feels like admitting you were wrong. Put a name and a date next to it.
4. Results go on TOP
This is the one almost nobody does. Most PMs ship, post a launch note and start the next PRD. Top PMs come back at 30 and 60 days, write down what moved and what didnโt, and put the results link at the top of the doc. Marty Cagan argued this week that experimentation should be โthe core competency of product organizations.โ This section is where that actually happens, or quietly doesnโt. We see the cost of skipping it in the feedback that runs through @trybagel
Before you start the next PRD, open the last one and read the results...
No results, no new PRD.
Go open the last PRD your team shipped and scroll to the top. If the newest thing on it is โStatus: Approved,โ that doc stopped working the day it got a yes.
Put a 60-day review on the calendar the day you ship, with the PM who wrote the doc and whoever owns the metric. If that meeting keeps getting pushed, you already know how the feature did.
ืขืืฉืื ืืื ืืืืช ืฉืื, ืฉืืืจื ืฉื ืืืื ืืืช ืืืฉืืืืช ืขื ืืืืื ืืืจืืืช ืฉื ืฉืดืก. ื ืืกืขืช ืืื, ืืืกืืช ืืช ืืชื ืืขื, ืฆืคืืจืืช ืืืืืืงื ืืืืืืื ืืืืจืฃ.
ืืืืื ืืืืชื ืฆืจืื ืขืื ืกืืืืช ืืกืืื ืืื
ืื ื ืืช ืืจืืืช ืืืืืกืืื ื ืื ืกืื ืืืืื ืืื
Replying to @Nadav_Eyal
ืฉืืื. ืื ื ืืืืื ืืืืืงืฆ'ืืื. ืื "ืืืขืฅ ืืฉืงืขืืช" ืืื ืงืจืืคืืืืจืฃ ืขื ืืืงืืืจื ืืชืืื.
ืื ื ืงืืจื ืืชืืืืื ืืื ืืชืืืืืช ืื ืกืืช ืืืกืืืจ ืื ืืื ืื ืืืฉืืื ืฉืืจืขืืื ืฉื ื ืชื ืืื ืืืื/ืืืืจ ืืขืืื, ืืื ื ืืกืืืจ ืืื ืืื ืืชื ืืืขืื.
AI video got basically free this month, and I already know what my feed looks like in three months. Every startup posting the same gradient launch video with the same voiceover and the same "we're excited to announce."
Most brand teams were hiding behind production value. A big video budget let them skip the uncomfortable question of whether they had anything to say, because the polish did the talking for them.
I did it too. I played with these tools a lot, with my own brand design system, a look I worked hard to make unique, and years of knowing how to cut a sharp video, and what came out was still kind of sloppy. Nobody would remember it a day later.
So the win is still in the human part, the dirty, imperfect stuff, as long as it's attached to a product, a feature or a message that actually changes something for someone. If your launch only worked because it looked expensive, it never worked, and a lot of teams are about to find that out in public.
Smart CEO is smsrt
90% of product teams canโt answer โwhatโs your P0?" anymore, and most of them have five agents running while they canโt answer it.
Donโt just take my word for it, ask Thariq Shihipar from the Claude Code team, who said that if youโre juggling a bunch of agents at once, you should slow down, because โmost of the stuff youโre doing won't really matter."
Heโs right.
Five agents in five tabs feels like a team of ten, and most weeks itโs a team of ten building P3s really, really fast.
The agents will build whatever, they donโt care if itโs stupid, so the hard part of the job moved upstream to picking whatโs worth building and being able to say why, and thatโs exactly where most PMs suck right now.
Hereโs what most PMs do with their agents vs what top PMs do, in the 5 moments that decide your week:
1. Picking the P0
Most PMs pick whoever pinged last, the sales escalation or the CEOโs message at 11pm on Sunday, and call it prioritization. Top PMs pick from evidence: how many accounts hit the problem, in which segment, and what fixing it does to the metric the company runs on, whether thatโs ARR, NPS, activation or some custom formula. If you canโt say the evidence in one sentence, what youโve got is a loud customer.
2. Starting the agents
Spinning up an agent per idea feels free. It isnโt, not even close. Every run lands on your review queue, and review is the one part of the job you canโt run in parallel no matter how many tabs you open, so ten agents just means ten things sitting on your desk by Friday. Top PMs put ONE agent on the P0 and actually watch it, and the spare tabs go to research on the P1. Cap your parallel builds at what youโll really review this week, and be honest with yourself about that number.
3. Feeding the agent
Most PMs plug in every MCP, every transcript and the whole ticket backlog, type โfigure it out" and go get coffee. Youโre toast. The agent burns tokens chewing on noise and builds the average of everything it read. Top PMs give it the decision: what weโre building, for which customers, why, and the exact customer words behind it. When we modeled this, 12.3M tokens of raw feedback came down to 904K tokens of validated decisions, so the agent reads less than a tenth of the stuff and all of it matters. One page per P0 with the problem, the accounts, the quotes and the metric, and nothing else gets in.
4. Saying no
Most PMs say โnot now" and let the request rot in the backlog, where it comes back to haunt them every planning cycle. Top PMs tell the customer and the AE straight up whatโs not happening and show them why. Slow down here, because a clear no with the evidence attached ends the conversation, and the AE stops bugging you about it every Monday.
5. After it ships
The day the last P0 ships, most PMs are already sprinting at the next one and never look back. Top PMs check 30 to 60 days later whether the customers who asked got what they needed, and thatโs how they learn which evidence to trust next time.
The replies to Thariq are split. On one side youโve got the maximalists, who want ten agents running because throughput wins, and on the other side the skeptics, who want everyone back to doing it by hand.
Both wrong.
Neither camp is talking about what the agents are pointed at, and thatโs the whole game, so run as many agents as you can actually review and point every single one at something you can defend with evidence.
So pick one thing, know why itโs the one, give it your full attention and let the rest wait, including the CEOโs Sunday night aha moment.
BTW, this is the whole reason we built Bagel AI.
When youโre scaling fast or sitting on hundreds of enterprise accounts, the evidence for your P0 is spread across sales calls, support tickets, chats and threads that no one has time to read. Bagel AI reads all of it, ties it to the metric you run on, and gives you and your agents the validated decision with the evidence attached, then checks 30-60-90 days after it ships whether it worked.
bagel.ai/platform-overview/
ืืืื ืชืืืื ืื ืื ืืชื ืืืฉืืื:
ืืืืข ืืื ืืงืืืฆืช ืคืืืกืืืง ืชืืืื ืฉื ืืื ืืื ืืืืืจืื ืืืื ืืืณื ืจืืื ื-AI ืืฉืืื ืฉืืื ืคืฉืืื ืดืืื ืื ื ืขืืฉื ืกืื ืง ืฉื ืืืืืจ ืฉืืงืืืชื ืืืื ืืืืช ืฉืืฆืจืชื ื[ืืื ืก ืืืื ืื ืจื].
ืืืงืื ืืงืื ืชืฉืืื ืดืื ืก ื-11labs ืชืขืฉื x" ืื ืดืชื ืกื ืขื ืงืืื ืืชืขืฉื ืืืืด ืืื ืืืฆืฃ ืืขืฉืจืืช ืชืืืืืช ืฉื ืืื ืืื ืืืืืจืื ืืืื ืืืณื ืจืืื ื-AI ืฉืื ื ืืื ืฉืขืืฉื ืืช ืื, ืื ืืืืื ืขืืฉืื ืืืชื ืืืจ, ื ืจืืื ืืืชื ืืืจ ืืืชืืฆืื... ืืื.
ืืื ืืืืืง ืืขื ืืื ืขื ืืืคืฉืจืืช ืืื ืืื ืืช ืื ืฉืื ืื, ืืกืืฃ ืื ืื ื ืืืื ืืืืืืืื, ืืฃ ืืื ืื ืืืฆืื ืืืืื ืงืืืื ื ืืช ื-WIZ ืืื, ืืฆืืขืจ ืื ืคืฅ ืืช ืืืฉืืื. ืืกืืฃ ืืชื ืืื ืื ืืช ืืืชื ืืคืืืงืฆืื ืฉืขืืงืืช ืืืจื ืืืฉืงื, ืืืกืคืช ืืช ืืงืืืืช, ืขืืงืืช ืืืจื ืืืฆืืืช ืืืืณ
ืืืฉื ืืช ืื ืืืืขื ืืืืื ืงืืืื ื, ืืื ืืืืขื ืืืืืื ืื ืืฉ ืืขืื, ืืื ืืฉ ืืืชื, ืืื ืืื ืืืจ ืคืชืืจื, ืื ืื ืืื ืคืืชืจืื ืืื ืื, ืืื ืคืืชืจืื ืืืชืจ ืืื.
ืืืจืื GTM ืืื ืืื ืืืกืคืืง ืืืคืจืื ืื ืืื ืฉืชื ืกื ืืืืืฃ ืืช ืื ื-GTM Engineering.
ืืืืืจ ื-RevOps ืขื ืขืื ืืื ืฉืืืื ืื ืฉืชืืกืืคื ืืื Engineer ืื ืื ืืฉื ื ืืช ืื ืืื ืื ืืืื ืืื ืขืื ืืื ืืืคืื ืืกืืฃ ืืืืืฉ. ืื
Don't f^$% with the CFO
โWhereโs the ROI on all this AI spend?" is going to be the biggest f*ckup moment in 2027 product and engineering budget reviews.
Accenture just capped Claude at $29 an employee, and every finance team is about to ask the same thing. Most product and engineering leaders will walk in saying โwe ship twice as fast now," which is telling the CFO the factory ran all night without checking if anyone bought what came off the line. Five answers are going to get people in trouble with the CFO:
1. โWe ship twice as fast now."
Nice! But the CFO doesnโt pay for speed.
Speed is a cost line until something on the business side moves, and a PR count canโt tell anyone what moved. Before the next build starts, product and eng write one line together naming the metric it should move. No line, no build.
2. โWe finally cleaned up our tech debt."
Engineers love (!!!) that sentence. Your CFO just checked their phone.
A cleaner codebase is invisible from finance. Name the feature that shipped because of the cleanup, and what it did for the customers who asked for it.
3. โThe team saves 10 hours a week."
Great. Where did the 10 hours go?
If they went into more features nobody asked for, you bought a faster way to fill the roadmap with guesses. Show which customer problems got solved this year that sat in the backlog last year, with the sales calls and support tickets that asked for them attached.
4. โUsage of the new features is up."
Up compared to what? Every launch gets a bump, and your own team clicking around counts too.
This is the one that hurts most, and we see it all the time in the feedback that runs through Bagel AI. A team ships a request, closes the ticket, and nobody checks whether the accounts that asked for it ever touched it. Three months later those accounts ask for the same thing in different words.
Go back 60 to 90 days after launch. Check the metric your business runs on for that segment (retention, NPS, expansion or active users), then check whether the customers who asked are the ones using it. If they arenโt, better you find out than the CFO.
5. โLetโs cap the spend until we figure it out."
Yeah. Good luck with that. A cap shrinks the bill and the output of your best builders with it. Pull your top 10 spenders across product and eng, list what each one shipped last quarter and what it moved, and keep funding the ones with an answer.
Budget season rewards whoever walks in with evidence and results. You still have a quarter to collect and prove it.
So this is the time to ask yourself: which launch or AI rollout from this year would you least want your CFO to ask about?
ืชืฉืืื ืื ืื ืื ืืืฆืจ ืืื ื ืจืื ืืื ืขืืืื ืฉืื ืืื 2025 ืืขืจื ืืชืงืืื ืืช ืืืชื ืชืฉืืื, ืืืื ืืืืื - ืขืื ืืื ืขืฆืืืืช ืืืื
ืื ื ืื ืืกืืื ืืืืื ืืื ืืืืฉ ืืื ืืื
ืืฉ ืืืื ืืชื ืืืืช ืืืฆืืจื ืฉื ืืืืืงื ื-AI ืืื ื ืืืื ืืืืื ืืช ืื, ืืื ืืกืชืืจ ืฉืืฉื ื ืืขืืช ืืืฉืืช.
ืื ืฉืื ืืืฉืฉืื ืืืืืื ืืื ืืฉืืืช ืฉืืืืืืงื, ืืงืฉื ืืื ืืืืืื ืฉ-AI ืืืื ืืืฆืืจ ืืืืืงื ืืขื ืืื ืช ืื ืืงืืจืืช ืืืืืื.
ื ืืื ืืขืืฉืื, ืืชืืฆืืืช ืฉื ืื ืืขื ืืฆืืจืช ืืืืืงื ืืื ืกืื ื, ืื ืืืืืืช ืืืื ืืืืื ืืืืื ื ืืื ืืช, ืืืืืื ืช ืืืคืงื, ืืกืืื ื ืืืขืืืื, ืืื ืืืจื ืืื ืืฉืืจืื ื ืฉืืขืื ืื ืจืืื ืืื ืืืืื ืืืืื ืช ืืืืืืืืช ืืืืงืกืืื. ืืืืื ืฉืฉืืืื ืืืืื ืืช ืืคืจืืืคื ืืืืื ืืืฉืคืืข ืคืืืื ืขื ืืืืืช ืืชืืฆืื.
ืืฉืืืข ืืฆื ืื ืืคืืืฉ ืฉืชื ืืขืืช ืืืฉืืช ืืฉืื ืืช ืืืืืืื ืืืืขื ืืืืืจืช ืื ืืืื ืืืืืงื ื-AI.
ืืจืืฉืื ื ืืืืชื ืืฉื ืื ืกืชื ืืืื ืืช.
ืื ืื ืืงืฉืื ืืฉืืจ ืืืื ืืืจืืงืื ืืืืฆืืข ืฉื ืืืจ ืืืื ืืชืืง, ืฉืื ืื ืืืืืจื ืืืื ืืงืืกืืืช.
ืขื ืืืกื ืฉื ืื ืื ืื ื ืจืืื ืชืืื ื ืฉื ืืืืืงืื ืืืื ืฉืืืจ, ืืื ืืฉืื ืงืฆืช ืืฉืื ื. ืืฉืืืชื ืื ืืืืืืงืื, ืืื ืขืื ื: ืดืื, ืื AIืด.
ืืืคืชืขืชื!
ืื ืื ืืืฉืื: ืดืื, ืชืฉืืข, ืื ื ืืืื ืืืื ืืืื ืืืื, ืื ื ืชืืื ืฉืืืข ืืืืืด.
ืืชืืืชื ืืืกืชืงืจื: โืชืืื, ืื ืืกืจื ืื ืืื ืืฉืืืช ืืืืฆืืข ืืื?โ
ืืืื ืขืื ื: ืดืื, ืืข ืื ืฉืื ื ืืคืืื ืืขืืืฃ ืืืืืงืช AI. ืืฉ ืืืืืืงืช AI ืืฉืื ื ืงื ืืืชืจ, ืืื ืืืืคืื, ืืื ืืืจืืงืืช. ืื ื ืืขืืืฃ ืืืืืงืช AIืด.
ืืกืืคืืจ ืืฉื ื ืฉืฉืืขืชื ืืฉืืืข ืืกืืคืจ ืืืืืืื ืฉืื ื ืืฆืื ืืื ืืคื ืืื, ืกืจืืื ืืืื ืืขืจืืฆื ืืฆืืืื ืืคืืคืืืจืืื ืืืชืจ ืืืืืืื, ืขื ืืขื ืืืืืืื ืื ืืืื, fsstoppers.
ืืืืฆืจ ืืกืคืจ ืืช ืืกืืคืืจ ืฉืื ืืื ืื ืฉืงืฉืืจ ืืืืืืงื ืืกืจืืื ืื ืฉืื.
ืืื ืืกืคืจ ืฉืืชืืืืช ืขืืื ืืืืืื ืืืืช ืืื ืืืืก ืืืืื ืืื ืฉืื ืืฉืชืืฉื ืืืืืืงื ืืืืคื ืื ืืืงื.
ืืืจืืข ืฉืขืืฆืื ืืืงื ืืืืืืช ืืืืฆืจืื ืืจืฉืช, ืืื ืืชืืื ืืงื ืืช ืืืืืงื ืืืืคื ืืืงื ืืืชืจื ืกืคืจืืืช ืฉื ืืืืืงื, ืืืืฉืชืืฉ ืืกืจืืื ืื ืฉืื ืื ืืจืง ืืืืืืงื ืฉืืื ืงื ื ืืืืคื ืืืงื.
ืืื ืขื ืืฉื ืื, ืืื ืืชืืื ืืงืื ืชืืืขืืช ืขื ืืคืจืช ืืืืืืช ืืืฆืจืื ืืืืืืงื ืฉืืื ืงื ื ืืืืคื ืืืงื.
ืื ืจืื ืฉืื ืฉืื ืฉืืืจื ืืช ืืืืืืงื ืืืชืจืื ืืืื ืืชืืจืื, ืื ืฉืงืจื ืืืจืื ืืืจืื ืืืื ื ืืืฅ ืืฉืื ืขืฉืจืืช ืืืคื ืืืืจืื, ืืชืืืขืืช ืฉืืคืกืื ืืื. ืืืืช ืืื ืืื ื ืืฆื ืืื ื ืืืฅ ืืฉืื ืืจืื ืืืื ืืกืฃ ืืขืืจืื ืืืื ืฉืื.
ืืื ืื ืืืฉืื ืขื ืฉืืืืข ื-AI, ืืืื ืืฆืจืืช ืฉืื ืืกืชืืืื. ืืื ืืืืฆืจ ืืกืื ื ืืืืืงื ืืกืจืืื ืื ืฉืื ืฉืืชืืืื ืืืืืง ืืชืืื ืฉืืื ืืืฆืจ.
ืืืืจ ืืืขื ืืื ืืืืชืจ ืฉืืื ืืกืคืจ ืื ืฉืืืืจื ืืฉื ืื ืื ืฉืื ืืืขื ืืื ืืชืืืืกื ืืืืืืงื ืืกืจืืื ืื ืฉืื, ืืื ืืื ืฉืืื ืืชืืื ืืืฉืชืืฉ ืืืืืืงืช AI ืฉืืื ืืฆืจ ืืืืคื ืืืฉื ืืื ืืื ืืืกืจืืื ืื, ืืื ืืชืืื ืืงืื ืืืชืจ ืืืืชืจ ืชืืืืืช ืฉื ืื ืฉืื ืฉืฉืืืืื ืืืชื ืื ืื ืืืืืืงื ืืืืช, ืฉืื ืื ืื ืืืืืื ืืช ืืืืืืงื ืืืืช, ืืืืืงื ืืืืืงื ืฉื ืืฆืจื ื- AI.
ืื ืื ืืชื ืืืฉืืื ืขื ืื?
ืืื ืื ืื ื ืืชืืืืื ืืืืื ืก ืืชืงืืคื ืฉืืืืืงืช AI ืืชืืืื ืืขืฆื ืืงืื ืืืืืืืฆืื, ืืืฉ ืืคืืื ืื ืฉืื ืฉืืขืืืคืื ืืืชื?
ืื ืืขืชืื?
youtu.be/gWHvr4-LW18?si=59vQโฆ
ืชืืืื. ืืชื ืื ืืืื ืื ืฉืื ืื ืืื ืืืืฆืจ ืืขืฆืื ืืืืืข ืืคืงื ืกืืจืืืกื ื ืขื ืืืืืจื ืืืคืฉืืช ืฉืื?












