Everyone will be doing this in five years.
The founders building media systems now will own the audiences everyone else is trying to reach.
But most founders will build the wrong one. https://x.com/i/article/2119833328249360384
Every Founder Will Have a Media System
I think every serious founder will have a media system within three years.
The ones building theirs now will own the audiences everyone else is trying to reach.
But most founders will build the wrong one.
They’ll use AI to produce more posts, clips, and articles from the same weak ideas. They’ll become more efficient at publishing without becoming any more interesting.
The founders who benefit most from AI will have a connected system for deciding what to say, creating it in formats people already respond to, and learning from what happens after it goes out.
I’ve been trying to solve this problem since 2023, although I didn’t always think of it as a systems problem.
For years, I blamed my inconsistency with content on discipline, focus, and time.
I’d have a good idea in a meeting and lose it before I could use it. Or I’d publish regularly for a week or two, then stop as soon as work got busy. When I did sit down to write, it was usually at the last minute.
The problem wasn’t a lack of ideas or discipline.
I just didn’t have a system for turning what I was already thinking into content.
The people who look spontaneous usually aren’t.
After selling my last business, I built a new company with Andrew Wilkinson. We built personal brands for some of the biggest names in the B2B creator economy.
I got a closer look at how people like Nathan Barry and Shaan Puri produced bangers every week.
There wasn’t anything magical about what they were doing.
They collected ideas and batched recordings to create content. They measured what worked and used that to inform the next session.
They had a simple system and stuck to it.
Most founders don’t have that.
And those that do, don’t do it like the pros. They’re asked to show up to content recordings where they’re met with basic questions and prompts.
That’s not good enough anymore.
So I started thinking about what a system for founders would need to do.
It would need to notice the ideas already hiding inside the business, know what content is attracting ideal customers, understand which formats are working and which channels those customers use, and create content that’s highly likely to win business, all without asking the founder to become a full-time creator.
That’s what became the New Media System.
The New Media System is a loop.
It has three connected stages:
1. Listen to customers, the company, and the market.
1. Create original content using ideas and formats supported by what’s working.
1. Measure who paid attention and use that evidence to decide what to do next.
Listen. Create. Measure. Repeat.
Once the loop works for one founder, it can be scaled across the rest of the company.
The system connects to tools the company already uses.
Granola, Zoom, and Gong hold the conversations. Slack and HubSpot hold more of the context. LinkedIn, X, Reddit, Instagram, and Meta show what customers are responding to in the market. Ordinal handles scheduling and publishing. Apify helps us collect public engagement, while HeyReach can help turn relevant signals into connection campaigns.
The stack changes regularly and can adapt to a client’s setup.
The tools aren’t the system. They allow the system to hear more of what’s already happening.
Without listening, content is guesswork. Without creation, the research sits in a document. Without measurement, the company never learns.
Here’s how the pieces connect.
Step one: Listen
Most founder-content systems start with an interview.
Someone gets the founder on a call, asks what they’ve learned recently, and hopes something interesting appears during the recording.
I think the recording starts too early.
Most founders have already produced enough raw material to begin. They’ve explained the company during sales calls, answered questions in internal meetings, spoken on podcasts, and shared opinions in Slack.
Their customers have also been telling the company what they care about.
We can connect Granola, Zoom, and Gong so the system can work through transcripts from sales calls, customer interviews, internal meetings, and product conversations.
HubSpot adds the customer and pipeline context. Slack contains the opinions, milestones, and half-finished ideas people share during the week. Podcast transcripts and previous content show what the founder has already said and which stories have worked before.
That gives us the view from inside the company.
Then we look at the market.
We use Apify and other tools like it to follow public activity across LinkedIn, X, Reddit, and Instagram. We look at what target customers are engaging with, who they follow, and which subjects keep getting their attention.
We also watch competitor posts, outlier content from thought leaders in the niche, ads running through Meta, and the formats already producing bangers in that market.
Any one of those signals can be misleading.
A competitor’s post doing well doesn’t mean the company should copy it. A recurring sales question might be a content opportunity, but it could also reveal a problem with the website or product.
The opportunity becomes more interesting when several signals overlap.
Customers are asking about something in Gong. People from target accounts are engaging with the same subject on LinkedIn. The founder has relevant experience buried in an old podcast or Granola transcript.
Now there may be something worth saying.
The system should bring back a small number of ideas with the evidence attached.
For each one, we want to know:
1. What’s the actual point?
1. Which buyer is likely to care?
1. What evidence supports it?
1. Why has this founder or company earned the right to say it?
1. Why is it worth talking about now?
That gives us something useful to create.
The system can start with zero new founder time
Most founder-content systems depend on getting regular time from someone who already has none.
I’d start without asking for another meeting.
The system can listen to the founder’s existing podcast appearances, internal transcripts, sales calls, and previous content. It can combine those sources with customer conversations and what’s happening in the market.
That’s often enough to identify the first group of stories and opinions.
Once the system is working, a monthly or quarterly 30-minute conversation makes it better.
That call isn’t a desperate attempt to make the founder invent content on demand. We arrive with supported ideas and ask for the part the system can’t find on its own.
The founder can share the story behind a decision, explain what everybody else gets wrong or add an opinion they haven’t said publicly before.
Those conversations supply more original thought leadership, but they’re supplemental.
The system doesn’t stop working if the founder gets busy.
Hiten Shah and I are teaching a free 30-minute lesson on November 19. During the lesson, we’ll show how the New Media System can turn one hour of founder input into 64 pieces of content.
Register for the free lesson.
Step two: Create
Once we know what might be worth saying, we have to work out how to say it.
The system looks for formats already working in the company’s niche. That could be a type of opening, a visual structure, a founder story, a breakdown, or a particular style of video.
Then we apply that format to an original story or opinion from the founder.
The distinction matters.
The format can come from the market. The substance has to come from the company.
If breakdowns of failed projects are working in the category, the founder still needs a real failure and something useful to say about it. If short opinionated videos are performing well, the company still needs an opinion it genuinely believes.
AI can help search the source material, find supporting moments, and turn the founder’s response into different drafts. Editors, writers, and video producers can then finish the work without asking the founder to spend the rest of the month making content.
We create for the channels where the company’s customers already spend time.
For some companies, that will mean LinkedIn and newsletters. For others, it might include X, YouTube, Reddit, Instagram, podcasts, or industry communities.
Different channels can play different roles.
We can publish a higher volume of ideas on X to find stories and arguments that earn a response. The winners can then move to LinkedIn, where they receive more development and amplification.
A short X post might become a detailed LinkedIn story. If the idea keeps attracting the right buyers, it could become a newsletter, video, customer story, paid Meta creative, or something sales can use during an active deal.
This isn’t about repurposing every recording across every channel.
It’s about using each channel to learn which ideas deserve more investment.
After the content is approved, Ordinal schedules and publishes it across the relevant company and executive accounts.
When it goes live, the system starts listening again.
Step three: Measure
Most content reports tell you how many people saw a post and how many clicked like.
That isn’t enough to improve the system.
We need to know which ideas worked, who paid attention, and what happened next.
Apify can collect the public likes and comments on the company’s LinkedIn posts. We can do the same for relevant competitor posts and identify people from target accounts who are already engaging with a subject.
We can enrich those people and companies with business information, then check HubSpot to see whether the account is already in the pipeline or has an existing relationship with the company.
Where it makes sense, HeyReach can add those people to a connection campaign. Sales can also use the subject that caught their attention as context for a more relevant outbound message.
Instead of telling sales that a post received 500 likes, we can tell them that three people from a target account engaged with an argument that also appeared during a recent sales call.
That doesn’t prove the post created buying intent.
The buyer may already have been evaluating the company. They may simply be interested in the subject. The content might have created the interest, influenced it or appeared at the same time.
We shouldn’t claim more than the evidence shows.
But the engagement is still useful when it’s combined with website activity, CRM history, and recent conversations.
Measurement also improves the content itself.
If a story attracts the right buyers, we can develop it further. If an idea performs on X but falls flat on LinkedIn, we can look at the format, audience, or timing. If target customers keep engaging with a competitor’s argument, we can decide whether the company has a credible response.
Those lessons return to the listening stage.
The next batch of ideas starts with everything the company has already learned.
What the complete loop looks like
Imagine several prospects start asking about implementation time during Gong calls.
The system finds the question in the transcripts. Apify also notices that people from target accounts are engaging with competitor posts about slow implementation.
The company has something credible to add because the founder explained on a podcast how the team gets customers live quickly.
Those signals create a recommendation: the company should make implementation speed part of its story.
We find a format already working in the niche and apply it to the founder’s original experience. A few versions of the argument go out on X.
The strongest story is developed into a LinkedIn post, then scheduled and published through Ordinal.
Once it’s live, Apify collects the people engaging with it. The relevant accounts are enriched and checked against HubSpot. HeyReach can invite the right people to connect, while sales gets the context it needs to decide whether a conversation makes sense.
Meanwhile, the comments, engagement, and later sales calls tell us whether the idea is still showing up.
That evidence goes back into listening.
If the story keeps working, it might become a customer story, sales asset, paid Meta campaign, or larger company narrative.
That’s the full system.
It moves from the conversations already happening in Granola, Zoom, and Gong to ideas supported by customer and market evidence.
Those ideas become content for the channels where customers already spend time. Ordinal gets the approved content published.
Apify, HubSpot, and HeyReach then help the company understand who responded and what it should do next.
Listen. Create. Measure. Then listen again.
Then you scale it
Start with one spokesperson, one buyer group, and one primary channel.
Once the loop works, more people can take part.
A product leader can explain why something was built. Someone in sales can address the objection they keep hearing. Customer success can show what the best customers do differently. The CEO can make the larger argument about where the market is going.
They shouldn’t all sound the same.
The advantage comes from giving each person access to the same customer evidence, company narrative, and understanding of what the business wants to become known for.
Eventually, the media system becomes part of the company’s memory.
It knows what customers have asked, what target buyers are paying attention to, what its leaders have said, which stories have already been told, and what happened when those ideas reached the market.
A competitor can copy the tools.
It can’t copy that history.
I’ve turned what I’ve learned since 2023, including work around companies like Salesforce, Bolt, Shopify, Loom, Buffer, and Drift, into a system other people can build.
If you want to build it, Hiten Shah and I are teaching a free 30-minute lesson on November 19.
We’ll show you how to decide what your company should say, turn one hour into 64 pieces of content, and use the response to improve what you create next.
Join us on November 19.