The question you are now getting
A large firm made media its stated differentiator, published the numbers, and the rest of the industry has spent the months since being asked to explain itself. The question arrives from three directions and means something different each time.
- From a partner: are we losing deals because we cannot promise a founder they will be heard?
- From an LP: what is our platform actually producing, given what it costs?
- From a founder in diligence: what happens on my launch day if I take your money?
Only the third one is really about media. The first is about competitive position and the second is about cost justification, and answering all three with a content calendar is how firms end up funding a publication nobody asked for.
What the benchmark actually is
Worth being precise before deciding whether to chase it. By their own published account, Andreessen Horowitz's media operation runs roughly a million followers on X, a daily newsletter at 250,000 subscribers, and a podcast at around a million downloads a month.
People tracking the effort closely put the marketing headcount above 70 against the one or two most firms carry. They produce launch films on compressed timelines, embed people inside portfolio companies, and offer the whole thing as a service to founders.
That is a media company with a fund attached, and it took years and a balance sheet almost nobody else has.
The useful move is to notice what that scale costs its owner, because industrial production has structural weaknesses that a small team does not.
Four things industrial media cannot do
These are not criticisms of anyone's execution. They are properties of scale, and they are where a two-person team is genuinely advantaged.
- It cannot be unscalable. Anything that only works when it is done personally, badly, and for one founder is off the menu once a process has to serve a hundred companies. A platform lead who reads a founder's draft at 11pm and rewrites the hook is doing something no operating manual reproduces, and founders can feel the difference immediately.
- It cannot be weird. Institutional output converges on a house voice because it has to be approvable by people who did not write it. Genuine strangeness, a strong take that might be wrong, a piece about something nobody asked for, all get filed down. That register is left lying on the floor for anyone willing to pick it up.
- It cannot go narrow for long. A general firm serving every sector has to broadcast across all of them. A firm that only does robotics, or only does devtools, or only writes for infrastructure buyers, can own a domain completely, and depth in one vertical beats breadth across ten for the reader who lives in that vertical.
- It cannot be disinterested. This is the load-bearing one. Every piece of praise a firm publishes about a portfolio company is published by a shareholder in that company, and readers know it. Commentators who follow the biggest operation closely make exactly this point: portfolio coverage reads as an interested party talking its own book. No budget removes it, because it is a fact about ownership rather than about the writing. The ceiling, in full.
The last one is worth dwelling on if you are deciding what to build, because it means the ceiling on a firm's media arm is belief, and belief is the thing your founders actually need on launch day.
The three questions that replace a content strategy
The most useful framing we have seen on this comes from commentary in the venture press rather than from any agency, and it is three questions rather than a plan.
- Who exactly are you reaching? Not "founders". A named, countable group. Seed-stage infrastructure founders in Europe. Heads of AI at insurers. If the answer is a category rather than a group, everything downstream will be generic.
- How do those people want to receive things? Ask fifteen of them. The answer is frequently not a newsletter, and it is frequently something unglamorous like a private group or a monthly call.
- What access do you have that nobody else does? The only durable reason anyone subscribes to anything. Proprietary data from your portfolio, a room you can convene, a diligence process nobody else runs. A firm with no answer here is about to publish opinions that are freely available elsewhere.
Most firms working through these honestly land somewhere the large operations are not, which is the point. The third question is the one that gets skipped, and it is the one that decides whether the thing is worth doing at all.
What founders actually want from you
Worth separating from the strategy conversation, because it is narrower and more solvable than it looks.
A founder in diligence asking about your media capability is rarely asking you to become a publisher. They are asking one thing: when I ship, will anybody see it? That question has an answer that does not require headcount.
- Introductions that are actually warm. The most valuable thing most firms have and the most frequently overstated. A real relationship with fifteen people who can carry a launch beats a list of four hundred you have emailed once.
- A specific opinion on the angle. Founders are usually too close to their own product to see which claim is arguable. Someone senior spending an hour on that before launch week is worth more than a film. What that conversation looks like.
- Reach bought per launch rather than staffed year-round. The thing a media arm provides on launch day can be purchased on the days it is needed, at a fraction of carrying it as salary, which is the arbitrage available to every firm that is not running a publication.
- Honesty about what will not work. Telling a portfolio company that their launch has no claim in it, before they spend on distribution, is a service. It is also the thing an in-house team promoting its own portfolio finds hardest to say.
Our platform team playbook covers the operating side of this, and helping portfolio companies launch covers how the support is usually structured across a fund.
What to say to the LP
The version that survives the meeting is short and does not pretend.
We are not building a media company, because the benchmark for that is seventy people and a decade of balance sheet, and buying a smaller copy of it would produce a newsletter with no reason to exist.
What we are doing instead is narrower and cheaper: we own one audience nobody else serves properly, we have a specific and honest opinion for every portfolio company on the angle their launch rests on, and we buy reach on the days our companies ship rather than carrying it as headcount every day of the year.
That is a defensible position and it is true, which is more than can be said for most platform slides. It also has the advantage of costing roughly a hundredth of the alternative, which is usually the part the LP was asking about.
Questions we get asked
Should our firm build a media arm?
Only if you can answer three questions without flinching: who exactly are you reaching, how do those people want to receive things, and what access do you have that nobody else does. A firm that cannot answer the third one is about to fund a newsletter nobody needed. The first two decide the format, and the third is the only durable reason anybody subscribes.
What is the actual benchmark we are being compared to?
Andreessen Horowitz reports roughly a million followers on X, a daily newsletter at 250,000 subscribers and around a million monthly podcast downloads, with marketing headcount reported above 70 people. Most firms run one or two. Matching that is not a strategy, it is a budget you do not have, and the useful response is to compete where scale is a liability.
What can a two-person platform team offer founders instead?
Introductions that are actually warm, a specific and honest opinion on the launch angle, and reach bought per launch rather than staffed year-round. A founder does not need their investor to be a publisher. They need to be seen on the day they ship, and that can be purchased when it is needed instead of carried as headcount.