Podcast, YouTube, or newsletter: which one should you commit to?

Ask which of the three formats grows fastest and you will have an answer inside an hour. Every platform, every agency, and every course has one ready.

Ask which one you will still be producing in three years, and the room goes quiet.

The second question is the one that settles this. All three are multi-year commitments that pay nothing for the first several months. The format that fits is the one you can keep feeding after the novelty wears off and a busy quarter arrives.

For most expert firms, that is the newsletter. Two conditions change the answer, and they are worth knowing before you buy a microphone.

What each format is for

A newsletter is an owned list. It arrives in an inbox because someone asked for it, and no platform can revoke that relationship. It compounds slowly and quietly.

YouTube is a search engine that happens to serve video. A talk you record this spring can still be answering a buyer's question in 2029, and the payback runs on that evergreen shelf life rather than on any single upload.

A podcast is closer to a relationship instrument than an audience instrument. The guest chair is the asset, which is why interview shows tend to repay their owners in conversations rather than downloads.

Three different machines. The advice that treats them as interchangeable content channels is skipping the part that decides your answer.

Two tests settle it

The first test is how your buyers verify you. When a referred prospect checks you out, do they read, watch, or listen? A general counsel skims a memo. A surgeon watches a procedure. A founder listens on a commute. You know the answer for your own field, and it outranks every platform statistic.

The second test is what you can sustain. Not what you could do for six weeks. What survives the month you take on two new matters and lose a partner to a sabbatical.

Write down the version you would bet on. Twenty minutes of dictation on a Sunday is one answer. Two hours in front of a camera every second week is another. Then pick the format that matches the answer you gave.

The attrition numbers should decide more than they do

Podchaser profiled every podcast that launched in the first half of 2026, all 153,767 of them. By the end of June, 41.7% had already gone quiet, and fewer than 1% reached any measurable audience.

Read that as a statement about calendars rather than about talent. Shows do not usually die of bad ideas. They die when the founder's week stops accommodating them, which happens to most founders within months.

The stakes are higher than a hobby quitting, because an abandoned show is a dated artifact carrying your name. A prospect who finds twelve episodes ending in March 2026 learns something about your follow-through that you did not intend to teach.

YouTube demands the same endurance on a longer clock. Practitioners running business channels tend to describe six to twelve months before the channel compounds into anything you would call a source of inquiries.

A newsletter fails more gently. Miss a month and you send the next one. That forgiveness is worth more than it sounds when the thing you are choosing has to survive years.

Why the newsletter wins for most expert firms

Three reasons, in order of weight.

It is the only one of the three where you keep the audience. Subscribers move with you across platforms, rebrands, and business models.

It is the cheapest per unit of trust. An hour of writing produces something a buyer reads in full attention, in a place where nobody else is bidding for their eye.

And it accumulates into a record you can read. Owners who have run lists for years describe the same shape: slow growth, then a steady share of inquiries that arrive without a campaign behind them. That pattern is what a long game is supposed to look like.

Now the part the newsletter advocates leave out. You own the list; you do not own delivery. Gmail and Yahoo have required authentication, one-click unsubscribe, and disciplined complaint rates from bulk senders since February 2024, and Google asks senders to stay under 0.1% of recipients marking mail as spam, never reaching 0.3%. Microsoft applied its own rules in 2025.

For a practice writing to a few thousand people, none of that is onerous. It does mean the mailbox providers are a party to your relationship with your readers, and any promise that email is fully owned is overstated.

When it is not the newsletter

Two conditions flip the answer.

The first is a field where buyers learn by watching. Procedures, physical craft, software, anything with a screen or a pair of hands in it. Video is the medium that carries the work, and YouTube's search behavior keeps each answer earning for years.

The second is when you need access more than audience. An interview show puts you in a recurring hour with people you want to know, and the guest list is the strategy. If your growth depends on twenty relationships rather than two thousand readers, the show earns its cost.

One caution on that second case. A show run for access is judged on who says yes, not on downloads, and the two get confused constantly. Decide which one you are buying before the first episode.

What AI changed, and what it did not

Production collapsed. Editing, transcripts, show notes, clips, and captions are now minutes of machine work rather than an afternoon of yours. The cost of publishing has fallen further in two years than in the previous ten.

What did not change is the input. Someone still has to sit down with the material, and that someone is you. The irreducible hour is the recording or the writing, and no tool has taken it.

So the choice gets simpler and harder at the same time. The old excuse was production. The remaining question is whether you will show up for the input every month for three years.

Cheap production also raised the bar on the other side. Every field is now full of competent, forgettable video and competent, forgettable email. What survives is the material only you have: the case you argued, the pattern you have watched for a decade, the number you track that nobody else in your field bothers with.

How to decide this week

Take four steps, in order.

List your last ten clients and name what each of them consumes when they research a professional. That is your buyer test, answered with evidence instead of instinct.

Then run a four-week trial at the real cadence, without announcing anything. Write four issues, or record four episodes, and keep them in a drawer. Almost everyone who quits would have known by week three.

Commit to twelve months in writing, on the calendar, with the recording or writing slot blocked like a client meeting.

And choose the one you would do for nobody. The format that survives is the one where the work itself is a reason to keep going.

The three formats do not differ much in what they can achieve for an expert. They differ enormously in what they cost you every week for the next three years. Choose on the cost you will still be paying in year three, and the growth question sorts itself out.