How I Grew My Newsletter to 50,000 Subscribers in 20 Months - Harry Dry

Harry Dry | Wynter

Abstract

A marketing writer takes a conference through how a newsletter reached fifty thousand subscribers in twenty months, starting from a web-developer job and a handful of case studies written on the side. The whole method reduces to two problems and only two: people have to see the work, and people have to choose to convert. Everything else in the talk is an intervention on one of them, which is what makes it possible to tell a real tactic from a busy one.

• Before any content strategy comes a map of the places the audience already gathers.
• Adding value in someone else's venue means republishing the whole article there, not dropping a link into it.
• Email is the storage layer, because a platform's reach is granted and can be withdrawn.
• Tripling conversion may cost a day; tripling traffic costs advertising money or years.
• The returns from one great piece far outweigh the returns from twenty adequate ones.
• What reads as a designed system is the residue of attempts that survived — two or three failed businesses came first.

Everything below follows the talk in order, through the distribution half, the conversion half, and then the questions — where the format's origin, the first hundred subscribers, the money, and the working day all get answered directly. For anyone building a personal brand the useful part is how little of it is technique: the two moves that built the site would both have been refused by a brief with a deadline on it.

Chapter summaries

00:17 - Publishing case studies until a sponsor appeared was the route in

The route in was writing marketing case studies online until a sponsor appeared, then leaving the job. Publishing the work was the thing that produced both the audience and the income, in that order — a web-developer job, case studies written on the side, a sponsor, and only then the resignation. The approach is introduced from outside as running against the received method: cross-posted across a great many groups, and working, in the framing offered, because the content is good rather than because the posting is clever. Cross-posting is the visible mechanism, and quality is named as the thing that makes it work rather than the volume.

02:55 - Newsletter growth is two problems only: being seen, and being converted

Newsletter growth is two problems and only two: people have to see the work, and people have to choose to convert.

"How do you grow a newsletter? I think it can be overcomplicated. Essentially you need two things."

— 02:55

Any growth tactic is an intervention on one of the two, so separating them stops effort landing in the wrong place. It is the same compression that runs through every version of this talk, and the value of it is diagnostic rather than motivational: a tactic that moves neither number is not a tactic.

03:43 - Before any content strategy comes a map of the places the audience already gathers

Before content strategy comes a map of the places the audience already gathers — his own term for them is watering holes.

"How do you get people to see your work? When I started marketing I didn't really know."

— 03:43

Attention already exists in those places, and building a destination requires manufacturing it from nothing. The map comes before the content strategy rather than after it, because the writing is shaped by where it has to land. His own term for them is watering holes, and the list is written down rather than remembered.

04:38 - Adding value in someone else's venue is not dropping a link into it

Adding value in someone else's venue is not dropping a link into it; busy people have to be impressed inside the place they are already standing.

"Adding value isn't just dropping a link on twitter. Nobody's gonna click that."

— 04:38

A link asks for a click before any value has been shown, and busy people have to be impressed inside the place they are already standing. The failure is not that the link is impolite — it is that it arrives in the wrong order, asking for the action before giving the reason. The value goes where the attention already is.

04:55 - Full republication beats teasing, because the reader gets the whole thing without leaving

The move is full republication rather than teasing: the entire article goes into each venue, with the subscribe ask at the end of every copy.

"What I did is I'd write these articles and I would write them out on my website."

— 04:55

The reader gets the whole thing without leaving, and the ask at the end is earned by what came before it rather than promised by what comes after. Teasing keeps the value on your own property and asks the reader to travel to it; full republication moves the value to them and asks for nothing until it has been delivered.

05:39 - The content stays constant and the wrapper changes for each platform's native form

The same content is tailored to each platform's native form rather than pasted across them.

"But what about if I did it on instagram and linkedin as well? Again, exactly the same principle."

— 05:39

Each venue carries a different format well, so the content is constant and the wrapper changes. A slideshow on one network, a carousel on another, a thread on a third — the same article, three times, in three shapes. The work is in the adaptation rather than in the writing, and it is the part most people skip.

06:03 - An image is the one unit every platform is suited to carrying

Where a platform cannot carry a full article, an image can — images are the one unit every platform is suited to.

"Facebook groups is a little bit harder, because you can't really share a full article."

— 06:03

The format constraint rather than the audience is what blocks distribution, and images route around it — every platform is suited to carrying one. Where a full article cannot be posted, the diagram or the single tip can, and it travels intact rather than as a summary of something else.

06:24 - A capability claim leaves the translation to the reader, a benefit line does it for them

The tip he uses as the specimen is the benefit swap: a line stating what a product is, replaced by a line stating what it does for the person holding it.

"Instead of saying the world's first digital media player: every song you've ever owned in your pocket. No one cares."

— 06:24

A capability claim asks the reader to do the translation from what the thing is into what it does for them, and most will not bother. The specimen is the oldest example in the field: the same product described by its specification and then by the experience of owning it. One sentence needs a translator and the other does not.

06:45 - The system is three stages: create value in the venues, transfer it, store it in email

The whole system is three stages: create value in the venues, transfer it to the site, store it in email subscribers.

"Six months later, with a bit of experience, I realised this diagram explains all of it. You create value."

— 06:45

Each stage has one job, and conflating them is what makes growth look complicated. Value is created in the venues, transferred to the site, and stored in email subscribers. The diagram is the whole method, and its usefulness is that it says where any given tactic belongs — which is also how it exposes tactics that belong nowhere.

07:00 - Email is the storage layer, because a platform's reach is granted and can be withdrawn

Email is the storage layer because a one-time visitor forgets, and the subscriber is a direct line that a platform cannot revoke.

"And then your email subscribers is where you store the value."

— 07:00

A platform's reach is granted and can be withdrawn by an algorithm change; a subscriber is a direct line that cannot be revoked by anyone but the subscriber. A one-time visitor forgets the site exists, which is not indifference but arithmetic: nothing in a visit obliges a return. Email is the layer that survives the platform it was built on.

07:49 - Membership of a community comes before any promotion inside it

The best self-promoters are not really self-promoters: membership of the community comes before promotion inside it.

"We talked a lot about self-promotion, and I like to say that I'm definitely a self-promoter."

— 07:49

A member's post is read as participation and a non-member's as an advert, and nothing in the post itself distinguishes them — the difference is the time already spent. He grants the label openly rather than denying it, which is what makes the distinction usable: the question is not whether you are promoting, it is whether you have earned the right to.

08:16 - The promotion move is ordered: the answer first, other resources next, the writer's own last

The promotion move is ordered: answer the question that was actually asked, link other people's resources, and mention your own thing last and as a consequence of your own interest.

"Someone writes, I want to get better at sales. So I reply, hey, I was in your shoes."

— 08:16

The answer earns the attention, the other links establish that the answer is not a pretext, and the mention of your own thing arrives as a consequence of your own interest rather than as the purpose of the reply. The order is the whole mechanism. Reversed, the same three elements read as an advertisement with a helpful preamble.

08:47 - Content promotion is a positive-sum conversation rather than a push

Content promotion is a positive-sum game and a conversation, not a push.

"I like to say that content promotion is a positive sum game. If you help other people."

— 08:47

Helping people produces reciprocal help, and forcing material at people is not how the mechanism runs. It is a conversation rather than a push, which sounds soft and is in fact the operating constraint: the exchange has no loser, so there is no reason to ration it, and reciprocation accrues rather than trades.

09:49 - Tripling conversion may cost a day, and tripling traffic costs money or years

Tripling conversion may cost a day; tripling traffic costs advertising money or years of content marketing. Both routes reach the same number, and only one of them is bounded by a budget or a calendar. Ten thousand visits a week converting at one per cent, against the same traffic at three — the second is a day's work on the page, the first is advertising money or years of content marketing. The cheaper multiplier is the one you already control.

10:24 - The generic subscribe box has no purpose behind it, so an update promises nothing

The first half of conversion is that the thing must be worth signing up for — the generic subscribe box has no purpose behind it.

"So conversion, there's two important things. Firstly, you've got to make something people want to sign up for."

— 10:24

A reader is deciding what they get, and an update has no content in it — so the generic box asks for a commitment against an unstated return. The purpose has to exist before the wording can carry it, which is why this half of conversion is not a copywriting problem.

11:06 - Stated effort, a named subject, a described format and a time cost supply the purpose

The rewrite that supplies purpose is built from stated effort, a named subject, a described format and a time cost.

"So let's rewrite it. Each week I spend 50 hours analysing a new startup trend. I share what I learned."

— 11:06

Each element is checkable and each converts a vague offer into a specific transaction: the hours behind it, the subject, the shape it arrives in, and what reading it will cost. Nothing in it is persuasion — it is a description precise enough that the reader can decide without being persuaded.

11:37 - Capture language holds the operator's goal, and the reader is answering a different question

The framing flip: stop thinking about capturing emails and start thinking about creating something worth subscribing to.

"The trick here is don't think about capturing people's emails. That's what people say, I gotta capture their emails."

— 11:37

Capture language puts the operator's goal in the operator's head, and the reader is answering an entirely different question. The flip is not a euphemism: thinking about capture produces tactics aimed at the form, and thinking about what is worth subscribing to produces work aimed at the reader. Only one of the two can succeed on its own terms.

11:57 - A named person, a cadence and a readership count tripled the conversion rate

His own box rewritten to a named person, a cadence, a described format, a readership count and visible social proof — reported as taking conversion from 1% to 3% overnight.

"And here's mine. So back in the day my subscribers, like the one on the left, it was: why not join."

— 11:57

The count and the testimonials answer the second question a reader has — is it any good — in the same breath as the first. The old line was an invitation with nothing attached; the new one names a person, a cadence, a format and a readership. The reported result is a conversion rate moving from one per cent to three overnight, which is the multiplier that does not need traffic.

12:32 - Every point on the site sits one click from signing up, because each step can lapse

The second half of conversion is ease, and the standard it produces is that every point on the site is one click from signing up.

"The second thing, as well as making something that people want to subscribe to, is make it easy for them."

— 12:32

Every additional step is a place the decision can lapse, so the standard is that no point on the site is more than one click from signing up. Four entry points rather than one, on the principle that the moment of deciding is not always in the same place — and the cheapest thing you can do for a reader who has already decided is not to make them look.

13:34 - An institution asking gets an institution's response, so the ask is written as a person

The ask should be written in human language — a named person describing what they actually do — rather than in the institutional formula.

"One of them, which is so obvious, is just asking people to subscribe in human language."

— 13:34

An institution asking gets an institution's response; a person asking gets a person's. The formula version is a sentence nobody wrote and nobody reads. The human version names who is writing, says what they actually do, and asks — which is the same content and a different transaction.

14:08 - Content is the multiplier on both halves, because the same input moves reach and conversion

Content is the multiplier on both halves at once: good work gets shared, which is reach, and good work converts better, which is conversion.

"I'll be lying if I said that content wasn't really the key behind all this. So good content."

— 14:08

The same input moves both variables at once, which is why content is not a third thing alongside reach and conversion. Good work gets shared, which is reach; good work converts better, which is conversion. Every other tactic in the talk moves one number, and this one moves both, which is the argument for spending the calendar here.

14:42 - The returns from one great piece far outweigh the returns from twenty adequate ones

The returns from one great piece far outweigh the returns from twenty that are merely okay.

"One thing I realised is that the returns you get from one great piece far outweigh."

— 14:42

Great pieces are the ones that get shared and cited; adequate ones return nothing and consume the same calendar. He started on one a day and stopped, which is the evidence rather than the theory — the volume experiment was run and lost. The ceiling is set by the best piece, not the average one.

15:09 - Commissioned work with a deadline and a keyword list cannot be the work that compounds

Commissioned content with a deadline and a keyword list cannot be the work that compounds, because the two moves that built the site would both have been refused.

"Gotta change your mindset slightly, because when managers might say, hey, we've got to get this."

— 15:09

The brief supplies neither the time nor the permission, so commissioned work with a deadline and a keyword list cannot be the work that compounds. Both of the moves that built the site — the hours per diagram, and the format nobody had asked for — would have been refused by that process. It is a structural claim about who can do this work rather than a complaint about clients.

16:11 - Each channel's ending is the next channel's beginning, so the set becomes a web

Loops between channels: each channel's ending is the next channel's beginning, and the set becomes a web rather than a funnel.

"So last thing is creating loops between your pieces of content."

— 16:11

A channel already holding attention is the cheapest possible launch audience for the next one, so each channel's ending becomes the next channel's beginning. The set stops being a funnel with an exit and becomes a web in which nobody is lost at the end — they are handed on.

17:44 - The format itself is the differentiator where the field competes on one standard axis

The format itself is the differentiator: the field runs on keyword volume from a research tool, and this is a self-made format.

"One thing about your approach is that nobody else is doing it."

— 17:44

Competing inside the standard toolset means competing on the standard axis, and the field runs on keyword volume from a research tool. A self-made format is not better on that axis; it is off it entirely, which is why the comparison the field knows how to make does not apply to it. The unoccupied axis is the cheap one.

18:18 - Inexperience left the field's central assumption untested, which was the angle

Inexperience was the angle: not knowing the received method left its central assumption — that the job is getting people to a website — untested.

"It's an interesting question. I think I had a benefit because I had no experience."

— 18:18

An assumption you were never taught is one you can notice, and the field's central assumption is that the job is getting people to a website. Not knowing it left it untested — which is a claim about the value of arriving late and uncredentialed, and it is hedged as a benefit he had rather than a strategy he chose.

19:03 - What reads as a designed system is the residue of attempts that survived

There is no secret formula: the method was trial and error and a long stretch of fumbling around not working.

"A lot of trial and error. It was just a lot of fumbling around, not working, for quite a long time."

— 19:03

What reads as a designed system is the residue of attempts that survived. The account he gives of it is trial and error, and a long stretch of fumbling around not working — which is worth setting against the diagram, because the diagram came six months in and describes what had already happened rather than what was planned.

19:46 - An engaged list is the asset, and sponsorship prices on traffic instead

What has been built is an engaged list and a following, not revenue; sponsorship is scale-gated and his own is modest.

"I've made very little money right now. I think that sponsorship is really scale-gated."

— 19:46

Sponsorship prices on traffic, which the list does not yet have, so the asset and the revenue are on different clocks. What exists is an engaged list and a following rather than income, and he says so plainly. The frame he borrows for it is a choice between two postures, taken from a rugby player: control, or explore. Choosing to explore removes the obligation to have a plan that survives contact, which is the honest description of where this is.

21:25 - The apprenticeship was two or three failed businesses before the visible start

The apprenticeship was two or three failed businesses; the thing that worked launched on one or two articles and a badly built site.

"I made about two or three businesses which failed, for months and months and months. That was where I learned."

— 21:25

Failure is where the learning happened, so the visible start is later than the real one. Two or three businesses over months, then a thing that launched on one or two articles and a badly built site — the asymmetry between the two is the point, because the second only looks fast if the first is not counted.

21:47 - The first hundred subscribers came from one share, and proximity was the mechanism under it

The first hundred subscribers came from a single share by one well-followed person, and the mechanism under the luck was proximity — being in the same groups as those people. One node with a large following can supply a whole first audience, and the mechanism under the apparent luck was proximity — being in the same groups as the people who might share it. It is not a plan you can execute directly, but it is a position you can occupy.

"I 100 agree. Find people your own size, growth together. That's how I built so."

— 22:21

Peers at the same stage will actually reciprocate; established nodes have no reason to.

22:47 - Quantity buys feedback while there is none, and the definitive piece is the goal after

Quantity is right while you are still learning what works; once there is an audience, the goal is the definitive piece on the topic.

"In the early days I was trying to do one a day, and I quickly realised that the returns from that."

— 22:47

Quantity buys feedback when you have none, and once feedback exists the ceiling moves. The goal after that is the definitive piece on the topic rather than another one — which is a different job, not a harder version of the same one. One a day was the right answer to a question he no longer had.

23:24 - Two bad releases and nobody opens the third, so the last one sets the standard

Treat yourself like a musician: two bad releases and nobody listens again, so the standard is the best thing on the topic and the reputation follows.

"If you treat yourself like a musician and you release two bad albums, your career is done."

— 23:24

The audience's decision to open the next piece is set by the last one, so two bad releases and nobody listens again. The standard that follows from it is not ambition but arithmetic: the best thing on the topic is what earns the next open, and the reputation is the accumulated record of those decisions rather than a thing you build directly.

23:50 - Aiming at the best article in the world raises the floor even when missed

The same standard stated independently from a decade earlier: aim to write the best article in the world on whatever topic you are on.

"My goal was to write the best article in the world on whatever topic I was on."

— 23:50

Aiming at the ceiling raises the floor even where the aim is missed, and the same standard is stated independently from a decade earlier by someone running a different site. Two people arriving at the same rule separately is weak evidence, but it is evidence — and the rule costs nothing to adopt.

24:31 - A format proven in a neighbouring field is validated and still unoccupied in yours

The format was imported from an adjacent field and then seeded from the channels he already occupied.

"Basically this designer, he started taking design, which was written in blog posts."

— 24:31

A format proven in a neighbouring field is validated but unoccupied in yours, so the risk has already been taken by somebody else. The second half is that an existing channel supplies the first audience for the new one: the format was imported, and then seeded from the places he already held rather than launched cold. Familiar in one field and new in another is the whole opportunity.

25:19 - Paid budget goes behind the single best artefact, because the work is already proven

Put paid budget behind the single best artefact and target the audiences of a handful of people whose audience you want.

"I think don't be afraid to spend ad money, if you're putting the work in."

— 25:19

The work is already done, so the spend buys distribution for a piece that has already proved it holds attention — which is the opposite of advertising something untested. The targeting he describes is the audiences of a handful of people whose audience you want, which is narrow enough to be affordable and specific enough to be worth it.

25:57 - Reposting old work is legitimate, because most of a grown audience never saw it

Reposting old work is legitimate, because most of a grown audience never saw it.

"Don't be afraid to post stuff twice. It's a bit of a boring tip, but most people, let's say."

— 25:57

An audience that has grown several times over is mostly new, so the old piece is new to them. He gives his own account roughly doubling between the first posting and the second, which makes the repost a first exposure for most of the people who see it rather than a repetition. It is the cheapest content decision available and it is mostly declined out of embarrassment.

26:38 - The audience is defined by appetite rather than by demographics

Appetite rather than demographics defines the audience: people who want to get their hands dirty and want the practical thing rather than the strategy conversation.

"My subscriber is people who want to get their hands dirty. It's marketers."

— 26:38

An appetite predicts what a person will actually do with a piece, and a demographic does not. The audience is the people who want to get their hands dirty and want the practical thing rather than the strategy conversation — which is a description of behaviour rather than of identity, and therefore something the work can be aimed at.

27:57 - Input redirects the day onto other people's priorities, so creative hours come first

There is no system for balancing community time against creation — the working rule is creative hours before input.

"How do you balance that? I haven't worked it out. I haven't worked it out. I'm trying my best still."

— 27:57

Input redirects the day onto other people's priorities before any of your own work exists, so the working rule is creative hours before input. There is no system underneath it, and he says so twice. He does all of it himself, with nobody helping out. The unstaffed function is the community and communication layer — the part his own disposition does not run, which is why it would be the first hire.

30:07 - Nobody has the process, so waiting to be given one is what stops the work

The closing advice is a disclaimer: there are no adults, everybody is making it up, so work it out yourself.

"I'm gonna give you a quote, because I haven't got the words."

— 30:07

If nobody has the process, waiting to be given one is the thing that stops the work. The closing advice is a disclaimer rather than a method: there are no adults, everybody is making it up, so work it out yourself. It is consistent with everything above it — the diagram arrived six months late, the format was borrowed, and the first hundred subscribers were one share from one person.

Personal Branding Lessons

The talk is about a newsletter, but the two variables underneath it — being seen and being converted — are the same two that govern any body of published work. Each move below acts on one of them.

Map the watering holes before writing the content strategy

Write down the specific places your readers already gather before deciding what to publish. Attention already exists in those places; a destination of your own requires manufacturing it from nothing. The map comes first because the writing is shaped by where it has to land — the format, the length and the opening all change depending on the venue. A list you can read back is the difference between a route and a habit. 03:43

Republish the whole article in each venue rather than teasing it

Do not drop a link. Put the entire piece into the venue, in that venue's own native shape, and put the subscribe ask at the end of every copy. A link asks for a click before any value has been shown; the full article delivers the value where the reader already is, so the ask arrives as a request for more of something they have just had. Where a platform cannot carry a whole article, an image can — it is the one unit every platform is suited to. 04:55

Answer the question first and mention your own thing last

When somebody in a community states a problem, answer the question that was actually asked. Then link other people's resources. Then, last, mention your own thing as a consequence of your own interest in the subject. The order is the entire mechanism: the answer earns the attention, the other links prove the answer was not a pretext, and the final mention reads as disclosure rather than as the purpose of the reply. Reversed, the same three elements are an advert. 08:16

Work the conversion rate before buying the traffic

Ten thousand visits a week converting at one per cent, against the same traffic at three: both routes reach the same number, and only one of them is bounded by a budget or a calendar. Tripling conversion may cost a day of work on the page. Tripling traffic costs advertising money or years of content marketing. Start with the number you already control, because it keeps paying after you stop working on it. 09:49

Replace the subscribe box with a named person and a described format

A generic box asks for a commitment against an unstated return, because an update has no content in it. The replacement is built from checkable parts: the hours behind each issue, the named subject, the shape it arrives in, and what it will cost to read. Add a real name, a cadence and a readership count and the second question — is it any good — gets answered in the same breath as the first. The reported move was one per cent to three. 11:06

Keep every point on the site one click from signing up

Every additional step is a place the decision can lapse, and the moment somebody decides is not always in the same place. Four entry points rather than one — a fixed bar, an end-of-article prompt, a pop-up, a dedicated page — so that a reader who has already decided never has to go looking. Write the ask in human language too: a named person saying what they actually do, rather than the institutional formula. An institution asking gets an institution's response. 12:32

Spend the calendar on one great piece rather than twenty adequate ones

The returns from one great piece far outweigh the returns from twenty that are merely okay, and the experiment has already been run — one a day was tried and abandoned. Great pieces get shared and cited, which is reach, and they convert better, which is conversion, so the same input moves both variables at once. Quantity is the right answer only while you have no feedback; once an audience exists, the goal is the definitive piece on the topic. 14:42

Import a format from an adjacent field and seed it from a channel you hold

A format proven in a neighbouring field is already validated and still unoccupied in yours, so somebody else has taken the risk and nobody has taken the position. The second half matters as much: launch it from a channel you already occupy rather than cold, because an audience already paying attention is the cheapest possible first audience for a new thing. The field here competes on keyword volume from one research tool — a self-made format is not better on that axis, it is off it. 24:31

Protect the creative hours before any input

Input redirects the day onto other people's priorities before any of your own work exists, so the rule is creative hours first and everything else after. There is no system underneath this, stated twice and without embarrassment: the balance between community time and making things is unsolved, and the first hire would be the community layer precisely because it is the part the operator's own disposition does not run. Name the function you avoid; it is the one that will stay undone. 27:57

Questions

Each answer ends at the moment in the recording where it is given.

How do I get people to see my work when nobody visits my website?

Take the work to the places your audience already gathers instead of trying to pull them to a site with no traffic. Map those venues first — groups, forums, professional feeds, chat rooms — then put the finished article inside each of them rather than a link to it. Attention in those rooms already exists; a destination of your own has to manufacture it from nothing, which is the slower and more expensive of the two jobs. 03:43

Should I share a link to my article or post the whole thing?

Post the whole thing. A link asks for a click before any value has been shown, so most people never take it; the full piece delivers the value inside the room, and the subscribe ask then sits at the end of something the reader has just finished. Adapt the wrapper to each venue — a thread here, a full submission there, a slideshow where images travel better — and keep the content itself constant. 04:55

Is it better to get more traffic or to improve my signup rate?

Improve the signup rate first, because the two routes to the same number are priced completely differently. Taking a site from one per cent to three per cent is described as roughly a day of work where nobody has optimised the page before; taking the traffic to three times its size costs an advertising budget or a couple of years of content marketing. The condition matters: that day of work is only available on a site nobody has looked at properly. 09:49

Why is nobody subscribing to my newsletter?

Most likely because there is nothing behind the box. Subscribe for updates names no subject, no format and no effort, so a reader has nothing to weigh — and an update is not a thing anyone wants. The fix is to make something worth subscribing to rather than to optimise the capture of emails. Reframing the same subscription as a specific, priceable offer is reported to have moved one signup page from one per cent to three per cent. 10:24

What should my newsletter signup box actually say?

Four things a stranger can check: the effort you put in, the subject you analyse, the format it arrives in, and how long it takes to read. One worked example states the hours spent each week, what is analysed, and a five-minute email. Add the number of people already reading and a few of their own words, because a count and visible testimonials shift the decision from your judgement to other readers' behaviour. 11:06

How many places on my site should ask people to subscribe?

Enough that no point on the site is more than one click from signing up. The worked example runs four: a fixed navigation bar, the end of every article, a pop-up after about sixty seconds, and a dedicated subscribe page. The count is not the target — the distance is. Every extra step between deciding and signing up is somewhere the decision can quietly lapse, and the repair is small tweaks rather than a rebuild. 12:32

Is it better to publish one great piece or twenty decent ones?

One great piece, once you have an audience. The returns from a single excellent article are described as far outweighing twenty that are merely okay. Great work also moves both growth variables at once: people share it, which is reach, and a reader who finishes it is likelier to sign up. Quantity is still correct earlier, while volume is what teaches you what works. 14:42

How do I promote my work in a community without being the person everyone mutes?

Join properly first, then invert the order of the reply. Answer the question that was actually asked, link other people's resources, and mention your own thing last, framed as a consequence of your own interest. A member's post reads as participation and a stranger's reads as an advert, which is why standing earned inside the room does the work no amount of posting from outside it can. 07:49

Where do I find a content format that nobody in my field is using yet?

Look one discipline over. The tip-image format here was watched working in design, then converted to marketing — validated somewhere else and unoccupied where it was taken. That is the unclaimed ground everyone benchmarking against their own field walks past, because competing inside the standard toolset means competing on the axis every rival is already optimising. Seed the new account from a channel you already have. 24:31

How do I define who my content is for?

By appetite rather than by demographics. The subscriber described here is somebody who wants to get their hands dirty and wants the practical thing — numbers, examples, more examples — rather than the strategy conversation with nothing behind it. An appetite predicts what a person will open; a job title does not, which is how one definition ends up covering marketers and founders without splitting into two. 26:38

Sources

How I Grew My Newsletter to 50,000 Subscribers in 20 Months - Harry Dry

The sections above follow the episode's own order. Each timestamp is the point in the recording where the idea is discussed, so the post reads as a map of the conversation rather than a rearrangement of it.

Quotations come from the episode transcript with spoken filler ("okay", "right", "you know", "um", "actually", "kind of", "sort of") and false starts removed. No words are added: square brackets mark anything inserted for sense, and an ellipsis marks any cut. British spellings are restored, and speech-to-text errors were repaired against context only where the intended word is unambiguous. Where a line appears both in a cold-open teaser and in its place in the conversation, the timestamp cites the second — the point at which it is actually said.

Quotes are stamped with the moment they are said rather than with a speaker's name.