That is what I told myself after 6 months of writing my first book full-time.

I was full of self-doubt.

Why was I spending so much time writing this book? Would anyone even read it?

I wasn’t delusional. I knew the odds. 98% of books never sell more than 5,000 copies.

And I didn’t have a publisher, a publicist, or an audience.

That didn’t matter.

My book, The Self-Taught Programmer, ended up selling over 200K copies in eight languages.

Most people think you need an existing audience to sell your work.

That you need millions of social media followers and a big publisher backing you.

The truth is, you don’t.

Having a publisher can hold you back because they force you to follow their playbook rather than creating your own.

I beat the odds by doing things my own way.

I didn’t rely on a big launch.

I focused on building something that sells every month for years.

First, you build a foundation of reviews. Without some initial social proof, nothing else works.

Then you layer in tactics that drive consistent sales every single month instead of betting everything on a big launch week (that could work, but probably won’t).

Instead, you focus on making progress every month.

And letting everything compound.

If you don’t have an existing audience, a publisher, or a big launch budget, this is the strategy for you.

The Problem with Traditional Advice

Publishers will tell you that selling your book is all about “the launch.”

I disagree.

When I write a new book, my goal is to create a perennial seller through years of effort, not to sell 10K copies in one week with some perfectly executed launch event.

That can work if you already have an audience. But most first-time authors don’t have one.

You are playing a multi-year game.

Act like it.

A Quick Win You Can Do Today

Here is something you can do to start making progress today.

Give your book away for free to friends and family and ask them to write an honest review if they like it.

Don’t try to sell it at first.

I asked friends, family, and people I hadn’t talked to in years if they wanted a free copy of my book.

Reviews are the only thing that matters in the beginning. Nobody buys a book with 3 reviews — but once you hit 25, something shifts.

Amazon’s algorithm starts noticing you.

Readers start trusting you.

And then you start to build momentum.

Next, enroll in Amazon KDP for the Kindle copy of your book and schedule a day to give it away for free.

Use Kindle giveaway sites like manybooks.net to drive traffic to your promotion.

They offer packages that guarantee 500 downloads for $39.

I used sites like these to drive thousands of free downloads for my book.

More people reading my book led to more reviews, which in turn led to more sales.

Find One Place to Grow Organically

While writing The Self-Taught Programmer, I tried all the major social media platforms.

I had a hard time growing on most of them.

Then I created a Facebook group called Self-Taught Programmers, which eventually grew to 365K members through word of mouth.

I did not anticipate that it would happen, and it started small. It took 5+ years to reach 100,000 members. The group started with just me.

Remember, slow, consistent growth compounds over time.

At first, I had to hustle to get people to join.

But once people saw that I would answer their questions about learning to program, more people joined, and the growth continued.

You need to find one platform where you can grow organically.

That means you need to experiment with many different platforms to find where you can build an audience, give value, and occasionally ask them to buy your book.

It is always easier to grow on a platform when it’s new.

Starting a Facebook group is way harder today than it was then. Because as platforms mature, they throttle organic growth.

But there are plenty of platforms today where organic growth is possible, like Substack, which is where I am focusing on growing to promote my next book.

The important thing is you need to find a strategy and a platform where your audience is growing every month.

Use Giveaways

Every month or two, I also did giveaways on Instagram.

People love giveaways.

It’s one of the only things on social media almost guaranteed to get engagement.

When I was promoting my book, I did a ton of giveaways on Instagram.

One of them got more than 30,000 likes.

You don’t need to give away crazy prizes to succeed with this.

The giveaway I did was literally offering a free copy of my book.

I’ve done successful giveaways with prizes ranging from my book to my book and a computer.

I recommend starting with your book and two other books similar to yours but better known. You can experiment with including a $50 or $100 cash prize or gift card as well.

The person who won the free copy of your book will most likely write you a positive review if they like your work, and everyone who didn’t win gets exposed to your work, and oftentimes, some percentage will buy a copy.

Get Press

I did not go for press right away.

I waited for nine months before I tried to get any press.

I ended up getting featured in Forbes by hiring a cheap publicist on Upwork.com.

You can also act as your own publicist. All you have to do is send emails to reporters, which isn’t as difficult as you might think.

The article generated a hundred sales.

Not the biggest spike in the world, but when your goal is slow and steady growth, you aren’t focused on huge spikes. Your goal is to move the needle slowly month after month.

Getting featured in the press helps with your book’s credibility.

You can use it elsewhere in your marketing, “as seen in Forbes.”

You can put it on your Amazon page, website, etc.

And if you act as your own press agent, all this credibility and growth is free.

You can also use a service like Help a Reporter Out to get quoted in different publications as an expert.

The platform lets journalists send out quote requests, and you can respond with your quote. If it’s a good match, the journalist may put it in their article.

I did this to get quoted in a few programming articles.

I was even quoted in CNBC.

Give Talks and Go On Podcasts

Eventually, as you continue this strategy, opportunities start coming to you.

It wasn’t part of my initial plan, but because of all the work I was putting in, I ended up getting opportunities like giving the opening keynote Speech at PyCon JP in 2019.

I also got invited to go on the Tom Style Show on YouTube, which ended up getting 681K views.

You don’t have to sit around waiting for opportunities like these to come to you, though.

Every major conference allows you to submit a request for proposal to give a talk.

And you can cold email podcasts with your pitch to get on them.

Boost What’s Working with Cheap Traffic

Once your book is selling consistently, reinvest some of your earnings into paid growth.

When I was promoting The Self-Taught Programmer, I would pay Instagram meme accounts each month to post about my book.

Influencers cost a lot of money.

But meme accounts are cheap.

I would literally spend $40 to have a programming meme account with a million followers shout out my book.

The ROI was insane.

Your Turn

Most people overestimate what they can do in one year and underestimate what they can do in ten.

You do not need to pray for your book to go viral with a perfectly executed book launch.

Instead, focus on making incremental progress year after year over a long-term horizon.

Ask your friends, family, and everyone you’ve ever met who may be interested in your book if they would like a free copy.

Build a social media following one day at a time, ideally on a new platform where you can grow organically.

Focus on leading with value, but don’t forget to also ask people to buy your book.

Run giveaways. Pitch reporters. Pay meme accounts $40 to reach a million followers. Submit talks to conferences. Cold email podcast hosts.

Everyone wants to rely on viral growth. But viral growth is luck. This is a system. I know which one I’d bet on.

Selling tens or hundreds of thousands of copies of your book over the years may not seem as glamorous, but it is a path forward that will work if you show up every day.

Every great book starts with zero sales.

Mine did too. I spent months asking everyone I knew if they wanted a free copy.

It didn’t feel like a strategy.

It felt like begging.

But it got me to 25 reviews and eventually over 2,000.

It was not glamorous. Building an audience never is. That is what nobody tells you.

But when you are sitting on hundreds of thousands of copies sold, you won’t care how unglamorous it felt at the beginning.

I’m going through the whole process again with my new book, The Self-Taught Marketer. Subscribe, and I’ll show you exactly what works this time.

Subscribe now

I had seven days left to get 1,000 new leads for my business by spending $1,000 or less on Facebook and Instagram ads.

That’s the challenge I had publicly set for myself, and if I was going to hit it, I needed images for my ads.

“I literally do not understand why you get the text wrong every single time,” I typed to ChatGPT after the 20th time it got the image I told it to make for my ad wrong.

I don’t have design skills.

And I didn’t have time to learn.

But ChatGPT kept telling me to do it in Canva.

“I DON’T KNOW HOW TO USE CANVA.”

Forget it.

I decided to test images with just a little bit of AI gibberish where there was supposed to be normal text.

It was close enough.

The Experiment

To get 1,000 new leads for my business for $1,000 or less by running Facebook and Instagram ads, I needed to spend $1 or less per new potential customer.

For the record, the average cost for a new lead on Instagram and Facebook is around $27.66.

Fortunately, I wasn’t starting from scratch.

In last week’s experiment, I wrote an ad that generated new leads for $1.57.

So I only needed to shave 57 cents per lead off the ad I was already running to hit my goal.

The ad I created last week tells the story of my ex-wife telling me not to go to the airport because the paparazzi would be looking for me.

Once I got their attention, I offered my book for free in exchange for their email address.

That’s called a lead magnet.

The idea is simple: give something valuable first, earn their trust, then sell them something later.

But before you earn their trust, you need to get their attention.

Good ads on social media are all about making someone stop scrolling by capturing their attention with something interesting.

My ad from last week did that by telling my paparazzi story.

But you don’t need a story as dramatic as mine to write a good ad: you just have to tell an interesting story instead of saying something generic about your business.

I will dive deeper into writing ads that convert in my upcoming experiments.

But for now, back to trying to get new customers.

I already had an ad.

I just needed to make it better.

To hit my goal, I decided to test new images, text, and any other changes I could think of to my ad to shave off those 57 cents.

If you are a self-taught marketer like me (the person responsible for marketing a product or service without a marketing team), and you haven’t tried running ads yet, you should consider giving it a try.

You can start with an ad that offers a lead magnet like I’m doing here.

A coach can offer a free consultation.

A contractor can offer a free estimate or a home inspection checklist.

A store owner can offer a discount or a free sample.

Consultants can offer a free chapter, a digital copy of their book, or a guide to solving a problem their clients commonly face.

Whatever you’re selling, leading with value before asking for money is almost always the right move.

The Results

On the first day of the experiment, I tested new images, text, and targeted different audiences to lower my cost per lead.

In the ad I created last week, I did not do any targeting.

That means I did not tell Meta (the company that owns Instagram and Facebook) which customers would be interested in my offer.

For example, I could have told Meta to only show my ad to people in California who are interested in entrepreneurship.

Instead, I allowed Meta’s magic algorithm to figure it out.

However, sometimes you can get better results by targeting, sometimes you can’t.

That is why testing is important.

So I tried targeting people interested in books, self-improvement, and entrepreneurship.

The result? My cost per lead went from $1.56 to $5.

Not good.

It turns out Meta’s algorithm is pretty good at what it does.

Next, I tried using the same text from the original ad with new images (without any targeting).

No luck there either.

Day two I knew I needed to try something completely different, so I decided to use the same premise as my high-performing ad but tell the story in video form.

I tried multiple variations of the video ad.

They performed a little bit better than the previous tests, but not by much.

I went back to trying even more new pictures in my ads.

I tried shortening the story in the original ad.

But nothing helped lower my cost per lead.

I started digging into the data to understand what was happening.

Last week’s ad was costing me around a dollar per website visit, but converting at 40% — meaning 4 out of every 10 visitors were signing up to get my book.

That’s incredibly high.

The reason: the book image filtered for intent.

Only people who actually wanted a free book clicked.

The video ad got cheaper clicks, but those people were just curious. They weren’t signing up.

That’s when it dawned on me: I couldn’t lower the cost per click to my website without also tanking my conversion rate.

$1.55 was my floor.

What I Learned

Even though I was unable to beat my previous ad with this experiment, I still learned a few useful things.

I now have a benchmark cost per lead to compare against all the other marketing channels I will test in upcoming experiments.

For example, I plan to pay influencers to post the same ad I ran on Meta. Now I have a number to beat.

This experiment was also a great reminder that curious clicks don’t drive real results.

When you create ads, clicks are worth nothing unless your ad attracts the right potential customers for your business.

That means engagement for the sake of engagement is a waste of time.

You need to engage people and filter for intent at the same time.

The simplest way to do that is to make your ad so specific about what you're offering that only the right people bother to click.

Furthermore, this experiment shows that sometimes spending a lot of time on optimization is not worth it.

I had an idea for an ad, wrote the hook, and the first version worked better than any others I tried.

Optimizing it further did not get me better results.

Sometimes, you need to be happy with your results and move on to other areas where you can improve your marketing.

What’s Next

This week, I was unable to get 1,000 leads for $1,000 by lowering the cost per click on my ads.

That does not mean I am giving up on my quest to get 1,000 leads for $1,000.

Next week, I’m shifting focus to the other side of the equation: the website where people are giving me their email address.

If Meta’s algorithm won’t give me cheaper traffic, I will have to make the traffic I already have convert better.

I’m running an experiment on read.selftaughtmarketer.com (the webpage I’m sending people from my ads to) to find out if I can get more people to sign up for less.

I’ll also dive deeper into the system I’m building to generate new leads and turn them into customers, so you can see how everything works and how I plan to make money from those leads.

As always, I will be sharing real numbers and real results.

Scoreboard

The Self-Taught Marketer Copies Sold (not out yet): 0

Sales Goal: 200K+

Substack Subscribers after my last experiment: 153

Substack subscribers this week: 267 (+114 since last experiment)

Substack Goal (pre-book launch): 5,000

Substack Goal (post-book launch): 50,000

Consulting Revenue (coming soon): 0

If you enjoyed this post, please hit the like button. It helps more people discover my experiments.

In my first experiment, I posted my book’s hook on social media to start building an audience of early supporters.

In this week’s experiment, I spent $157 on Facebook and Instagram ads using that same hook to see if I could acquire more early supporters, and at what cost.

The Experiment

My prior social media experiment was successful, and the results were encouraging. But I can’t just post on Instagram every week and expect to get to 200K+ in book sales.

That’s why the stakes for this experiment were terrifying: it would answer whether the world cares about my book outside of my existing warm audience.

When I was getting ready to run this experiment, I had a lot of anxiety. Publicly announcing I was going to sell 200K+ copies of my new book by running marketing experiments in public was frightening, bordering on crazy.

First, I created a simple website to send people to after they clicked on my ad.

I used Lead Pages to build it using one of their premade squeeze page templates.

It took me around two hours to build my squeeze page. There was no coding involved.

I decided not to invest in a fancy setup to import the email addresses into my system and instead manually add them to my Substack.

The offer I created was simple: enter your email to join my Substack and receive a digital copy of The Self-Taught Marketer for free when it’s released.

Once I created the squeeze page, I had to install my Meta Pixel to track conversions when people subscribed. That took about thirty minutes.

After I installed the Meta Pixel, I created a privacy policy. I used Termly.io to get it done quickly (maybe 10 minutes).

All that was left to do was create the ad. For the text, I used my book’s hook, followed by a call-to-action to go to read.selftaughtmarketer.com to join my Substack and get the book for free when it comes out. That took maybe 15 minutes.

I didn’t manually target a specific audience; I let Meta optimize based on conversions. I also didn’t restrict placements (Instagram, Facebook, etc.) or test any variations. These are the results from my first try.

The Results

The first result came quickly, much faster than I expected, and at a lower cost, with my first subscriber costing me $1.77.

To put this in perspective, let’s look at how much it normally costs to acquire subscribers for a newsletter/Substack. Here are some benchmarks to compare that to:

$2.50 = Good,

$2 = Great,

$1.50 = Excellent

$1.00 = World class

So, you can see why I was excited about $1.77.

The last three days of the experiment, each new subscriber cost me $1.57. That’s when it hit me. This might actually work.

And here is my Substack growth since I ran my first experiment:

What I Learned

The main takeaway from this experiment is that the hook I used on my warm social media following also attracted new Substack subscribers from a paid cold audience. This confirms that my book’s premise resonates with people who haven’t heard of me before.

That’s what I needed to know, and it was a huge relief.

One of my weaknesses in marketing is letting my perfectionism slow me down. In the ads I ran, I used a picture of the book with a subtitle I no longer plan to use, just to move faster.

I also skipped setting up automation to auto-subscribe people to my Substack after they signed up on Lead Pages. In the past, I would have let that derail my whole experiment.

I did eventually set up a Kit automation to send a welcome email and accidentally triggered a series of premade emails about motherhood. So if you got an email with the subject 'Things no one tells you about motherhood' from me, sorry about that 🤦.

Let’s run some numbers.

At $1.57 per subscriber, it would cost me $7,850 to gain 5,000 subscribers.

Many Substacks convert 1% to 5% of their free subscribers to paid subscribers. My paid tier is $8 a month (I have not worked on promoting paid much yet).

Let’s assume I can hit a 3% free-to-paid conversion rate. At 5,000 free subscribers, my list would generate roughly $1,200 a month.

But my Substack isn’t my only potential revenue stream. I’m sending a free digital copy of The Self-Taught Marketer to everyone who subscribes when it’s available.

Even if only 10% of those 5,000 subscribers also buy a paperback because they also want a physical copy, want to support the project, or want to give it as a gift, that’s 500 paperback sales at launch. At $7 profit a copy, that’s an extra $3,000.

That means I would break even on the ads within the first year.

Ad spend builds the list, the list buys the book, and book revenue funds more ads.

That’s my hypothesis. My next few experiments will tell me whether it is correct.

What is Next

For my next experiment, I will try to get 1,000 subscribers for $1,000 or less.

I am already at $1.57, which is excellent. But if I can get to sub-$1, my entire model becomes much more powerful because, under my current assumptions, it would take up to a year to break even on my ads.

The faster I break even, the more ads I can run. And the more ads I can run, the faster this whole thing could scale.

If I can’t lower my cost per subscriber, I will have to rely more on free-acquisition channels like viral marketing and cold outreach to hit my sales goals, which will be more difficult and may not work. Next week we find out.

Scoreboard

The Self-Taught Marketer Copies Sold (not out yet): 0

Sales Goal: 200K+

Substack Subscribers after my last experiment: 57

Substack subscribers this week: 153 (+96 since last experiment)

Substack Goal (pre-book launch): 5,000

Substack Goal (post-book launch): 50,000

If you enjoyed this post, please hit the like button. It helps more people discover my experiments.

Marketing. That is the short answer. The longer answer takes some explaining.

That’s the opening of the book I’m currently writing: The Self-Taught Marketer.

It teaches the lessons I learned marketing my programming books (200K copies sold in eight languages) and helping launch a podcast with more than a billion views.

When I first started writing it, I planned to write it for venture-backed founders. My last two books were about software engineering, and I’m from Silicon Valley, so the startup space seemed like a natural fit.

Before committing to that audience, I talked to potential readers. None of them were venture-backed founders. They were authors, aspiring podcasters, course creators, consultants, app builders, and small business owners. People with something real to sell but without venture money or a marketing team behind them.

If I had written the book for venture-backed founders, it probably would have flopped.

I came so close to violating my own book’s rules and marketing it to an audience without validating that anyone in that audience actually wanted it first.

That’s when I realized it might be interesting to document the experiments behind building an audience and a successful product: talking to customers, testing what works on social media, getting press, finding partners. The things successful products do that nobody ever shows.

That is why I created this Substack. I’m using my book’s own marketing system to launch it in real time. My goal is to sell 200K+ copies and outsell my previous books. And I’m documenting every step so you can see exactly how the methods I teach in it actually work in practice.

My hope is that these experiments help you grow an audience and sell more of whatever you’re building.

My First Experiment

I designed my first experiment to test one thing: how compelling the opening hook of my book is.

Marketing begins with one question: Can you get people to stop scrolling and pay attention? If you can’t, nothing else matters.

The test was simple. Was my hook strong enough to interrupt someone’s attention and get them to subscribe to my Substack?

I posted the following on my Instagram story and X:

The Results

First, some context. I have 42K followers on Instagram. But I rarely post. I think I’ve posted once in the last year. So, I was expecting the algorithm to punish me and not show my story to many people. Furthermore, a lot of the people who follow me do so because of my programming books, not my marketing advice. On x.com, I only have 3K followers and have been even less active on there over the last few years.

With that said, here are the results.

My Instagram story was viewed 7,336 times (with a few hours left to go). My tweet (my X?) was viewed 143 times.

Those two posts took me from 2 Substack subscribers to 57, with one paid subscriber. However, the paid subscriber was my mom. Thanks, Mom!

57 subscribers from 7,336 unique views means a 0.78% conversion rate.

That is a promising conversion rate given that most of my followers came for programming content, not marketing advice.

What I Learned

The experiment worked. My book’s hook is compelling enough to get people to subscribe before I’ve posted a single piece of content. That matters because a hook that doesn’t stop people from scrolling means nothing else in your marketing will either.

Now I know the hook can drive attention and convert subscribers, which gives me a channel to promote the book when it launches.

Doing this experiment was nerve-racking. At first, I thought I wouldn’t get any new subscribers. I kept refreshing, and nobody subscribed. I thought, “Oh no, I came up with a stupid book idea, I’m going to embarrass myself.”

But that is the thing. Good marketing requires discomfort and vulnerability. Most people won’t ask their best customers to get on a call and tell them what’s actually working. Or talk about a difficult subject. Or risk public embarrassment. But those things are often what makes the difference between a successful launch and nobody buying.

What Is Next

My next experiment is testing whether this same hook can convert cold paid traffic from ads. I will be running Meta ads using the opening of my book, targeted at solo builders, people marketing their own books, podcasts, apps, and products and services without a team. I’ll share exactly what I spend, who I target, what it costs me to get a new Substack subscriber, and whether the economics make sense as a long-term strategy for selling my book. If it works, I’ll have a repeatable system for acquiring subscribers and driving attention to my new book. If it doesn’t, I’ll tell you exactly why and what I’m trying next.

Subscribe if you want to see whether this experiment actually works.

Subscribe now

And feel free to DM me with any marketing questions @coryalthoff on Instagram.

Scoreboard

The Self-Taught Marketer Copies Sold: 0

Goal: 200K+

Substack Subscribers: 57

Goal: 50,000