Execution Isn't Everything: How to Design a Launch Strategy with Learnings from Indie Hackers

Written by Guan

I'm about to launch a second indie software product.

Before even writing a single line of code, I did a ton of research on how to design a business so that I'm setting it up for success, and reducing the chances of it failing.

This essay is a roundup of those findings.

Indie Hackers is the go-to resource for finding and learning about successful indie businesses. It's super useful for help in generating ideas for the kinds of businesses to start, as well as choosing which one to work on.

I've been constantly digesting its content ever since Courtland launched it, and over time, I've come to notice a pattern of a few key attributes that successful businesses tend to share.

These are:

1. Make something people (already) want.

2. Charge more, and from the start.

3. Launch now, and keep launching.

4. Choose an idea you'll be happy working on for the next 5 years.

I'll be going into detail about each point, as well as share some examples of how I've implemented these concepts in a software product of my own: WriteMapper, the mind-mapping desktop app for writers (which I used to draft and outline this piece of writing).

Let's get started.

1. Make something people (already) want.

Like writer's block, starting out with a blank slate of a business idea can sometimes be a little difficult to overcome.

My approach to this is to start with the subset of business ideas that have already been proven to work.

Be in a market that exists.

To validate new product ideas, most of us would have been taught to start by trying to find product/market fit.

While that's one way of doing things, I find that I prefer to skip this step entirely, and take a shortcut—be in a market where there are plenty of examples of competitors. This is proof that this is a market that exists, and is viable.

My absolute favourite post on Indie Hackers is Trevor McKendrick's interview on his Spanish-language Bible app for the iPhone:

"I didn't do any validation other than scrolling through the App Store... I think a lot of entrepreneurs focus too much on making something new, which then has to be "validated," instead of looking around and noticing industries that already exist and are doing well.

I used an approach very similar to Trevor's in coming up with WriteMapper, looking through the Mac App Store and deciding on a mind-mapping app.

Finding product/market fit can take a really long time. On top of that, it's not a guaranteed outcome. In the process of building something original that people want, even well-funded startups can die or take years to find product-market fit. This adds significant risk of failure to your endeavour, which is obviously undesirable.

Make a thing that already exists.

On top of being in an existing market, you also want to ensure that you're making a thing that people already buy. Examples of such "things" include e-books, online courses, desktop apps, mobile apps, etcetera. Customers already consistently and commonly look and pay for these things.

With these, there's no need to educate the customer about the format of your product. Customers will understand what an e-book is, and what they will get after they give you their money.

Don't invent new product formats that people don't already pay for, which will force you to spend time educating your customers. You'll also waste time going through the whole product/market fit dance.

Stick to tried and tested business models and formats to make and sell. It's interesting how this doesn't get echoed more, because of how advice like this is uninspiring and unremarkable.

There's many examples of uninspiring, unremarkable, but very successful businesses on Indie Hackers. Tyler Tringa’s StoreMapper, is basically a map of store locations that can be embedded on any website. Philippe Lehoux's ConferenceBadge.com is just printed conference badges for events and conferences.

You can also see this in WriteMapper, which is a desktop app. People already buy desktop apps from app stores all the time — it's a market that exists.

“Make something people (already) want.”

Most founders have to abandon their original ideas anyway to arrive at product-market fit, pivoting (sometimes multiple times) in favour of letting the customer’s wants define what they build, instead of the idea that orignally existed in the founder’s mind.

If you don't want the odds stacked against you, don't make something entirely brand-new.

Leverage prior experience.

Another clue to help decide which product idea to work on, is to leverage your past experiences and successes, and build a product that complements that.

One such example is past product launches. Nathan Barry of ConvertKit started out publishing e-books and selling them online. He then used what he learned from that experience to decide on building a SaaS email marketing product:

"I decided to build an email marketing tool built specifically for authors/bloggers like myself who needed a good solution for selling digital products."

Another way you can approach this is to teach what you've learned at your job to an audience that will benefit from that valuable knowledge. Justin Jackson published Marketing for Developers doing exactly this. From his 2017 year in review blog post:

"Most of what I teach was learned in the trenches while I was Product Manager at Sprintly and Mailout."

Just like how Nathan built upon a smaller, first product (the e-book) of his to start a bigger second thing, Justin intends to do the same. He intends to start a SaaS product in 2018—I wouldn't be surprised if it were in the same domain as his prior ones.

The benefit from this pyramiding upwards is at least two-fold: First, you're able to create a much better second product now that you're armed with this experience. Second, you can give yourself a boost in getting your second product off the ground, by having an existing audience to introduce it to.

I've shared how deciding to work on WriteMapper was quite an arbitrary decision, but it's allowed me to grow a tiny base audience with whom I will be able to share my second product, which will be an iOS version of the app for the iPad.

2. Charge more, and from the start.

Now that you have a general sense of what you'll be working on, it's key to not get carried away by the idea.

It's important to acknowledge that creating profit is still the imperative, and you're creating this business for this objective.

Don't do free.

First off, make sure that you're charging for your product at all.

I'm much more partial to the word 'business', than I am to 'startup'. 'Business' implies that the objective is to sell a product or service to a customer for revenue, with the objective of turning a profit.

A startup, on the other hand, is defined primarily by its growth. I guess it wouldn't be such a bad thing to make your product free if you have a ton of VC funding, and your primary objective is growth, not profit. For indie businesses, however, this is not the case.

Hence, it's only logical and necessary that you charge for your products, so that the business is able to generate revenue.

To quote Pieter Levels, whose product Nomad List is Courtland's inspiration for Indie Hackers:

"If you have an app with traction but never charge money it's not traction, it's an expensive hobby".

With WriteMapper, where the goal was to achieve some form of initial business success, I explicitly made it such that there was no free version of the app, and you had to pay for it if you wanted to continue to use it. I even went as far as to make special mention of this on the FAQ section of its pricing page, taking much insipration from Pieter's FAQ page on Nomad List.

Charge more.

Businesses almost always charge too little for their products initially.

Patrick "Charge More" McKenzie elaborated quite a bit on how creators tend to underestimate the value of their products. From the Indie Hackers podcast episode he's featured in:

"We [creators] see, because there's no magic involved in bringing something into the world, that since there is no magic and since it is free for me to create software, it must be free for people to buy software, which is the opposite of the truth, because it is totally impossible for them to create the benefit of software without paying for it. They're willing to pay a lot of money for it because software is fundamentally magic if you can't actually make it."

Also, if you're selling to businesses, they can well afford it even if you significantly increase your prices: "In a business context, anybody who can pay $19 can pay $29, but that's 50% extra to your bottom line immediately".

On top of that, as shared by John O'Nolan of Ghost, pricing too low attracts a very undesirable type of customer for a business to sell to. They tend to nitpick, clamour for discounts, and constantly require your attention in customer support due to their inexperience and indecision:

"$5/month customers are just terrible. They have the highest rate of failed payments, the highest rate of credit card fraud, the highest amount of support tickets submitted, and are the least friendly people."

As a rule of thumb for how much to actually price your product at, John says in the same interview to charge at least $10 per month, while Patrick recommends a stock pricing model of $29, $99 and $249 per month for any SaaS business just starting out.

WriteMapper isn't a B2B product, but a consumer desktop app. It runs on the one-time sale business model that's familar to the desktop app consumer audience. It started out with the three price points of $26.95, $38.95 and $89.95, and I've since increased the pricing to $39.95, $49.95 and $109.95, with no discernable negative effects on sales.

Charge from the start.

Although we've already gotten some degree of validation for our idea by adopting exsting ideas, as previously discussed, we still do want concrete validation of our product, by having customers give us money for it. Besides, the whole point of this work is to make a profit.

“Charge more, and from the start.”

This means that at the time your product's v1.0 is launched, it should be launching alongside a fully operational "Buy Now" button that will let your customers put money into your bank account. Never launch without a payment button.

In Courtland's words:

"Getting mailing list signups is the easiest way to validate your idea, but trying to convince people to become actual paying customers is the most accurate method. Start charging as early as you can!"

WriteMapper comes with a 7-day free trial, so when I launched it on Product Hunt, I wasn't expecting anyone to make purchases until about a week later. I was wrong. I received the first sale on the day I posted it to PH, and more came afterwards. It was only then that I was able to say that the product idea was validated.

3. Launch early, and keep launching.

The commonly echoed advice when it comes to new products, is to build an MVP and get it launched as quick as you can. Perfectionism be damned!

However, it can be logically inconsistent to blindly produce an MVP without first understanding why you're doing it.

Launch early but ready.

Let's step through the objective of an MVP—to validate a product.

  1. We know that to accurately validate your business, you need to charge for it, and see if you get validation in dollars.
  2. To validate the product idea as early as possible, you should charge for it since the day of launch.
  3. To start charging from day one, it follows that you have to have a product worth paying for from day one.
  4. To have a product worth paying for from day one, your MVP can’t be a shoddy, poorly made product.

So, the product should not only be fully functional at launch, but also feel well-designed, polished and above all, ready for the customer. Doing things like labelling your launch a "beta" are simply counterproductive to the objective of validating your idea, as it devalues the product and deters your customers from paying for it.

Your customers need to take your product seriously, or else they'll never put money down for it.

From this gem of a comment on Indie Hackers:

"Yes, you're supposed to launch soon to get feedback as early as possible. However, I think that early feedback is going to be useless if it's nothing but people saying "This is shit" because you made a sloppy product to launch in a week."

The founders of Canny have also experienced this in their launch, as they've described in an Indie Hackers post:

"If you sell to tech companies, Product Hunt is a great place to launch. It’s a one-time trigger, not your marketing strategy. If you do it sooner, you’ll get valuable feedback. If you do it later, you’ll get paying customers. I’m glad we did it later."

Rather than trying to launch a sloppy product that would clearly be seen by customers as not ready, I took a couple months to build out my product. I made sure to launch both WriteMapper only after I felt like the product was able to hold up as a product that customers would pay for.

Make a kick-ass half, not a half-assed whole.

Launching with a fully functional product that's ready for your customers does not equal launching a feature-complete product.

You can shift the goalposts for what’s defined as "complete" for your v1.0, by stripping out everything but the absolutely crucial features for your product to work, to get the launch out as soon as possible.

You can always schedule more features to be added after launching, by putting them on your product roadmap.

Be particularly aware of this especially if you have an idea you really like, because it's very easy to get carried away thinking twelve steps ahead.

From the same Indie Hackers comment in the last section:

An MVP is minimal. That means you're grinding your idea down to just the absolutely necessary elements for launch, which means you can launch sooner than if you had a fully featured product.

I did this with my launch of WriteMapper, launching only with the features I determined to be the bare minimum, and adding more features after launch via software updates.

After WriteMapper launched, users also able to give feedback on the features they wanted in the product. This helped me to better prioritize which features to work on and include in the next product update.

Know your distribution channels.

Getting your product in front of an audience means that you have at least an initial launching strategy.

But, like the founders of Canny have shared, launching on avenues like Product Hunt and Reddit is "a one-time trigger, not your [entire] marketing strategy".

This agrees with what Sharon Pope, Head of Programs and Marketing at YC Continuity said during her YC Startup School (2017) lecture, How to Think About PR:

"[Getting a story in the press is] not guaranteed to drive a specific business goal because it's not an engaged, measurable customer acquisition tactic."

The tendency for a first-time founder is to build a one-of-a-kind, special product that manages to get press initially, but is drastically harder to market after that first launch, since the product isn't a thing that people are already actively searching for.

“Launch early, and keep launching.”

I learned this the hard way—With WriteMapper, I wanted to make something original and remarkable. While I did get a sizeable amount of traffic during its launch, it didn't last. As of now, WriteMapper doesn't have a sustainable marketing strategy.

Instead, to keep the business continually functioning after the initial launch, you have to have a sustainable distribution or marketing channel in place.

One way around this is if you can structure your business to always be launching.

The founders of Creative Tim have structured their business to employ multiple freemium mini-products that create value for communities, whenever posted to social media platforms, such as Reddit, Hacker News and Product Hunt:

"The basic idea is that those freebies are always appearing on top 10 lists in those big communities. Each post that's in the top 10 gives you between 1,000 and 15,000 targeted users in one day."

I've learned that to make recurring revenue, your marketing efforts can never fully take a break.

With WriteMapper, my intention is to exhibit it on this blog, using content marketing as my distribution channel, whilst showcasing my use of WriteMapper in each post.

4. Choose an idea you'll be happy working on for the next 5 years.

Lastly, your business is probably not going to be an overnight success. You'll have to work hard consistently, as you continue to improve it.

In the blank canvas, ideation stage of a new business, it pays off to make a good decision about the type of business you'll be working on. You need to be sufficiently motivated to work on it, even as time goes by.

The best case scenario is to build something you’ll be enthusiastic about for a long time. If you’re hesitant about the idea at the time of ideation, forget it. You’ll not be enthralled enough to get a product up and running within the next three months, let alone persist through the next five years.

In the words of Patrick McKenzie (patio11), from his Indie Hackers interview on Appointment Reminder:

"My biggest challenge is that I gratuitously violated the Peldi Rule, which is to work on something you'll be bouncing up with enthusiasm for for the next five years. This impacted both my desire to execute and also our marketing success."

One anecdotal observation I've made about idea/founder fit that I think is worth sharing: where there is good idea/founder fit, it should feel like the idea keeps growing on you. The more you think about the idea, the more it resonates with you, and you start thinking about directions you could take this, ideas for growing it... you just like it more and more.

“Choose an idea you'll be happy working on for the next 5 years.”

With deciding on the WriteMapper idea, I started off by first knowing that I wanted to publish my own blog as a long-term content marketing strategy.

It felt right that I would be able to showcase the product alongside my writing, using WriteMapper to turn my ideas into essays that could be shared with a larger audience on the blog.

And those are the attributes I've tried incorporate into the launching strategies for WriteMapper, a mind-mapping desktop app which I used to draft and outline this piece of writing.

Thanks for reading, and I hope you've found this helpful. Subscribe to the blog below if you would like to hear more from me!