How I Made & Launched prodmgmt.world

This is the story of making & launching & growing?prodmgmt.world, a super simple micro-SaaS.

I hope it surprises you & makes you laugh. Let’s go!

The Botched Launch?

We’ll get to the story of how I got to the product idea soon.

But before I wanted to tell the story of my nearly botched launch on?Product Hunt.

Midnight Run

I launched?prodmgmt.world?on?Product Hunt?in?October 2020.

Product Hunt?was one of the first channels I wanted to launch on.

Easy access to millions of people, for free, with a way to collect further feedback? No Twitter or Facebook can provide that. Especially if you’re launching something in the tech space.

But I was freaking out about the launch schedule on Product Hunt.

Reading?MAKE?&?Demand Curve’s Product Hunt launch guide,?I knew to time it to when the clock rolled over midnight in San Francisco.

This is so that when the day updates, your product is at the top & more people see it and upvote it. I am based in New Zealand and the time difference with the Pacific US is always a bit weird since we’re almost a day ahead.

The Product Hunt launch flow has a scheduler, but it was a bit janky. You know: plenty of async JS, several steps — so you’re never sure if you’ve saved anything in the end.

There was a ton of landing page copy & assets I had to create and upload to prepare.

But finally, everything was ready to go.

As I clicked through the launch setup flow, it looked fine.

The Comeback Kid

But at the very last second right after I clicked the button, it suddenly looked like it had scheduled the launch in the past. Wait, my past or SF-time past? I was confused.

So for a whole day, I wasn’t even sure if I had slated it for a release at all. That was on October 13th.

A few hours later, I found a URL for my “launched” product through my profile.

It looked like I had launched it. And maybe it ended up somewhere at the end of the previous day, not at the beginning of a new one. It had few upvotes.

I steeled myself. I had clearly failed, but that was ok. I moved on to other things.

Then, on Monday October 19th, I got a sudden stir in my mentions on Twitter from Product Hunt.

There was an email to let me know that I hit a milestone.

I checked — and there it was!

Instead of being at the bottom of the ladder, I ended up close to the top.

While I didn’t expect this to happen, I had been prepared anyway.

Later in this post, I’ll tell you how I set up the site for the launch.

But before that, I wanted to explain why I even made it in the first place.

The Product Management World… Is a Mess

Inspired?

As a newbie in product back in 2017, I leaned into practices and frameworks. I read Marty Cagan’s “Inspired” cover to cover looking for a lifeline.

“Inspired” offers a barrage of techniques you could try, without much context to how to do them well. It often felt like I was engaging in “product improv”.

(Many folks on Twitter & Reddit commented on the dangers of reading this book as a new PM. Whatever people say, the world of product would be different without Marty Cagan. I love the book & we owe Marty a huge debt of gratitude.)

Back in 2017, I knew about the Lean Startup. I read “UX for Lean Startups”, “Lean UX” & “Running Lean”. Yet I couldn’t see which techniques would apply & where when it came to the real world.

Besides, those were all tools for new products. My product was mature & riddled with dependencies.

There was a knowledge & cultural gap.

Whenever I said we should 404 Test some bright new idea that came to one of the executives, I got blank stares. “Why would you test something with such a bad user experience if the idea is a sure shot?”

On top of it all, there were org challenges. I knew I had to work with my design & engineering cohorts to explore the problem & the solution space. Yet the org structure prevented this.

The Internet was even a hotter mess of random content on near or adjacent topics.

When to do a 404 Test? When to do a Wizard of Oz test? When to just work on better strategy?

There was a disconnect.

Fast forward to March 2020, COVID struck our travel ecommerce company. Several people in the company lost their jobs. I was one of them.

Deadlines matter

Losing work was hard, but losing the shackles of org design was a blessing in disguise.

Now I didn’t need to sit in meetings with people who were not sure what they wanted and where they wanted to take the business. I decided to build something.

After a few explorations in the HR & content space, and a few failed job interviews to boot, September came.

It was a stressful time, and my patience was running low. It felt like I needed to jump back in the game and not worry about distractions and cute side-hustles.

But I kept getting pulled to this problem over and over. I felt that making something would build up my self-esteem, destroyed in the wake of my redundancy. (In many ways, it doesn’t matter what action you take, as long as you take it.)

At the time, Makerpad were running a no-code competition which I entered. The challenge was to build anything with a no-code tool in 30 days.

This was a great push.

Stuck in a no-mans land between idea & implementation? Use a forcing function like an external deadline.

Only intrinsic motivation lasts, I know. But for some of us right at the start, we need a push & we need structure.

But I still needed an idea for the challenge.

At the same time, the book by?David Bland?called?“Testing Business Ideas”?came across my desk.

Thumbing through its pages, I saw the beginnings of my product.

One big idea

One of my favourite?David Perrell?essays is the?“One Big Idea”. In it, the author talks about?Rich Barton, the founder of?Zillow,?Expedia?and?Glassdoor.

David’s argument is that you could live your life out building on just one big idea in perpetuity. In Rich’s case, it’s?“Power To The People”.

When I think about my One Big Idea, it’s something like?“Connect Ideas With Metadata”. I know I sound like a cheap?Google?knock-off (to organize the world’s information and make it universally accessible and useful). But it’s true!

Over the years, tech has developed its own lore. There was a loose set of methods that startups used to develop new products & test new ideas.

In most orgs, people have no clue about these methods to this day.

In?“Testing Business Ideas”,?David Bland?gave structure to this set of random methods.

If you’re a Wardley Map geek like me, this was a move from Genesis to Custom-Built on the Evolution axis.

I had known about many of the methods for years now, but in David’s book, I found a new way to connect them.

The first cut was obvious.

Risk it all

If you wanted to figure out if someone would pay you for the product, you’d need to test that one way.

And to test whether the value prop resonated with an audience, you’d need a different set of options.

Those would be different from testing how to make it work & scale.

This was?Marty Cagan’s 4 Risks, and David started to map the methods to the risks.

But there was more.

We discover new product & growth content every day.

How do we know how actionable a piece of content is?

And is it appropriate for the discovery stage of product development? A bit later during concept validation? What if it’s a growth stage opportunity? Who knows!

None of the content come with any caveats.

While David’s book is a treasure trove, it focuses on a few methods and data points. I knew there was still work to do.

Hello, World — meet prodmgmt.world

What documents did I develop?

A key cornerstone document I craft right at the start is a clear & concise pitch. It outlines the general problem, the appetite & the solution in broad strokes.

Here it is:?prodmgmt.world — Google Docs

Weird flex, but?it got featured in?Lenny’s Newsletter. I was so proud.

I wanted to be clear on the problem & the solution. It gave me power to write it out.

From an amorphous blob of ideas, it became concrete. I still didn’t have a good sense for what to build first and what next, but it gave me a sense for what I didn’t want to build.

Sometimes, spelling out the boundaries of the thing is all that it takes to take action. It will still be a mess, but a controlled mess.

Part of this document was a section on?Success Metrics.

I reasoned that if at least a 100 people visited the site 30 days post-launch, it would be validation to keep going.

I also knew that I wouldn’t have anything to measure stickiness, so I wanted half of the traffic to be return users. That’s how I thought to measure the fact that it wasn’t a mere blip of someone’s curiosity.

Everything else, from product usage metrics to revenue was too early to consider.

The thing I most cared about was learning more.

How did I prepare the site for launch?

Around the same time, I went deep on the Jobs-to-be-done theory and practice.

I had known about it for a while, but only on a surface level. Most people know about it on the surface level because it’s deceptively simple. Yeah, it’s the milkshake stuff. Yeah, people hire products to do jobs. Yeah, drill vs. hole anecdote.

But to get the insight, you need to do some work. It’s not a matter of throwing up post-its with some improvised jobs on the board and calling it a day.

So I knew that when time came for launch, I needed to gear up for interviews. People would come, use the product, come back, and then I would catch them with an interview request.

I knew that I didn’t want to spend $100 or more per interviewee. I decided to offer the interviewee to name their price for the interview.

So some time before the launch, I set up a Hotjar poll. I asked anyone who’d spent over 30 seconds on the page & engaged with the product to leave their email. Then I asked them how much money they would need to agree to a 1-hour interview.

Of course some people said “1 million dollars”, but many said “I’ll do it for free” or “I’ll do it for $10–20”.

While there may be bias here, I was happy to do some interviews with anyone who my campaign had brought in.

This was my free Online Ad experiment that I ran through Product Hunt, not Google Ads. The outcome was similar.

Someone was interested enough to click through to an “ad” — a listing on Product Hunt. Then they were interested enough to stick around long enough for me to ask them their email. And they were still interested enough to give their email to me.

But where would people be coming to in the first place? Next, I’ll tell you more about how the product came together.

How did I make the database?

I spent over 500 hours on finding the content for the core prodmgmt.world database.

I scoured the web. I looked under every rock to find templates, frameworks, techniques with good explanations.

I spent days tweaking the categorisation engine and meta-data. I modelled & re-modelled the database to make it easy to search & discover content.

I knew early on that I would likely use Airtable for my database of choice.

Airtable is one of the best tools on the planet.

At first, I thought it was a pretty Google Sheet.

But it’s so much more.

I am a hardcore database geek. I had to learn SQL on the job. The learning had to be fast. It happened under severe pressure of potentially losing work & my immigration status.

I know how steep the learning curve for getting started with databases is. So Airtable is like Prometheus stealing fire from the Gods.

But it still felt a bit naked to have a database as a product. The product needed some clothes.

What is the core innovation?

Remember, the content and the meta-data are the core innovation.

My core innovation was:

• add more metadata
• find & add more methods
• make them easier to access & filter

The content was there. The new thing was how I put it together & categorised it.

However, the UI almost didn’t matter. In my first iteration I was competing with a paper book & tools that had no metadata (just a list view). Anything would be an improvement. The UI was the last in the list of risks to tackle.

The whole project was an experiment, an MVP. If you know how to build fast and cheap, you research as you build.

My initial idea was just a landing page highlighting benefits (+ some feature designs). The main CTA was going to a Typeform that would ask the user some questions. At the end of it, Typeform would generate top 1–5 options the user could take.

Imagine if you were to tell me that you were working without any UX designers or data analysts. You were trying to validate the viability of a product you built under HiPPO’s instructions. What should you do? What techniques or methods could you try? The product would give you some answers.

I ran into feasibility issues.

I was about to invest in a recommendation engine, when I didn’t even have conviction in the concept.

The recommendation engine would either be too crude to be helpful (rules-based). Or I would have to collect a lot of data to navigate this space well. I felt like there would be a threshold in the survey at which users would drop off.

Since the goal was learning, I wanted to get more people into the funnel at the start. I didn’t want to figure out how to plug a risky ML model into an unvalidated product.

I did what any PM would do in this case and opted for a Concierge experiment. People would still be able to ask the questions. The responses would come within 24 hours. The man behind the machine: me.

How did I develop the visual look of the product?

That couldn’t be the shtick. Forcing people to take a survey is always a funnel-narrower. I wanted a bit more meat on the bone.

I’m skipping the part where I turned the database & the brand into a website. This part was the least interesting bit. I’ll touch on it briefly.

I searched for a tool that would work with Airtable. Luckily, right on Product Hunt, a tool had just launched that did something like it. I tried it, it worked well for my needs, and I used it. Kudos to Sam from Pory.io!

But what I did to design the look and feel of prodmgmt.world might be complete blasphemy in the design circles.

I designed the site’s whole identity in under an hour using a Figma plugin, an emoji, a free font & an Instagram account.

Here’s what I did:

1. Went to?colorpalette.cinema?on Instagram.
2. Found a few palettes that I liked. Put them into Figma.
3. Used the Able accessibility tool to make sure the colours worked for accessibility. Tweaked a bit.
4. Picked a monospace font (because I love them).
5. Created a logo. A blob plugin in Figma + an emoji that isn’t used very often & works with the playful nature of the concept.

What came out is what you see on the site today. It may be violating a lot of rules and sensibilities, but I did it in under an hour total. What’s non-essential, you either avoid, or have fun with it as long as it doesn’t take too much time.

But take my advice with a grain of salt.

If you’re a real designer, obsess over the colours & typography for years before your MVP is perfect (I’m kidding).

Why didn’t I monetise sooner?

After making sure that the UI wasn’t too complicated and did the basics, I moved to launch.

By now, you know the launch was a success. I got over 3000 users to visit this unknown site over the course of a few days. I got over 200 emails and some great opportunities to interview people.

I got backlinks and a steady stream of daily users that come to the site to this day.

So why didn’t I monetise this right away?

Let’s be clear. Anyone who is new to business always charges less than they should.

It’s an ego thing.

It’s why so few people do standup comedy, or when they do, don’t ask to get paid right away.

You don’t want someone to tell you, “I don’t see your value, you’re too expensive.”

But you don’t know what your skills are worth unless there’s a market that can help benchmark those. Even then, it’s imperfect.

It’s very much about competition.

If you don’t know who you’re competing with, how can you know what people are even comparing you against?

Were people going to compare?prodmgmt.world?to?David Bland’s?book.

And if so, should I have charged them?more?or?less?for the product I made?

It’s about learning.

Asking someone to pay you is the question that removes a thousand questions. But we don’t want to learn the answer.

In my case, my original goals were different at first.

I didn’t even know I could enter the Pantheon of Product Hunt launch medalists. Why? My reality was that “This is what other people do. Not me. Who am I?” This was the first mountain to climb.

The sooner we realise that other people are like us & that most glass ceilings are not real, the better. Social media & advertising thrive on the editing of the visual reality so we see a different one — a reality that isn’t real. This makes things that are close & possible seem far.

In 2007, on a boring language exchange trip to Soria, Spain, I read a book by Bret Easton Ellis called “Glamorama”. That book made me question the reality as I knew it. It was such a shocker, it made me into a different person for at least 3 months. I questioned the reality because I couldn’t believe a book like that could exist.

Launching prodmgmt.world was my “Glamorama” moment.

Hope this was helpful!

If you liked this, you might like my other writing:

• How To Become a Senior Product Manager
• Start With Why — But Don’t Stop There
• The Micro-verse of Strategy
• How To Get Buy-in Without Trying
• How Elon Musk, Charlie Munger & Aristotle Had You Fooled

Make sure you?say hi on Twitter.

And of course, check out?prodmgmt.world!