Here’s what I learned hanging out with Jason Fried

Discuss on Hacker News.

I went to meet Jason Fried so I could learn how to stop selling software by accident.

Since I started programming 10 years ago, I’ve made a fair amount of money online. But those sales were mostly coincidental.

By that I mean, I never thought deeply about how and why products were bought. I would build something, release it and drive traffic to it. Sales would almost always trickle in. But I never took the time to understand who was buying and why. I never worked on refining my sales copy, or understanding which forms of traffic brought the most customers.

Once I released a product I would do one of two things: start working on adding new features or start working on something else. I didn’t take time to understand what was working and what wasn’t. I wasn’t really interested in refining and simplifying – just seduced by the prospect of building something new.

You can learn things by doing this. I’ve learned a lot in the last 10 years. But I think getting from being good to being great requires something else. Building great products requires constant practice at the art of building understanding. It also requires getting rid of everything except what’s absolutely necessary.

With this in mind, what my co-founders and I have been working on over the past few months with Firefly is learning how to sell deliberately. When we make a sale, we want it to be because the copy addressed customer pain and offered a solution they could connect with on an emotional level. We don’t want to make a sale because a customer is smart enough to swim through a list of features he doesn’t care about, and come up with a reason to pay money on his own.

Selling deliberately requires building understanding, and Jason is probably one of the best people in the world at doing this.

In November we set up a meeting over Twitter. Three months later I was in 37signals’s office sitting across a round wood conference table from him. We spent about two hours chatting about Basecamp (his product), Firefly (mine), and a bunch of stuff in between. Here’s what I learned.

Building products creates an information problem
There are two obstacles to selling deliberately: lack of information or too much information.

When you’re just starting out with a product you don’t have very much information. Given this vacuum, most people guess at why someone might want what they’re building – but frequently those initial guesses turn out to be wrong.

So what most of us do is talk to everyone for feedback about what we’re building. I have 37 notes in my Evernote from separate conversations with people about Firefly. And those are just the ones that I bothered to write down.

But talking to everyone creates an information problem. It’s very difficult to sort the wheat from the chaff. And the information must be sorted through. If you base your copy off of the feedback you get, it’s impossible (and not helpful) to use feedback from everyone. It’s a common aphorism in sales that if you sell to everyone you sell to no one.

The information problem can be solved
Jason’s solution to this problem is simple and effective: the only two people who can give you real feedback about your product are people who just purchased it and people who just canceled.

You can certainly gather feedback from other people. But in general it’s going to be less valuable.

The customer who’s been with you for 8 years and loves everything you do isn’t going to give you useful feedback. She’ll just yes you to death. And neither is the guy who will definitely sign up and pay you Big Bucks if you just include, like, these four features. He’ll just string you along and eventually disappear.

A customer who just signed up can give good feedback for this reason: they’re only recently removed from the buying process. Because of this, it’s easier to get useful information about why they bought your product, what emotional event caused them to look for it and which part of it is actually useful for them.

If you interview your recent customers you can use their words to sell your product. It helps you get into their mindset, figure out what’s important to them, and find what it is about your product that convinced them to hand over their credit card details.

Similarly, a customer who just canceled can tell you exactly what you’re missing. They signed up and put their credit card information in because they thought you would be able to do a job for them. If they cancel, your product didn’t deliver what they expected.

Sometimes this a problem with your copy: you didn’t provide what your website promised. Sometimes this is a problem with your product: maybe it’s too slow or too buggy or doesn’t solve the problem that you intended it to solve. Either way, in-depth conversations with customers who have recently canceled can help you puzzle out these issues and fix them for the future.

Don’t have customers yet? Jason had an answer to this too. Find your competitor’s customers and talk to them.

How to ask questions that get actionable answers
Just because you know who the right person is to talk to doesn’t mean you know what to ask them or how to ask it. This is actually my biggest issue with the customer development methodology: it tells you that you need to ask questions but it doesn’t say how to ask or what to ask about.

Let’s start with how to ask.

The human brain is a tricky, messy thing. It doesn’t deal well with high-level abstractions like “what are your problems” or “how can we make this better”. If you ask someone a high-level question about problems with their job you’ll generally get nebulous answers about how they hate their nosy co-workers or are afraid of angering their bosses. These answers are useless because, in general, abstract human problems can’t be solved by software.

But the human brain is very good at talking about specifics. Questions like “can you walk me through what you do everyday” or “are there any repetitive tasks that you do day-to-day” will lead you down a much more interesting path. You’ll find problems that your software actually can solve.

Now that we know how to ask questions, let’s discuss what kind of answers are useful to have when figuring out how to sell your product.

What job does your product do?
The first thing that you need to find out is what job your product does.

On the surface it’s easy to answer this question. A shower helps you get clean. A backpack helps you carry things. A wallet keeps your money together. Firefly helps you screenshare with your customers.

But people aren’t rational. They don’t make buying decisions because of what a product does in general. They make product decisions because of what a product means to their lives.

After we had talked for a while, Jason brought a YouTube video up of Harvard Business School professor Clayton Christensen. Clayton contends that “products find a certain market only when they help their customers get done the jobs that they have already been trying to do.”

He tells a story about a fast food restaurant that was trying to understand how to sell more of its milkshakes. They did the traditional market research that any big firm does. They figured out their target demographic and brought those people in for focus groups.

In the focus groups they asked participants: “What can we do to make our milkshakes better for you?”

And they got feedback. People wanted more chunks, or more chocolate, or a new interesting flavor. The company would go back, and according to customer feedback, modify their lineup of milkshakes.

But it never had any effect on sales.

So they brought in a consultant to help them figure out how to sell more milkshakes. The consultant approached the problem by asking a very specific question: “What job do people hire a milkshake to do for them?”

What he found was that 50% of people bought their milkshakes in the very early morning.

So he made a point to go out the next day and start talking to customers who had just purchased their milkshakes. A very interesting pattern emerged.

Nearly all of the people who buy a milkshake in the early morning are on long commutes. And while they were on these commutes they needed something to do.

It turns out a milkshake is a perfect food for a long drive to work in the morning. It’s sweet. It takes a long time to eat so it lasts for most of your drive. It’s in a cup so that it doesn’t make a mess. And it’s pretty filling.

So for these people, a milkshake wasn’t just a sugary drink. It was something to keep them occupied on their commute.

Once you figure out what job your product does, there’s another important thing to find out.

What are people switching from to use your product?
In the milkshake example, it turns out that the people who buy milkshakes in the morning had tried other things. Some had tried to eat bananas. Bananas are healthy and they taste good.

But unlike a milkshake they’re gone pretty quickly – they don’t last for your whole car ride. And they’re also not very filling.

Other people tried to eat donuts. Donuts work well because they’re sweet.

But they get the steering wheel sticky and make a mess in the car. They also don’t fill you up very well.

It turned out people were switching to milkshakes because they were the perfect solution to keep them occupied on their commute.

The same goes for any other type of product.

You say people aren’t switching from anything to use what you’ve built?

That means one of two things: either you don’t understand your product, or no one wants what you’re selling. Every product has competitors. Sometimes they’re other products and sometimes they’re human processes.

Knowing what they are switching from tells you a lot about how your product should be marketed and what problem it needs to solve.

And when you do the research, you’ll find that what people are switching from is often-times surprising.

Your competitors may not be who you think they are
In the milkshake example the fast-food restaurant initially thought that their competitors for the milkshake market were McDonalds and Burger King. But as it turned out, they also had a broader range of competiton that they had no idea existed.

This applied perfectly to my product as well.

Firefly does download-free screensharing for customer support. And one of the things that we think about a lot is how commoditized the screensharing market is. There are bunch of free screensharing products out there. So in order to charge money for our software (which we do) we need to understand our customers’s needs better than anything else on the market.

I gave Jason a demo and he seemed to like it. He called in his customer support person and we demoed it for her as well. To figure out what job it might do for her, we asked her what she does right now when she’s on the phone with a customer who needs help.

We found out that right now her process doesn’t include screensharing. It can’t, because any type of screensharing that requires a customer to download software is a nightmare for a support rep to explain.

What she does instead is she asks the customer to describe the page he’s looking at and what he wants to get done. Then she, from memory, instructs them on which buttons to press to fix their problem.

Many times they describe the page to her and she has no idea what they’re looking at. Other times, she’ll tell them to press a button she believes is on the page they’re looking at but is told that the button isn’t there.

Worst of all, she can’t tell whether there’s an actual bug in the product or the customer just isn’t looking in the right place.

Firefly fixes this problem: she can see exactly what the customer sees without interrupting their browsing experience at all. But what’s interesting is that we’re not really competing against other screensharing software in this case. We’re competing against her memory and her willingness to explain things from memory day after day to customers.

This is key to understanding how to sell it. Just because we ostensibly do the same thing (allow someone to see a customer’s screen) doesn’t mean that we do the same job as traditional screensharing.

And if we reflect this in the way we market our products it will help us find more customers and charge more money.
Think about who your real competitors are. To figure it out, find what job your product does and what people are switching from to use it.

After my meeting with Jason, my girlfriend and I spent the rest of the day in our hotel room pointing out different products and trying to decide what job they do. It’s a fun game, and some of the things you’ll find are surprising. For example, a shower doesn’t just get you clean – its job is to be a place to get away from everything and be alone with your thoughts.

You may have noticed that a lot of what I’ve said in the preceding pages seemed to apply both to product development and product marketing. This is because understanding your customers tells you what features to build and also how to pitch your offering. That brings me to my next point.

Great marketing and a great product come from the same place
Jason never explicitly told this to me, but it seems to ring true from everything else he said. If it turns out he doesn’t actually believe it, I’ll take full credit.

When a lot of people think of marketing or sales they think of tricks that fool people into buying something. But great marketing doesn’t do that. Great marketing comes from understanding exactly what the customer needs on an emotional level, and showing how your product will satisfy those needs.

A great product comes from the same understanding. When you know what your customer needs emotionally you can build something simple that fits those needs exactly.

When you understand what each UI element in your product means to your customer’s life, you’ll understand how to sell it better. What your product is and the way it is sold are completely symbiotic. And both depend on detailed understanding.

This realization is both simple and beautiful.


I think the most important thing that I took away from all of this wasn’t necessarily the specific advice he gave, although that was certainly valuable. What was most important was how he gave the advice, and what it reveals about his approach.

The first thing that I noticed about him was that he wasn’t trying to convince me that he was right.

“This is what we’ve found works for us, and it could be wrong, but it seems to work,” he said gesturing at an image of Basecamp’s homepage being beamed to the wall by a projector wirelessly hooked into his Macbook.

I had come prepared with questions, a lot of them, and he had answers. But not quite the typical kind.

You see, when smart people give advice they tell you what you should do.

But what I’ve found over the past few years is that when the smartest people give advice they do three things:

1.     Before saying anything, they ask what you’re hoping to accomplish
2.     When they do give advice, it comes with a few caveats
3.     They don’t hesitate to say “I don’t know”

Jason was no exception. He was interested in why I was doing things just as much as what I was doing. (Aside: as a philosophy major nothing makes me more excited.) And he didn’t seem concerned with trying to appear right at all costs.

What he seems to do really well is concentrate on finding everything unnecessary about his business and removing it. That requires a deep understanding of his customers that he seems to pursue relentlessly.

It’s very easy to do things without taking the time to go back and evaluate how well they work. What’s difficult is to constantly and deliberately revise until you’ve cut to the bone of the problem.

That’s the mindset he seems to have. And that’s what I want to take with me.


Thanks to Arielle Shipper, Ryan Shea, Mark Bao, Paul Singman, Joanna Weibe, Justin Mares, and Tony Diepenbrock who all read drafts of this post.

7 Jan 2013, 3:41pm | 32 comments


Never miss a new post