Fighting Subjectivity When Building New Products
May 23, 2020
If you answered “A. Marty is a nuclear engineer.” you are correct! Yet on average people have been shown to answer “B. Marty is a nuclear engineer who loves socializing.” (Try it on your friends.) Why is that? Our brain is naturally primed to want to answer B. It sounds more plausible given the bias I purposefully introduced about how Marty is the life of the party and all. Yet it is less probable. This is termed the Conjunction fallacy.
Note: Why is A more probable than B?
Let’s say Marty has a 1/10 chance of being a nuclear engineer. And let’s say the probability that he loves socializing is 1/2. (You can choose any arbitrary probabilities.)
That means the chance of A is 1/10.
The chance of B is 1/10*1/2=1/20 by the Multiplication Law of Probability.
1/10 is larger than 1/20 which is why A is more likely.
If you’ve ever successfully been pitched, persuaded, or sold you’ve likely experienced this phenomenon at play. One compelling story can be more convincing than tens of charts and stats. Stories put us in the shoes of the hero which spins up our imagination and makes us experience events more naturally through time. Probabilities on the other hand are abstract and unnatural. We don’t think in terms of 2.4 Jims. Yet probabilities expose the reality of the situation more accurately.
Like most things in life, a combination of both is most relatable and compelling. Great examples of this are certain documentaries that tie together numbers with stories. I believe the “Explained” series on Netflix is a good one.
So how can we fight this plausibility bias to be more realistic with ourselves?
Plausibility Busting Mindset
I recently followed a link to a live YouTube video and it was Michael Siebel and someone else talking. I only watched it for a few minutes but got to hear him say this which made me laugh because I imagined a lot of 🙁 in the audience:
- Your startup is dying right now.
- It feels as the next big thing is around the corner. It is not.
What at first sounds discouraging and might feel uncomfortable is actually great advice. The point is to constantly remind yourself that any positive emotion you derive from your work is likely an illusion. Only evaluations you can hope to be objective are measurable results.
Strong emotions lead to subjective decisions which are bias prone. When you are at a point where there are no results yet, all of your decisions are subjective. This is very dangerous. So what is the common advice? Focus!
“It’s all about focus!”
“Focus on one thing.”
“Focus. Focus. Focus.”
Man do people love saying shit like this. You’ve probably heard some form of this well intended advice doled out before. Does it mean something specific? Is it vague folk wisdom? When we first started on our value creation journey at Asgard Analytics in 2018 I remember thinking:
- Focus on building just one thing? What if it’s not something people want?
- Stay focused, as in, get off Reddit? Can’t be, too obvious.
- Focus on one outcome like making money? Ya, but too simple.
- Ignore new customer information to stay focused? Hmm.
Two years in and we’ve probably misinterpreted focus in every possible way. But today our daily focus is always clear: systematically eliminate risk. Keep in mind that I still don’t have $1B exit so it is probable that this is wrong.
Fighting Plausibility With A Risk Based Mindset
Risk is an interesting plausible sounding yet probability based concept. That makes thinking of things in terms of risk a powerful plausibility busting tool.
When building a new product there’s a seemingly infinite amount of paths to take. It’s easy to lose focus when a problem seems to have an unlimited number of solutions. Yet risk allows us to flip the problem on its head and componetize the huge dimensionality of what it means to build a product into just a few categories. At the very least, it allows us to answer the question of what we don’t want to have. The task of succeeding becomes the process of elimination and infinity can be brought down to roughly four parts.
The Four Big Risks
From Marty Cagan’s excellent blog post, the four big risks when it comes to building a product (questions by me):
- Value Risk – whether customers will buy it or users will chose to use it
- How badly do people want the emotional or functional outcome?
- How much does the solution help with achieving that outcome?
- Usability Risk – whether users can figure out how to use it
- Is it understandable?
- Is it usable?
- Feasibility Risk – whether our engineers can build what we need with the time, skills and technology we have
- Is the team capable of building the solution in time?
- Business Viability Risk – whether this solution also works for the various aspects of our business
- Can the team reach enough people?
- Are there enough people who need it?
- Are people willing to pay enough for it?
Not All Risk Is Equal
If a problem is not important, no solution for it will be valuable. It doesn’t matter if it’s the most usable design you’ve ever seen or a technological marvel. And if it’s not valuable it’s certainly not viable to be a business. As YCombinator likes to say, “Build something people want.”
Put another way, value is king. If you’re a visual thinker, stay out of the “No One Cares Zone” in the Importance versus Satisfaction matrix. (Adaptation of Dan Olsen’s framework.)
Risk Distribution Per Category
The specifics of your product category will dictate the amount and size of each risk. Some common examples:
- In a B2C product, usability is a top risk because it is the gate keeper to testing your other risks.
- Consumers are bombarded with too many offers and too much information to give you more than a few seconds.
- If it takes >5 seconds to understand, farewell.
- In a B2Enterprise company where the sale is often not made to the end user, desirability by a decision maker is a top risk.
- The cost of end users suffering is not something that’s easily measured. If it’s not measured, it “doesn’t exist”.
- Which means the CIO/CTO will probably keep buying it as long as the account manager can maintain perceived value.
- In a B2SMB company where there’s so many companies and they come and go so quickly, viability is major risk.
- Startups might be the most excited about your product.
- You’ll have trouble growing unless you have a self sustaining acquisition strategy or you help SMBs avoid death.
- In a deep science company, feasibility and viability are the primary risk if you’ve already proven your technology works.
- Think nuclear fusion and the ability to teleport.
- No, you’re not automatically a deep science startup if you add AI to your name.
Risks are not always independent. For example:
- Value risk can be an inverse relationship with feasibility.
- Eliminate value risk by open sourcing or making it free but significantly increase the business viability risk.
When new companies endeavor to build their first product from scratch there’s so much risk they more often than not explode. But if you largely eliminate desirability, usability, and feasibility risk (product/market fit) you begin scaling to test viability risk (business model fit).
Depending on your plans to grow, what you consider business model fit will vary:
- A lifestyle business is happy to plateau as long as it’s a stable source of income for many years.
- A venture backed company needs to grow 30x to 100x in a matter of ~11 years.
Note: Don’t assume you’ve eliminated Value Risk.
Unless you’re 10x-ing a fundamental human need, like constructing food from raw molecules, don’t assume your product is valuable. No matter how obvious it seems to you.
As an example, a friends of a friend built some extremely impressive AI technology to reduce the need for a nurse do an interview before a doctor appointment. He thought it was a home run. Personally, when I saw it, I thought it was too.
To his dismay, this didn’t even rank in the top 10 problems for hospitals. The cost associated with implementing it and training doctors to interact with it was simply not worth the benefit. There’s a hidden cost that we as builders tend to not see. The cost of implementation, training, and change management.
This cost is so significant and important we see ripples of it throughout many industries. A tragic example of it is the 737 MAX and its two fatal crashes. Boeing knew historically that to sell a new plane they need to be able to say that pilots did not need to be retrained as that incurs significant cost on their clients. Because of this they created an automated MCAS system that would account for the plane’s engines being significantly further forward which changed the mechanics of piloting the plane. As we now know, due to a lack of extensive testing and an off switch for the auto functionality, hundreds of lives were lost.
It’s a terrible example of misaligned incentives between the lives of those passengers and the lengths large corporations go to avoid the costs associated with selling change. When billions of dollars were at stake with the rush and pressure to meet shareholder needs a tragic trade-off was made between safety and profits.
Depending on your desire for growth, as you scale you’ll need to use what you learn to quickly de-risk similar but larger or more lucrative markets. This becomes easier and easier as you get larger though since you have access to so much more information a new company would not have access to.
The art of building a company is finding an irregular combination of risk resolving strategies that shouldn’t be possible but are. Big companies have a hard time letting go of their risk resolving strategies as it’s tough to accept that something that brought so much success can become ineffective.