RETHINKDB founder teaches you how to build a great product

About the author

My name is Slava Akhmechet, I am one of Rethinkdb's founders, RETHINKDB is an open source, distributed database designed to help developers and operators to process unstructured data while building real-time applications

How to build a great product

If you agree with the principle of "selling the Universe", then if I say that most startups have already sold their money to the market before selling it for sale, Yuan Fang, what do you think of it, drubbing? If you're looking for an explosive, fast-growing market, you need to find a way to build a great product, Yuan Fang, what do you think? [1]

Building a great product is difficult, and if you don't have a good way to analyze the product and function a workable model, it makes the difficulty even worse. Without this model, you will never cease to be trapped in the vortex of the functional point of the product, which is like groping in the dark to shoot with feelings. Of course not rule out that a small number of people can hit the target success, because they have a different from the beginning of the normal product intuition. However, most of us ordinary people do not have this ability in the first place.

I started out with a very bad intuition about the product (and worse, I didn't realize it). Over the past 3 years, when I stared at our user matrix every day, trying to create a feedback loop (feedback loop) to train our brains to find out how to build a good product. Fortunately eventually I finally became very good at predicting the impact of a functional point on the product, so I started thinking about building a model to try to capture the knowledge that I had learned to share with you.

Three barrels of wood models

The most important aspect of product management is to classify the function of a product into 3 different barrels: subversive function (which is enough to change the game's rules), indispensable function (the function that is necessary but does not bring great benefits), and insignificant function (which can only make the user "Oh" a sound function). When I first started trying to start a business, all the features of the product were almost as important to my eyes. But as time went by, the "three barrels model" was gradually shaping up in my head, and today, I have been able to classify it automatically in the 3 barrels mentioned above in every function I have encountered.

An example is described. If you're building a new mobile phone product right now. First of all, the phone must be able to call, otherwise you can not be called as a mobile phone, the ghost will buy. But on the other hand-people will not buy your phone because of the phone call function, because which phone on the market does not provide phone function. So in this category, the ability to call this function in your product should be placed in the "indispensable" barrel inside.

On the other hand, if your mobile phone product can also project the video being played, this other phone does not have this function, so it should be placed in the "subversive" bucket, because this feature will make most users scream. But the other possibility is that most users may not be interested in this function, so the function should be in the "no-no" barrel.

The above example shows you the 3 barrels that a given function should be categorized into:

• Disruptive features : Users will be happy to pay for the functionality of the package because of this feature.
• indispensable function : The user does not purchase because you have the function, but he certainly will not buy it.
• innocuous function : This function and whether the user choose your product relationship is not very small.

Practice has proved that a successful product will have one to three subversive functions, dozens of of which belong to the golden mean function of not or go, and very few insignificant function. As an entrepreneur, the thing you need to do is to build a keen intuition to categorize different functions. Although sometimes the situation is still very subtle (how can an embedded mobile projection module be a disruptive feature or a trivial feature?) , however, at least the "three barrels model" gives you an aggressive plan that can be implemented.

Three barrels How to guide the allocation of resources

If you have enough time to waste, of course you can also ignore the "three barrels model" mentioned in the three product functional categories, and through the endless blind iteration to obtain the market recognition. But you don't have that much time to splurge. The longer you need to build a great product, the easier it is for you to burn your money, the easier it is to blow your morale down, or the easier it is to see the opportunities in the market sneak away under your feet. The product management model brought by the "three barrels model" is very valuable here, because it allows you to really treat product management as a resource allocation problem.

If you put too much resources into the function of innocuous categories, you are wasting resources. This is obvious.

If you devote too much of your resources to functions that are more than necessary and indispensable, you are wasting resources! When the iphone first generation just came out of the lack of copy and paste features, this feature may be an indispensable function for some users, but Apple is very right to decide that even if the feature is missing, there will be enough users to buy their products, so there is no need to postpone the release of IPhone1.

You are wasting resources if you continue to invest in resources when an indispensable category of functionality has reached a predetermined level of acceptability! IPhone1 just came out when the call quality is actually more fucked, but enough to be able to talk. Most users choose to accept. It can call, and there is no way to hear how bad the other person is talking. If you were to increase the quality of the call 10%, it is not very important to enhance the user's purchasing degree.

If you're developing more than 3 disruptive features at the same time, you're wasting resources! Practice has shown that almost no disruptive product is good at dealing with a lot of problems. It is difficult to deliver disruptive features. 3 This level of functionality may be the limit you can accept, even if your product can have a large amount of subversive features.

Finally, if you don't pour enough creative resources into the set of disruptive features, you're wasting resources! However, if a disruptive feature doesn't embarrass the user, it's unlikely to be a disruptive feature-it's just a trivial function. In this case, you don't need to compromise to waste resources.

It is acceptable to make some mistakes in the process of making products. Not many products have made mistakes before they are actually accepted by the user. But most of the first-time product managers are constantly breaking the rules that must be followed for products, perhaps because they are unaware of the existence of these rules. Then your own kind of bitter fruit can only be swallowed by yourself. The less mistakes you make than your competitors, the more chances you get to win. The cost of every mistake can be incredibly huge. If you make these mistakes enough, your opponent can easily throw you a few blocks.

Tips for driving three barrels of wood models

To build a product the most test is the need to learn exactly how to put a product's various functional points into the "three barrels model" inside the three barrels inside, and then know when these barrels are already full. Back to an example mentioned before, whether an embedded projection module opponent Machine products is a disruptive function or irrelevant function? If it is subversive, is this feature enough to capture the hearts of a large number of users? Or do you need to combine another disruptive feature to do this? If you increase your phone's call quality by 50 percentage points at this point, will this improvement be a disruptive upgrade? Or is it still just an essential level? What about the 200 percent increase? And what exactly do you need in order to create an eye-catching product with the number of functional points in the bucket that is indispensable?

Because I'm not familiar with the mobile phone market, I can't give answers to these questions. If I'm in the industry of unstructured data I'm good at, I can easily classify any of the functions I've encountered into their corresponding barrels. Although I sometimes make mistakes, it's not a big problem, as long as I make fewer mistakes than my competitors.

The way to build this intuition is to talk to a lot of people. To chat with potential users. What do they think of it? Go and talk to people in your circle who have failed to start a business. What experience can you gain from their failure? Go and talk to your competitors. How do they solve the problems they encounter? Talk to the engineers at the big company. What information will they tell you about the current state of technology? Talk to other entrepreneurs in similar industries, talk to investors, talk to reporters, talk to students, talk to professors, and even talk to people who disagree. The best way to get sensitive touch in a circle is to plunge yourself into the industry and talk to as many people as you can reach.

Instead, opinions

The sooner you get acquainted with the history of your industry, the technical status quo, the ideas of potential users, and the direction of your competitors, the sooner you can paint a clear blueprint for the industry, the sooner you can create a unique vision for your product. But it is important to be careful that when discussing with different people, sometimes it is easy to adopt some wrong people's advice.

If you decide to design your mobile phone in the form of a walkie-talkie, the target customer base is the construction worker, and the best way you decide to sell out is to get the building manager in the top-down way. If you talk to the builders first, maybe they will be attracted by the fancy icons and a unique color shell of the walkie-talkie, and you'll think that these unique designs should be disruptive in your mobile phone product. But don't forget, buy you this walkie-talkie is not the construction workers to decide, but contractor, is the building manager and the right to write your cheque oh. The beautiful design is good for contractor, but it's not a disruptive feature for him. Because these beautiful icons did not help him better than before to push his career to another stage.

In complex business marketing, you have to be careful about the situation of each group and serve all stakeholders to make them happy. If construction workers have a huge impact on contractor's decisions, it may not be a bad thing to spend more time on these unique shell and icon designs. Otherwise, you will be wasting resources.

This needs to take into account the situation of various groups and stakeholders in fact, even in the general consumer goods also exist. If you're designing a luxury phone, and it's more expensive than any other phone in the market, you need to think about whether your client needs to persuade his wife to buy the luxury goods. Is it common for most families to decide whether to buy luxury goods? Or are the family members individually deciding whether to make a purchase or not? If they need to persuade their wives to make a purchase, have you ever considered adding an extra feature to make the wives easier to accept and to be persuaded? Think about it!

Need to be careful with noise. Learn to identify those who will buy your product, and those who are only commenting on your product. It is tempting to design a product based on feedback from industry authorities on the functional points of the product. But a product feature that becomes disruptive is because the user who buys the wallet thinks it is really a disruptive feature, otherwise, it can only be considered a trivial function. Industry pundits can play an important role in knowing what's going on in the industry, but they're not always the people who buy your product. If you build your product entirely around their feedback, you will eventually find that your product is not selling at all.

It can be seen that you cannot create a great product without experiencing the same kind of life, eating, and breathing as your target users. You need to pinpoint exactly what your users are, what they are going through, what they think about your product, and who will help them make a purchase decision. Your intuition needs to reflect exactly how your customers perceive your product. The functional classification of the "three barrels model" only works if you can correctly predict the response of your real target customer to that function point, or you are not sure which bucket the function point should be in, then you will be wasting your life.

Disruptive features of aggregation are undesirable

Here we begin to discuss some of the subtleties of the "three cask model" we have not discussed before. Some of the independent function points themselves are not attractive enough, but once they are aggregated with other functional points, they become subversive functions immediately. For example, suppose you are now designing a set of pictures for your mobile phone product. Will this be a disruptive feature point? Most likely not. What about a separate set of unique, beautiful color combination templates? This does not look like a disruptive feature. What about a unique type of phone case? It's hard to imagine a person going to buy a cell phone because of a beautiful phone case. [2] This is not the maiduhuanzhu of the foreign article? So what if you put all of the above features in the same product? A unique design direction came out, a novel icon design, a unique color combination template, matching the distinctive mobile phone shell such an aggregation function sounds enough to become a disruptive feature to attract enough consumers.

It is a very dangerous choice to create an aggregated subversive function by aggregating the function points for three reasons:

• First, it's hard to choose which feature points are aggregated or not a disruptive feature point.
• Second, the cost of achieving the disruptive functional points of aggregation is very expensive-you now need to make a number of correct decisions for several function points, rather than just making the right decisions for a disruptive function point.
• Finally, it makes it easy to convince yourself that if you add a function point again, you will succeed in getting a disruptive aggregation function.

So it's hard to build a great product, and now it's compounded by these subtle situations.

There is no denying that many products have been successful in this way, but if possible, try to avoid this situation. If you really have no choice but to use aggregation function points to form a disruptive combination of functions, it is likely to indicate that you are in the market is actually very mature. Well, there's no way to do this, but maybe it's time for you to have a deep self-reflection-is this really a market worth investing in? Or are you supposed to find the next market where you can make innovation relatively easier?

Product Mission Verification method

Assuming that you can now use your established product intuition to apply the "three bucket model" to your industry, you can easily (and correctly) classify the function points into different barrels, and you are already better than most product managers. But that's not enough. You need to know that this method is actually flawed in some way:

• If you are simply classifying functional points by a speculative intuition, you will easily be persuaded by the rhetoric of another voice in your mind to make the right decision after making a mistake.
• When you build the product, you are bound to hands every big and small decision, because others simply have nothing to guide them.
• Your siege Lions will be frustrated because they will assume that your decision to speculate is unfounded.
• Before the product comes out you will need a lot of other people to help you-journalists, investors, potential employees, and customers. It will be very difficult to persuade these people, because the decisions you make are intuitive.

A good way to solve these problems is to define your product mission. You can think of it as a function that accepts a function point as a parameter and then returns a value that is one of the three casks. The definition of a good function must be concise, readable, and reusable. Ideally, after reading this product mission, most members of your team will be able to classify function points into the corresponding barrels as you would.

Here's what we think of as a productive, humorous product mission definition for RETHINKDB:

Database Tools should be the same as Magic
Provide developers with the coveted development tools to help them build real-time and data-driven Web applications to surprise and wonder. And to create a great software PRODUCT process to bring pure interest and simplification.

On the surface, these two sentences actually seem to be saying nothing, but if you dig deeper, this product mission is actually full of information. It tells people that what we're building is a database product; It tells people that we first treat the product as a development tool, which eases the tensions between developer-related features (such as the query language of the database) and functions related to software operation (such as monitoring). Be aware that all of our disruptive capabilities revolve around solving developers ' problems. We consider the function point of how the database should be run as an indispensable function. The mission of this product resolves our expectations of how users should use RETHINKDB to build real-time, data-driven Web applications. This product mission also gives you an idea of how much we can achieve (surprise and magic) for a particular feature. It's not enough for developers to be good enough. These people spend a lot of time on our products every day-we need to make them feel good in the process. The product's mission also implies that we will accept more complex implementations to make life easier for our users. It leads us to the ability for developers to write new, rather than already existing, applications. It has a sense of self and embodies a healthy sense of humor that we have as a team. This gives the public a general understanding of what kind of team we are.

With this product mission, we can use it to verify the various feature points proposed. On top of that, with some team-sharing knowledge, our team members are able to classify function points independently in a broadly consistent way.

It took us three years to figure out what kind of plane we were doing before thinking about the product's mission. If we were to define this mission on the first day, our development time could be half-reduced, maybe more. When you build a product, the product mission should be the first thing you need to fix. If you have a mental model that is good enough to write a product mission that will inspire all of your company, other things will naturally be solved.

[1] What I mean here is not to imply that finding a good market is easier than building a product. In fact, the two should in turn, compared to finding a good market, product management is quite easy to do, so here I plan to solve this problem first. To be successful, there are many other aspects, such as publishing, economics, regulated markets, and others, besides the need for a great product and a fast-growing market. But the failure of a start-up company for these reasons is dwarfed by the fact that the company has been killed because of the choice of a niche market or the lack of a timely way to market a great product.

[2] What happens in reality is that people do buy the unique colors and skins of their phones. But here we overlook this subtle aspect, and let's focus on the big things first.
——— finished ———-

English Original: http://www.defmacro.org/2013/09/26/products.html
Make/Translator: Slava akhmechet/Heaven Zhuhai Branch Rudder
Public Number: Techgogogo
Csdn:http://blog.csdn.net/zhubaitian

RETHINKDB founder teaches you how to build a great product