How do internet products work?

This title has two key contents: "Internet products" and "work ". specifically, the reason why "Internet products" are mentioned is to be different from traditional software products. The "work" here not only refers to development, at the same time, it refers to the entire operation including R & D.

I am not qualified to fully elaborate on such a large set of theories because of my low qualifications. In this article, I only write my own feelings and hope to communicate with colleagues, if it can help you develop projects, it is even more unexpected.

On my Netease blog, I made such an announcement for myself:

"As an Internet product, the most important thing is its continuous update, continuous development, and continuous improvement capabilities.

Never think about launching a nearly perfect product at a time. There are so many demands and changes for Internet users. If you only use traditional software development theories for demand and planning, obviously, this is not suitable.

We can make our products perfect and easy to use only by bravely pushing our products to the market as soon as possible, allowing users to test, find new demands from users' opinions, and find new highlights ."

Internet products mentioned here refer to all software products that can be used only by the Internet in a broad sense. In a narrow sense, you can understand these products as follows: online games, blogs, searches, Im, network Video and so on.

I want to talk about the following aspects:

1. How do internet products meet requirements?

Similar to traditional software, the development of any Internet product must go through the demand analysis phase. I regard the demand here as the basic requirement, which is far from a complete requirement.

I have always opposed to formulating and conceptual the entire process of software development. This is not only difficult to operate, but also easy to get out of reality and fall into blind development: what developers think is a very important feature, users think they are dispensable, and developers turn a blind eye to what users think is important. There are already countless such examples.

The needs of Internet users have two major characteristics: complicated requirements and frequent changes. therefore, if you still use the traditional regular army to deal with Internet users, it will always be carried by users and passive.

Based on this, I believe that the most economical and effective way is to bring products to the market after the basic needs of users are fulfilled, gradually improve the product with the latest user requirements and consolidate the product.

Since we cannot guarantee that we can make a prophet that can predict all user needs, we can at least follow user needs. as long as we can quickly adjust our product direction based on user needs, the user will also be subject to our reverse incentives and will continue to strengthen the product adhesion.

To achieve this, we must consider setting up such a framework in advance to meet the ever-changing user needs.

If you still feel dizzy and cannot grasp the user's needs, there is a minimum principle that you can follow: Never think about killing any small function limitation, but consider that it may change, there is a possibility of adjustment. We need to consider this situation and leave room for future expansion.

Of course, we still cannot guarantee that a framework with super scalability can be created from the very beginning. The establishment of this framework is gradually being improved.

Another small principle can be considered: If your development cycle is more than half a year or more, you can consider adjusting any demand on the Internet, because no one can predict what the situation will be like for half a year or even one year later, maybe this market will no longer exist by then.

2. How to develop Internet products?

For R & D technicians, we should consider not only meeting the existing requirements, but also considering future system expansion and leaving space for expansion. if you do not consider expansion now, you will pay much more in terms of expansion in the future.

However, we should not fall into the blind expansion situation. If that happens, you will find that your work is no longer available. The expansion we can consider is from now on to a very short time, the requirement for expansion has been determined. there is no need to think about the future, and you cannot think about it.

In fact, whether it is demand analysis at the project management level or technical R & D at the specific implementation level, as long as you have this concept, how can you grasp it in each specific project, you have to rely entirely on yourself.

Speaking of specific R & D methods, I have also recommended a development method we adopt to others: scrum. this development method is just a secret to the original spirit of Internet development. However, whether or not this development method can be used by everyone depends on whether the R & D team is suitable, this development method requires the R & D team to have strong self-management capabilities, strong learning capabilities, and a solid professional foundation.

"Quick R & D and quick launch", all around it. never build a car behind closed doors. A product has been developed for a year or two. When the product comes out, the market is no longer that market, and users are no longer those users.

However, in other words, for a team with poor professional foundation and poor development experience, any development method may not be of great help to them, what they need to do is to strengthen themselves, rather than looking for a panacea that can solve their worries.