A Guide to Cloud design

[Guide] Improve the user's use frequency, lengthen the use of a single time. If you can not do this two points, then you decorate the cloud is a disturbance, or even a user experience of the destruction, then please do not decorate the cloud.

The "cloud" classification is divided into two major schools, one is divided into three categories by service type: IAAS (Infrastructure cloud, typical Amazon AWS), PAAS (platform cloud, typical Google App Engine), SAAS (software cloud, typically Salesforce.com). The other is divided into three categories: the public cloud, the private cloud and the mixed cloud.

In fact, both of these methods belong to the "hindsight", both of which can let us analyze the experience of others, but can not let us guide the product Manager to design the future "cloud" products. So in kind I introduce a new classification method, which is design-oriented, and the goal of this approach is to help the product plan the cloud at the design stage.

I divided the "cloud" into two categories, called "Push Cloud", is the product built-in, by the background "push" to the user. A simple example of "push Cloud" is the micro-letter "plugin", which constantly pushes new functionality to the user. The other is "self-born Cloud", is the customer's initiative to contribute to accumulate. A simple example of "self-generated cloud" is the "software score" in 360, how many people download, update, and how many people unload the data, all of which come from the user's own accumulation.

With these two concepts in mind, product managers have two issues to think about when planning a product:

1. When you have a certain installed capacity, in the future to push what to customers?

2. To "motivate" users to contribute, accumulate what data?

"Push Cloud"

When the product has a certain installed capacity, we must take advantage of this Hard-won foundation, the establishment of beachhead, to quickly seize the fort, expand the outcome. As Tencent has QQ, Qihoo has 360, in the user terminal is only the first step. This requires the "cloud" to be laid out in advance. There are three kinds of clouds:

Resource clouds, storage space, bandwidth, and computing power are three typical resource clouds. If your product is very consumption of traffic, or occupy space, you are blessed, try these three kinds of clouds! Initial space 1G, not enough? Pro, can add to 10G Oh! " Too slow? Pro, proprietary acceleration channel, 5 times times faster Oh! " 3D rendering to wait for a long time? Pro, high speed CPU exclusive, shortened by 70%. "Ripe unfamiliar?" What is the cow x product?

Application cloud, what other features will users use before, during, and after using your product? Pick out some of these features and add them to your product incrementally through "plugins" and "markets." When we use the browser, we have a lot to do while surfing the Internet: Listen to music, write the schedule, send and receive mails, send micro bo, download things, anti-virus trojan, the content of the Web site to the other places ..., do these things we will use other software, why can't we develop these functions into plug-ins? Let the user do all these things without leaving the browser-this is the Golden Hill "Cheetah" dry method.

Data cloud, what kind of data is useful to customers? Try to put this data into the cloud. If you want to do a navigation software, preset into "Real-time traffic information" is very valuable. If you want to do an EHR, preset into the job description, the position of the salary data, evaluation indicators, the HR manager will be ecstatic. If you want to do a invoicing software, preset into the social commodity database, the user can be "ready to use", do not have to go through the "warehousing, input merchandise information" This cumbersome procedures, how many store managers will be grateful to you.

"Self-born cloud"

As long as your product is in use, will continue to accumulate data, the problem is that some data is rubbish, accumulated no more useless. What kind of data is useful and valuable? How to motivate users to provide the data we want? These two issues are considered good when designing a product.

What data do you want? The more you can "map real life", the closer the "money" is, the more valuable the data is, the less valuable it will be. For example, just record the user "like" what movie, it is better to record the user "what movie theater to see what movies", if you can record the user "buy movie ticket" information, it is fantastic! See here you will find "can" recorded what data, in fact, at the beginning of the software design decided: "Like" is the film review class, " Cinema "is the registration category," Buy a Ticket "is the electric business class. Follow this idea, you can grasp the direction of product design. Try to avoid "there is no (valid) user data" products, for example, most of the reader, the electric tour will almost accumulate no data, do this kind of product risk is very high, there is a "taxi small secret" application, let users provide "here at this time the difficulty of a taxi", This data accumulates to answer a question: at what time and where is more difficult to take a taxi, this data has little value.

How to motivate users to provide data? " Location Check-in "is a high value data, but the" cut off "but play. The reason is that the user's "incentive" is insufficient, you give me a lot of strange medals why? What's the use? " The original design is to allow businesses to do "precision promotion", a customer for the first time, send him a trial install it, the other one came five times, give him a 80 percent! Unfortunately, offline business integration does not give the force, too little promotion, users quickly lose the power to sign in, product failure. "Baby whole Plan" has a function, enter your baby's height and weight, it will give you a curve, so that you can clearly see compared with the same month-old baby: Your baby than how many babies heavier, than how many babies high. A lot of parents are willing to go in and see, the more people willing to try, the more accurate the curve, this is a forward loop.

What is the purpose of the cloud layout?

Word: Increase the frequency of user use, the longer one time use. If you do not have these two points, you decorate the cloud is a disturbance, a user experience damage, do not decorate the cloud.

Let me end this article with an application of an alarm clock. Imagine an alarm clock, we use day and night, one time, half a minute, not even look at it, also accumulated no data, so the alarm clock this product is a "small and tasteless" products. Can "cloud"? Let's try it.

Alarm clock To set the time, "set" this action in addition to their own, and "who" has the power to set the alarm for you? Aha, your boyfriend has this motivation, so we add the "remote setting" feature to the alarm clock, so if you authorize it, your boyfriend can start setting the alarm for you every night. Tell you to get up, you can not use the system bell, every day in the clouds Pick a love song, send a sweet morning message how? Get up, your boyfriend wants to remind you of something? Lunch date, Evening Hi song, your anniversary ... Let him set the bar: "Pro, restaurant booked, old seat, 12:00 noon waiting for you." "Some things may be you want to remind him, but also for him to set a alarm:" Kiss, Tomorrow is your father's birthday, gifts I help you ready, remember to bring home AH! " Boyfriend with gifts home, dad very happy, very like, so take a photo sent back to you, thank you for your understanding, by the way sync to micro Bo, bask in your sweet.

Look, the alarm clock has become a private and strong social application, songs, sharing, photography, filters, calendars, spaces ... You can continue to push the new cloud, and as lovers, friends increasingly close interaction, will accumulate more and more valuable from the birth of the cloud. Think about how many times a day would you use if you really had an "alarm clock"? How long do you spend each time?