Grayscale Publishing of Internet products

Source: Internet
Author: User
Keywords Release function shunt can

In the Internet product release process also more use this kind of publishing way: The product release process is not overnight, but gradually expand the scope of users, from the company's internal users-> loyalty higher seed users-> a larger range of active users-> all users. In this process, the product team according to the user's feedback in time to improve product-related functions. This way of publishing, in accordance with Chinese characteristics called "Grayscale release", "Gray volume", "split release."

In the traditional software product release process (for example, Microsoft's Windows7 release process), will generally experience pre-alpha, Alpha, Beta, releases candidate (RC), RTM, general availability or General acceptance (GA) and several other phases (refer to Software release life cycle). It can be seen that the release phase of the traditional software is from the internal-> external small scope of testing > External large-scale Testing-> officially released, the number of users involved is also a gradual process.

In the Internet product release process also more use this kind of publishing way: The product release process is not overnight, but gradually expand the scope of users, from the company's internal users-> loyalty higher seed users-> a larger range of active users-> all users. In this process, the product team according to the user's feedback in time to improve product-related functions. This way of publishing, in accordance with Chinese characteristics called "Grayscale release", "Gray volume", "split release."

The origin of the term "grayscale release" is not explored. Just according to Chinese traditional philosophy, it is in line with the Chinese way of thinking: all things in nature are always in the form of symmetry, complementarity, harmony, such as black and white, yin and yang, positive and negative, blessing and evil. In the two-yuan opposition between the elements of the transition between the stage, the so-called "misfortune and blessing of the lean, Fu Xi Curse of the Volt." Specific to black and white, in the black and white between the middle color-gray. So there's been a lot of gray: grey-box testing, grey management (highly recommended ren: The gray level of management), grey income, gray area, and so on. So for grayscale publishing is actually a process that never publishes, and then gradually transitions to a formal release.

1, grayscale release of the role of early access to user feedback, improve product functionality, improve product quality to allow users to participate in product testing, and enhance interaction with users to reduce product upgrades affected by the user range ... 2, Grayscale release steps

1), define the target

2, selected policies: including user size, release frequency, functional coverage, rollback strategy, Operation strategy, new and old system deployment strategy, etc.

3, filter users: including user characteristics, the number of users, users commonly used functions, user scope, etc.

4, Deployment System: Deployment of new systems, deployment of user Behavior Analysis System (Web analytics), set streaming rules, operational data analysis, triage rules fine-tuning

5, Release Summary: User behavior Analysis Report, user Questionnaire survey, social media opinion collection, form product function Improvement List

6), Product improvement

7), a new round of gray or full release

3, FAQ 3.1, sweeping

1), Problem characteristics:

A, the selected sample is not representative;

b, the sample is representative, but the selection of sample user habits does not cover all core functions
2), the solution:

Sample selection to diversify, the combination of samples covering most of the core functions

3.2), the Curse of knowledge

"The Curse of Knowledge" from the "stick" in the experiment, you can search the specific. We are very familiar with the products we develop, so we take it for granted that users should be able to understand the design ideas of products and the function of products.

1), Problem characteristics:

A, the results are not quantitative means;

b, only rely on the user survey;

C, no web analytics system;

D, operational data is not comprehensive, only core business indicators (such as transaction volume), no user experience indicators

E, analysis of results, only select the information that is favorable to the release, and ignore others
2), the solution:
A, product design to consider product quantitative indicators
b, the result analysis is based on quantitative indicators rather than feelings

3.3), released No Turning back road

1), Problem characteristics:

A, the old and new system user habits are too different, not compatible with the original function

b, the old and new system because the function difference is too big, cannot run in parallel, only forced to upgrade

C, the new system is only the implementation of the old system part of the function, users want to complete the use of all functions, in the new and old system switching

D, the new and old system database data structure is too large to run in parallel

2), the solution:

Pre-product planning focuses on these issues, including:

Rollback scenarios, new and old system compatibility scenarios, user experience consistency, user usage continuity, new and old system data Model compatibility

3.4), the user participation degree is not enough

1), Problem characteristics:

A, expect the user to dig all the features themselves. For a product, most users often only use part of the function, most users are also lazy, do not take the initiative to mining product features
b, a single interactive channel
C, fall into the "Curse of knowledge", do not respect the views of participating users
2), the solution:
A, good to eat crabs sample users, including the user to participate in the test of small rewards (such as MS to participate in WIN7 Test users genuine license), to the user title
B. Interact with users through new media such as mail, forums, communities, blogs, Twitter, etc.
C. provide product function Guide. The latest upgraded functionality in Hotmail Tip,gmail tip has a similar product feature orientation. In the product will prompt similar to: You know, XX also provides XX function, through which you can xx.

4, Grayscale release VS. A/b test

Grayscale published in the Internet companies commonly used A/b test seems to be more similar, foreigners do not seem to have the so-called gray-scale concept. According to the Wikipedia definition of A/b test, A/b test is also called: a/b/n testing, multivariate testing, so in essence grayscale testing can be considered a special case of A/b test. Just for the terminology not to be equated with confusion, to talk about their own understanding of the differences.

Grayscale release is the release of a product gradually expand the use of groups, also known as gray level

A/B testing is focused on selecting the best solution in several scenarios

About A/b test can refer to this article: A/b Test Ultimate Guide

5, Grayscale publishing engine

For a typical small system, you do not need a separate grayscale publishing engine, you can refer to the practice in A/b test to implement streaming rules on page JavaScript or on the server side. But for large Internet applications, a separate release engine for managing user streaming is necessary. "Money dispensers" streaming release mode refers to the original Ali software used in gray-scale publishing engine, design ideas are universal, can be used for reference

6, reference documentation

"Money dispensers" streaming release mode

Baidu Encyclopedia: Grayscale Release

A/B Testing

The ultimate guide to A/b test

Source: http://www.yeeach.com/

Related Article

Contact Us

The content source of this page is from Internet, which doesn't represent Alibaba Cloud's opinion; products and services mentioned on that page don't have any relationship with Alibaba Cloud. If the content of the page makes you feel confusing, please write us an email, we will handle the problem within 5 days after receiving your email.

If you find any instances of plagiarism from the community, please send an email to: info-contact@alibabacloud.com and provide relevant evidence. A staff member will contact you within 5 working days.

A Free Trial That Lets You Build Big!

Start building with 50+ products and up to 12 months usage for Elastic Compute Service

  • Sales Support

    1 on 1 presale consultation

  • After-Sales Support

    24/7 Technical Support 6 Free Tickets per Quarter Faster Response

  • Alibaba Cloud offers highly flexible support services tailored to meet your exact needs.