Shulou(Shulou.com)06/03 Report--

"Isn't it just a small request? What took you so long? "

"Your demand is not listed in the App, so it will not be approved."

"The requirements document written by your product is not the effect we want. What we value is..."

"Queue up, queue up, there are not enough R & D people, and the queue is already in the middle of next month. "

"The function test has passed, but the App has no plan to launch in the near future. It will wait until the beginning of next month."

"In special periods, APP containing this content will not be approved."

........

It's too difficult. How many difficulties and dangers does a demand have to go through from proposal to project approval, to R & D completion, to official launch?

More heart-breaking than the long online process is the function that is not easy to go online, missed the best window period, and did not achieve the desired effect.

As shown in this flowchart, even if a function is launched, even if the process is streamlined, it is difficult to upgrade it due to the organizational structure and software architecture. Accompanied by complex processes are increasingly rich business scenarios and demands:

Business segment: The market is changing, and it is necessary to speed up the response speed of demand and keep up with market hot spots in time.

Operation Department: There are thousands of operation theories, and App cannot obtain user portraits, so it cannot support refined operation at all; and the iteration speed of products seriously limits the launch of operation plan.

Product department: The demand is like a tide, and the priority ranking that cannot be handled is endless. It blindly follows the business demand, which leads to more redundancy and lack of bright spots in the App

R & D department: scheduling, lack of people, rush work, constantly do demand, there is no extra energy to review code, optimize architecture, improve performance

Is this really a dead knot? Is there no way to satisfy multiple demands simultaneously?

Yes! Try a greyscale release!

What is Grayscale Publishing?

According to the traditional definition, grayscale publishing refers to a publishing method that can smoothly transition between black and white. A/B testing can be carried out on it, that is, let some users continue to use product feature A, and some users start to use product feature B. If users have no objection to B, then gradually expand the scope and migrate all users to B. (From Wikipedia)

Today, when we pay attention to the concept of "growth hacker" and need refined operation, grayscale publishing is not only a publishing method, it can:

Quickly validate marketing plans or product ideas and achieve more attempts at lower cost

Provide numerical proof for new functional verification, replacing subjective judgment with scientific experiments

Advance "pilot" new function, through experimental group comparison, get more clear what to do next

"Finding out" User feedback that will be unpopular features "stifle" in the cradle

Shorten R & D testing processes and accelerate team agility

In fact, although grayscale publishing looks "wonderful," in the real world, limited by the complexity of the grayscale experimental design process, few teams can improve the overall conversion of the App through grayscale.

The difficulties include:

Because the gray scale needs the experimental group and the normal group, the design of the experimental group cannot directly reduce the workload of R & D.

It is difficult to obtain effective, true and timely user feedback after the App lacks "data grip" and gray scale is put in.

There are very few options for grayscale publishing, and the App itself can only achieve uncontrollable grayscale ranges through limited tools.

Grayscale is more suitable for limited small functions, such as button color and text fine-tuning, which can only improve the conversion rate of some services.

For example, the gray scale of large version is limited by tools, and it enters the endless loop that is difficult to control the gray scale range.

How to make a good gray release?

Let's review the difficulties in grayscale publishing. To sum up, it is nothing more than:

Grayscale functions also need to be developed, corresponding to a complete development process, which seems inevitable.

Grayscale publishing requires more rigorous scheme design, aiming at what people, what content, how to analyze and apply the analysis results, which requires people who are proficient in business, understand products and users to carefully design.

Grayscale release coverage should be more accurate, coverage methods must be diverse and can meet the needs of gray scheme

The data after grayscale release must be properly recorded

Grayscale publishing should not be limited to small functions, but only for H5 pages, which cannot meet larger and richer business scenarios. All these can be solved in Fantai Mini programs open platform.

Let's take a look at the design and development process of the entire grayscale release:

Fantai Mini programs open platform is dedicated to helping enterprises "fragment" App, breaking the original "indivisible" whole into functional modules composed of Mini programs one after another. Through Mini programs, enterprises can split different business modules, each business module updates and iterates by itself, and is no longer limited to the overall App release process. Business flexibility and development agility will be greatly improved.

In the grayscale publishing process, once the business personnel put forward the corresponding business requirements, they can immediately put into the corresponding R & D process. The advantages of this process are:

Business functions (Mini programs) are independently developed internally, so there is no need to take up too much development manpower, and the demand response time will be greatly shortened

Mini programs only need to complete internal testing and review. The hot update mechanism of Mini programs can bypass the long review cycle of App application market and quickly complete the function launch.

The design and implementation of gray scale experiments are changed from developers to business personnel. Business personnel design gray scale ranges and gray scale rules by themselves, which will be able to complete the setting and practice of experimental groups at a higher level.

Agile iteration of functionality and business is implemented, and good ideas are not obscured. The next business explosion point may be hidden behind this small idea.

Fantai Mini programs open platform gray distribution composition

1. Rich and visual rule configuration to meet personalized publishing settings

Fantai Mini programs open platform visualizes complex rule configuration, and personalized business rule setting can be completed in just a few simple steps. These settings can be the most common user list, age, city; can also be some technical indicators, such as mobile phone model, system version; more free to add rules with business characteristics, such as whether it is a premium paying user, whether it is associated with other products of the company, etc.

2. Simple grayscale release creation process, so that the grayscale experiment gets the best embodiment

Based on the rule base, Fantai abstracts the most complete, easy-to-use and scientific grayscale publishing scheme creation process according to various grayscale publishing scenarios. In this process, business personnel select versions of Mini programs that have been developed and tested, and select corresponding release time windows, rules, rules between rules, and miss handling according to the gray scheme.

Once the grayscale scheme is successfully created, it can be synchronized to the App in real time without waiting for the function review again. For the release with extremely high timeliness requirements, the most direct and real-time data feedback can be obtained. Mini programs can also be "rolled back" to existing versions in the event of bad feedback.

3. SDK Sandbox automatically reports relevant data to achieve a complete closed loop of grayscale release

There is no need to carry out business buried point development for each Mini programs. The SDK of Mini programs released by Fantai will automatically collect and report some common data during runtime; meanwhile, for Mini programs requiring complex business data return, since the SDK has a standardized data reporting protocol, only a small amount of development is needed to realize the most accurate data reporting and return.

Try it! With Fantai Mini programs open platform, realize all your ideas about "thousands of faces"!

Want to know more? You can pay attention to Fantai geek public number

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.