Ali enterprises are always busy , I have the honor to participate “ The whole process of mobile terminal from online to iterative ” Analysis of the duplicate disk of , Come and have a look ,enjoy~


1.  Sort out the idea of the second round - Avoid building cars behind closed doors

Don't take action when you have a task , Please make your own plan before action ! Do one thing blindly , Practice makes perfect . Not suitable for changing needs , Not from “ Toolmaker stage ”, Grow up to be an independent thinker “ Product owner ”.

The first time to receive the re disk task , I didn't do it as usual .“ replay ” It's different from other clear tasks , Just follow the process . Behind it may be that the leader simply wants me to do such a thing , Maybe it's through this thing , To test whether my handling of such matters has met my expectations , Or I want to know if I have the ability to play my own role in the next stage .

therefore , Alumni in the product manager community , After the communication between teachers , Let's start to have rules and methods to deal with this matter , I would like to thank the students and teachers in the community who answered questions during this period .

1.1  Reasons for mining a second disk

What is the reason for the resumption , To whom , What is expected ?

Eliminate blind thinking , Ask the superior reasonably , Intent to quickly implement requirements , Unified thinking ! Adoption and leader Direct communication , I learned that the resumption was a temporary decision ( Maybe there's a reason ), On the one hand, show leaders , On the other hand, it can be used for department reflection .

* Review project for leaders , Reporting ,
* Resume project for department meeting , Reflection work .
1.2  Thoughts on sorting out and restoring

What results do you want to achieve , What materials need to be prepared , General process ?

Report to leaders , To refine ! Originally used Excel Table by table project details , But considering the reporting time and scenario priority , We chose to adopt PPT Form of , Simple and quick report “ Project overview ”,“ Replay results ” and “ Later planning ” These three points , Reporting time is controlled at 30min within .

* Review project for leaders , Reporting .
time -30min within

place - Leading office

scene - Take it PPT Propaganda

expect - The leader was very moved , Recognized the work of the Department ( promotion increase )

Meeting with colleagues , To be grounded ! Rigid meetings like those on TV are not in line with our team atmosphere , And most of them are R & D students , So we use Excel Form a list of items to meet , Simple and direct connection to specific details and R & D students . It is mainly divided into four steps “ Affirm the achievements in the project ”,“ Point out the deficiencies in the project ”,“ Review the project history together ” and “ Discuss specific solutions ”.( A meeting is not an arrow with feathers , We all like to give a candy bar first and then hit it lightly , So I'll talk about it first, then point out the problem, and my colleagues will give me the solution , So that the work can be carried out easily , Become a qualified team manager .)

* Resume project for department meeting , Reflection work .
time - unlimited

place - Conference Room

scene - As per project list , Talk about the recent situation in detail

expect - During project sharing , Summarize potential problems , Make the future project implementation process better

2. Implement key processes and outputs

2.1  Version management

According to the version , R & D easy meeting , Leaders are easy to understand ! one side , Long project cycle, hard to remember details , For the first time in a year , It's normal for R & D students not to recall specific points ; on the other hand , Leaders don't care what's in each version , List version number and main functions .

(1) Online version (0-1)

This step lists the online version (0-1) Main function items in .

* What are the function modules ?
* What was the expectation then ?
* Whether it meets the expectation ?
* ……
(2) Iteration version (1-100)

* What are the function modules ?
* What was the expectation then ?
* Whether it meets the expectation ?
* ……

It is worth mentioning that , I didn't know until I made a replay , Good housekeeping habits are really important . Because the project is in a hurry , A lot of information is not updated to the demand pool in time , Weekly, daily and even SVN and GIT There was no comment on the requirements for more details , It has made a certain impact on the data check of this round up .

Possible documents : Version iteration record , Demand pool , Product requirements document , Test cases, etc

If you also have agile development ( Catch up ) Situation of , Some data are not right . You can go to R & D students for help , such as : daily , weekly ,GIT,SVN, There will be records on these .

2.2  Project management

ahead “ Version management ”, It's for meetings . Next, this “ Project management ”, It's time to have a meeting with the Department , Focus on it , Let's see the document , So we have to do it more carefully .

(1) Estimated time and online time

The purpose here is to give feedback “ Progress ”! Online time is always delayed , It's been discussed many times, but it doesn't solve the practical problems , So take this opportunity , Let's talk about it , benefit by mutual discussion , Think about measures .( When reporting to leaders , This data can be modified as appropriate , It will be uncomfortable for the leader to see it all postponed , So give it to the superior in advance , Make everyone comfortable )

About progress management , I quite agree “ Opinions of Senior Product Manager of ”:

Today's agile development really can't guarantee the result in the planning stage ‘ quality ’.

1. We need to / Function split , Confirm whether it is a short-term project or a long-term project , Clear priorities and ‘ Key components ’;

2. Schedule requirements after splitting ( Design , development , Test each stage ),‘ Key components ’ and ‘ Preconditions ’;

3. If it is a short and smooth project, it should be carried out every day 20 A station meeting in about minutes , If it's a long-term project, hold a weekly meeting ;

4. Follow up project progress in real time , grasp ‘ Key components ’ and ‘ Preconditions ’ essential factor ;

5. Communicate well , Reporting , Wording needs to be precise , clear , Don't show up ‘ ready to accept either course ’ The answer to .

(2) Function overview and details

The purpose here is to help colleagues in the Department , Recall follow up XX Process in demand , Step by step reflection and summary “ Difficulties encountered ”,“ Interim solutions ” and “ Better solutions .”

meanwhile , Let's evaluate whether the function achieves the expected effect , Review the problems encountered in the R & D process , Solutions and suggestions , Lay a foundation for dealing with similar problems in the future .

(3) Task category and priority

This step , It's for summary .

Task category
We can see where our R & D energy is mainly invested , Classified as “BUG repair ”,“ New features ”,“ Functional iteration ”,“ Interface optimization ” etc .( After all, it's always fixed BUG It's a very problematic thing , We need to know where we're going )

Priority is a matter of opinion . it is said that , look for 10 A woman can't have a baby in a month . So does the project , Row 10 The highest priority cannot be achieved in a day .

For prioritization , There are too many methods and conclusions . But let's think about it , Why prioritize . Let's say we have a demand , There was an urgent need , Priority is needed ! Usually do it according to the priority or work overtime , But it's not necessary .

For urgent needs , Students of general products can think of “ Replace low priority requirements ( Non urgent needs )” perhaps “ Overtime completion ”. In Hezhong soft 10 Senior Product Manager , in my submission “ Urgent needs ” The priority of can be split , Don't stop analyzing him because it's urgent , Let's analyze this “ Urgent needs ”, What's the priority , Is there an alternative . One level one level split down , And the natural plan came out . Can greatly avoid meaningless overtime .

3.  On the methodology of the double play

3.1  Dai Minghuan (PDCA)

Dai Minghuan (PDCA) method

P(Plan) plan : Target prediction and determination of product reliability , Reliable planning and determination , Reliable organization and division of labor

D(Do) implement : Reliable task motivation , Command and Implementation

C(Check) Check : Evaluation and evaluation of product reliability , Reliable operation control and audit

A(Adjust) correct : Looking for improvement plan to make the next plan more perfect

From Wikipedia -2018 year 6 month 19 Revised on

3.2 Six ways (5W1H)

Six ways (5W1H,6W)

who (Who)

why? (What)

when (When)

Where (Where)

How to understand (Why)

how (hoW)

Tip: You can call it anything ,5W1H and 6W The difference is right “how” It's a different definition , Some people prefer “How” So it's called 5W1H, Some people prefer “hoW” So it's called 6W.

From Wikipedia -2019 year 9 month 2 Revised on

3.2  Your own summary of the plan

The best way is to suit yourself , Otherwise, it's Handan , the loss outweighs the gain .

The way you walked , Books read , Scenery seen , Loved ones are quietly guiding you in your direction .



above , yes “ Wood depth ” Recently, I made a summary of my own project's re analysis work , Yes “ Duplicate analysis ” There are some new conclusions and Reflections , Share with you after finishing , Hope to have small partners to exchange and study together .

©2019-2020 Toolsou All rights reserved,
First knowledge python Skills summary GDOI2019 travels c Linguistic 5 Three common sorting methods Python Basic knowledge and notes " Cephalosporin wine Say go and go "? DANGER ! Don't drink alcohol when taking these drugs Thorough explanation from Zhongtai vue The component page height is adaptive according to the screen size Classical algorithm - recursion ( The case of raw rabbit ) Big data tells you , How tired are Chinese women about keras use fit_generator Encountered in StopIteration