By the previous section, we can know, the estimated delivery time, delivery costs, profits obtained on whether the project may have a significant impact on the ground.
Agile estimate basis:
- Why should estimate: estimates may allow the team to understand project specifications to calculate ROI and IRR, form the basis of an executable permission, with the estimates, the market can advance to prepare for post-market;
- Who performed the estimate? : Product Owner, Agile Coach;
- When meeting? Nature is the sooner the better, between the whole project, also with the gradual improvement more information, estimates should continue. Agile advocate: embrace change, embrace change that since estimates have to do adjustments, plus the manpower necessary to add staff. Do not blindly count on overtime to reduce costs, as after 90, 00 after entering the market, this generation pay more attention to the quality of life, after all, work to live; do not yourselves, do not say the exploitation of high-sounding;
- How to estimate create? There are a number of ways: from the various dimensions, such as time aspects, labor, materials and other aspects;
- Relative Size: mainly used for user stories;
- Point value: delivering value and results. Only deliver software available to maximize the interests of customers, customers will pay the balance due us.
Project size measurement: by the amount of code, the time dimension, the human dimension, the functional complexity of several considerations.
- Story point: relative measurement is relatively independent, compared with each other;
- Story distribution point to consider: the complexity of factors, inputs and risks;
- Step story point estimation: the story should be split into small, independent. Each story should be completed with a person within no more than two days, otherwise it will fall into a stalemate, the cost is often enormous. Meanwhile team to reach a consensus, agility is an idea, team members need to change their thinking, have a dropped, it will affect the stability and speed of travel. Need to continue the process of dialogue, coordination, improved (annual leave, leave and other unknown factors, tends to affect the integrity of the schedule and story points the team).
Point of the story analogy estimate: tell a story compared to other stories, if the story is similar to its accuracy, completion time can not be that much difference. R & D team is also recommended to assess together to avoid one-man.
- Ideal day: do not bother anyone, be the only focus of a job if all information is available; many companies promote all-rounder, that a person can do the job A job can also do B, which is considered highly efficient? This is wrong, Imagine encoding one hour 10 minutes meeting, the meeting thinking, encoding the meeting prepared to speak thinking. Hey ...... price is also quite high, and the strange thing is a lot of corporate executives choose to ignore, no one to make such negative phenomena CEO.
- "Programming for 10 minutes, 1 hour Chit-Chat" by the sentence a joke, we can see that the work is not mechanical R & D staff is creative, it is necessary to maintain a relatively intimate space, free from external interruptions. Here to say a few words: "A lot of the initial research and development to a managerial position brothers, tend to make a mistake, personally pitched; dismissive of the evaluation meeting, considered to be messing around this is actually thinking has not changed over the post. . height, decided to post daily to be the focus of the work is focused on management positions for planning, coordination, avoid risk, paced, and not personally involved in the development came in one sentence summary: the evolution of an adult, do not use thinking of a monkey, your concern is a banana trees, instead of one or two banana is good or bad. "
Contrast story points and ideal day: cut the more help drive cross-functional behavior (coordination of resources, and so support the work of answering the line); estimate story points will not decline: to determine the unity of thinking through continuous dialogue; story Point Properties , hijack a growth estimate of time. The latter: there are differences from team members; ideal day of uncertainty, will the outside world that is tricky, in fact, that is, do not fly.
Estimate the scale: quick assessment based on a reasonable forecast estimates should not pursue 100 percent accuracy. There are several ways:
- Wideband Delphi: used to collect accurate estimate of the project, estimated in a meeting to discuss only problem you might encounter, the estimated cost spent itself and not discussed. After the meeting, the team everyone quantified individually, must be independent, refused twinning-type estimates. After the crew finished estimate, we collect all of the estimates, and drawn in the painting table, showing the differences will be discussed. One thing to note, which is anonymous, that do not show their names, I remember one time, the group do satisfaction surveys, the survey volume as well as the name of a column, the result is as I expected, the administrative department collected 100% satisfaction. But the real problem is covered, the loss of this role to be played, it is a waste.
- Step: Select team members set up, kick-off meeting: made it clear game rules and procedures, and personal preparation, estimation of the meeting, configuration tasks: collecting estimate single, summary, the assessment mission: to find out the differences and reach a consensus.
- Program Poker: Because I am extremely disgusted with poker and mahjong, there is not introduced himself Baidu;
Affinity estimate: mainly used for large-scale estimate advantages: fast, simple, visible and transparent decision-making process, active cooperation rather than confrontation;
step:
- Silence relative size: Product Owner to provide production product backlog, arranged on the wall, every kind of goods in execution time spent by members in consideration efforts. Not to discuss the technology.
- Editing wall: According to the consistency of the size reached by mobile size.
- classification
- Product Owner Challenges faced: According to the estimated size, may not meet the ideal product manager. It would need to re-evaluate or other measures in accordance with the actual situation;
- Data storage: The final step, all of the above steps, are as a result: reserve data;
Compassionate Size: Advantages and affinity similar to that team members decided to reference each size; L, XL, XXL benchmarks to be unified;
Program Poker: Not introduced himself Baidu;
Determine the size of the project: productBacklog mainly to see how many items, keep in mind that, ProBacklog dynamic rather than static; the need for constant viewing, adjust throughout the life of the project; it is dynamic, because we determine the size of the team, team the number, duration of the sprint, the sprint and the number of the target ceiling versions;
Estimate to assist our work, rather than the KPI \ KBI assessment, the agile manifesto "to adapt to change rather than follow the norm" feature determines the time we estimate the cost it takes, early investment in R & D is the only choice , spend too much time to determine its estimated 100% accuracy, in fact, it is a waste.