Make writing a habit together! This is my first day to participate in the “Gold Digging Day New Plan · April More text challenge”, click to see the details of the activity.

Voice-over:

I’ve been cleaning up other people’s messes, so here’s how to make a functional design that everyone can admire.

This series is mainly written for products, the best practice dare not say, I have a lot of failure cases to share and ridicule.

Research and development, please feel free to point out any inappropriate place.

1. Don’t make wheels

Yes, you read that right. If you can, don’t.

In life, there are so few things that can be done well that one or two 80 points will be fine.

A lot of people want to encapsulate the function, build a wheel, feel powerful, write a PPT, brush performance (product or r&d).

Then every day to push a variety of downstream business access, trial, get at the end of the year, promotion pay job-hopping.

In fact, a chicken feather, the bottom of the people dare to anger dare not speak. In all sincerity, it’s enough for wheels to make themselves sick.

And of course, thanks to these people, they left and then they rebuilt and they created jobs again.

Especially in these years the typhoon is very big, each factory team seized a pile of mountains.

If you want, every API can create a middle platform.

Do a simple H5 activities, have to find more than a dozen Taiwan joint tune.

This is known as “best practice”.

If you don’t run fast enough, you’ll end up cleaning up the mess or getting grossed out.

Two, the first dish organization

Suppose a business does need a wheel.

Remember organization first, Conway’s law that “system complexity increases as organization expands.”

Decide who to dump it on, or don’t do it if you can.

Sometimes bar, always feel first out to support the business, brainwashing everyone “no boundaries”, there will be a variety of natural split.

For one thing, most of the responsibilities of the corresponding team can be found to undertake within the organization.

Instead of being poked off and copied and replaced, or forced to hand over and marginalized,

Or combine team functions, better not to get involved in the first place. (If you’re running off with your resume, ignore it.)

The other side does not have manpower, can let the business side HC out of the budget. Slow schedule, can let the business side up, even procurement.

The problems of being able to spend money are really small problems. The team’s manpower is limited, the business demand is infinite.

Is there really enough rich manpower to do it well and maintain it for the long term?

This is a question for both product and development.

If you need to start a new team, think again about putting in your own people.

Three, find the right backer

If you’re really unlucky, you need to masturbate yourself. Or maybe you just want to build a wheel. Find the right business to back you up.

Use the business to create a “moat” for yourself, driving up the cost of invasion and replication of rivals.

Jack Ma has a saying to the effect that if it takes 10 years to do something well, 99 percent of people will give up.

Not many people want to be with us in 10 years, 2 years or 3 years.

So think about whether you can strongly bind to your business. UI component libraries, for example, are not really necessary.

Instead of technical volume, others rest you want to graphQL, not only disgusting others, but also disgusting yourself.

But picking the right business is an art:

This track is not popular, is the capital still invested, to what stage of development, what is the profit margin, can make money?

Is it being positioned on offense or defense? Is the offense a publicity stunt, a public relations stunt or a way to make up the numbers?

Political awareness is really important, and the best choice is certainly in a new business with a big market. For example, domestic transfer abroad.

The main feature is at least the creation of a new first-class department. If the evaluation makes sense, run with it.

If not, build on your current business:

The best advice is:

In the revenue or flow of this business and other main functions, and find a business old products or old operation belt.

The architecture design/implementation scheme does not have to listen to him, but at least the needs of others out of someone. Consider using it or extending it.

At least not like some big factory looking for some 985 fresh products to do a bunch of bullshit features,

For example, design a nine-grid lottery:

To run the dynamic effect of the lantern, 9 prizes let people highlight and default a total of 18 icon images.

Not to mention the heavy workload of operations and art, and the “who the hell did that?”

Technically, one slice is 50KB, and 18 slices are 0.87MB. Don’t worry about the first screen speed and port time?

Oh, if you happen to be a product and you’re in the engineering business. Advise you to understand more business, think about the whole link.

If you really don’t understand, you should ask questions. Look for a colleague to ask, grass-roots level seniority is the oldest operation which a few, one by one chat.

Especially men, quite good face, most “pontifical” haha ~

As for how to think and design this wheel, welcome to the next part of this series: The basics of Building a Wheel.

Good Luck.