Under the kitchen founder Wang Xu: Recipes website How to community

Intermediary transaction http://www.aliyun.com/zixun/aggregation/6858.html ">seo diagnose Taobao guest cloud host technology Hall

For people who cook, in addition to the content of some different, recipes are the same, and the traditional menu site, but also the menu is printed directly to the Internet, the user is still doing query recipes, reading recipes and then left, the slightest reflection of the characteristics of the Internet. But Wang Xu, the founder of the kitchen, added a "upload" link to his recipes.

"Work" is the key to driving the community

Wang the next kitchen, not only can read recipes, users can upload their own reading recipes after the study, and then the community interaction will naturally occur. On the other hand, "upload" users to the information to the social network, but also to attract new users to the "kitchen" to learn to cook. "Work" is the key to driving the community.

"Works" is how many people have done this dish. There are 120,000 of users in the "kitchen" upload their own "works", on average, under each menu has 7 "Works", which we are in all menu applications in the lead.

In our definition, the core value of the menu application is "work", just as the core value of "public comment" is the user's evaluation. "Work" allows users to know which recipes are currently the hottest, and what other people are doing when they do this, can help us figure out what's eating now, and improve the efficiency and quality of what we help users find. This is much more efficient than manual editing.

The key to community is to make people feel equal.

How to get more people involved in our content system, this is the core of our product design. In my opinion, the key to community is to make people feel equal.

First, we make the threshold of production lower. It's very tedious to write a cookbook. Cooking is easy, but you have to be a photographer and write steps. We will try to simplify the process of creating recipes.

The users of the recipes are the majority, and we will deliberately guide them to upload their own works. The first is to let users know that the work can be uploaded. The second is follow-up, upload the work immediately after someone and you greet, this will have a sense of accomplishment.

In the past, we have to study the work in accordance with "point praise number" to sort, but users see others do well, often embarrassed to upload their own works. Then we adjusted the "upload time" to sort, and the user saw that the dish was cooked a few hours ago, indicating that the dish was very popular. The recipe picture of this dish is very good, but the newest study work may be very general, so the user is confident to upload his work. After this adjustment, the number of "works" has been slightly improved.

Second, we let the threshold of interaction lower. We have designed a few paths, people and people have a "concern" relationship, and mutual praise. To the kitchen friend "point praise" is very easy, the kitchen friend interaction lets the user have the confidence to upload own work. The ecology of feedback makes the producers of recipes happy. I write a prescription and a lot of people do it and it's a sense of achievement. There are also circles in the community that are scattered all over the country, because food lovers get together, they collectively change their heads, or send gifts to each other, and many popular recipes are detonated by these coterie.

Help users make the quickest choices

Our latest revision highlights the shift from tool to community. Before, the "Down Kitchen" of the first page is around the discovery of content. Now, we're exposing people, and all the features are centered around the display. "Add Kitchen Friends", "attention of the kitchen friends", especially each recipe has emphasized the author. People can make content have a sense of labeling. Just like people go shopping only to their favorite shop, people is the brand of recipes, can solve the choice of barriers to help users make the fastest choice.

But the tools and the community are sometimes contradictory, such as our search function. Search is a typical path for users to use the "Down kitchen". Over the past two years, we have been using the most energy to solve the search. Our definition of search is not only active search, but also the homepage recommended. "Down Kitchen" the recommended content is the algorithm implementation, from the previous few days of popular search and last year, the same period of hot search combined impact.

We already have a hundred thousand of recipes, how to help users find what they want most quickly is the problem we need to solve. Our search list is not sorted exactly by popularity, and one of the places in the search list is for the new recipes. New recipes have just entered the community without any social and social data. We editors, like the government, give a certain amount of resources to a curious person to see. We usually give 5,000 exposures first. High-quality recipes do not need editing intervention, the main role of editors is to help the new recipes.

Personalization and precision are the problems that content distribution wants to solve

Enabling new authors for community reasons can affect the search hit rate. The first screen hit rate is definitely high if you sort by "popularity". But this will form a Matthew effect, hot recipes are getting hotter and less information is rolling. New users will feel that the menu is very classic, but the old users will feel always these, not a fresh. Personalization, precision is the content distribution is to solve the problem, we are also constantly adjusting.

Many users want our search to have a sort function. The sort itself may not be complex. But users feel that search needs to be sorted, translated into our logic, that is, we are not accurate in the sequencing. If everyone sees is different search results, are their favorite, do not need a strong sense of the sort of operation. The user needs to reach the content, not control the sort.

In the precision, "down the kitchen" North wide home recommended content is different, but has not done a single product search. The demand for food is not the same for the north and the broad three. If you want to do a single product search, you may need to dig more information. In the optimization of the details, we do not want to allow users to create confusion on the choice. The ultimate goal of the user is to find the dish he wants. We want to do a lot of things, but users are still very concise.

It's easy to get out of the community. Users to find content, user production content, distribution in the middle, we have to always think and solve the problem.