First of all, we need to be clear about what big data is. "Don't do ineffective marketing" gives a certain answer: in an affordable time range, conventional software tools can't capture, manage and process data sets. "Baidu Encyclopedia" also puts forward the 5V characteristics of big data (proposed by IBM): volume (mass), speed (high speed), diversity (diversity), value (low value density) and authenticity.
It will be found that after the level of data volume rises, the benefits and weights will increase geometrically. It seems that we can also win the popularity of House of Cards with big data like Netflix mentioned in the book, bringing phenomenal ratings and publicity.
These are the resources that our operation department urgently needs. We really need "big data" to guide and correct our actions, constantly erase the pen marks in the wrong direction and move towards victory.
On the road to victory, if you are not careful, the three major misunderstandings of "big data" can bring us into the ditch in minutes.
Many people talk, but few people know.
A product or a certain behavior is exploding in a short time, and it is being discussed all over the street. At this time, I always feel that this is "big data." I must analyze and think. As a result, I will find that there are many people who say it, and few people really analyze a little routine.
(2) sell dog meat by hanging sheep's head.
All companies or tweets use this word, and my # # conclusion is based on "big data", which is particularly effective. In fact, the so-called big data may only be tens of thousands of pieces of data last week, which is not particularly significant, because the timeliness is too short, but people who don't know it are easily misled by it and enter the strange circle of changing company settings.
③ excessive myth
I always thought that the analysis results close to "big data" were particularly awesome. From analyzing data to setting up departments, I carry the slogan of "big data" and then draw conclusions based on a bunch of calculation methods and logic. What will happen to XXX I will find that the conclusion is completely inconsistent with the direction of work and completely useless. It may even be because of this wrong conclusion that the company's strategy has gone astray.
So, since these three misunderstandings are obvious, how do we use the knife of "big data", especially in the education industry?
As we all know, all analysis must be based on data and cannot be imagined out of thin air. The guidance of "big data" to the education sector should mainly focus on two aspects, one is the teaching level and the other is the market level.
Let's talk about the teaching level first. The first benefit brought by "big data" is personalized learning analysis. It can be found that if there are too many people in the class, the teacher's control over some children will be biased, but big data will help us, but we can make good use of the data of "entrance exam"+"going out exam" and several after-school classes to give students corresponding personalized learning analysis, and can also guide children how to make up lessons according to the personalized results.
The second advantage is to quantify the learning process, because we have "VPS textbooks" and "answering machines". In addition to the entrance examination, we can also input the classroom situation into the computer continuously, generate the corresponding students' classroom mastery, track the students' learning situation in all directions, and remedy the problems as soon as possible.
When the first benefit and the second benefit are mixed together, you can describe "student selfie", so that every parent can know what his child's situation is and how to adjust it. When the teacher gets this "selfie", he can also understand the children's problems and explain them to the students in class or after class in time, so that the three parties (parents, students and teachers) can take a selfie together.
At the market level, "big data" gives more help, mainly in three directions.
① Shift change rate. Operators can always get all the data of summer vacation or winter vacation, thinking it is big data and thinking that they can write a lot of things. After careful analysis, it turns out that the low shift change rate is due to the low shift change rate of Class 0 and Class 4, and they come to a general conclusion that Class 0 and Class 4 will not be held next year.
In fact, this way of treating "big data" has entered the third misunderstanding mentioned above, the transitional myth. Holding big data, drawing shallow conclusions and pointing out work according to shallow conclusions is a transitional myth. When we get out of this misunderstanding, we are multi-directional year-on-year or month-on-month, and then think, what is the situation as usual? Never think subjectively that only in this way can we find out the truth and really help our later work. For example, if a teacher's attendance rate is low, it is easy to subjectively think that the teacher is irresponsible, so you need to come up with specific data. Without data, everything is empty talk. At the very least, it needs to be selected from "big data" in real time. Is she the only one with a low continuation rate, how to attend classes, and even the teachers in this campus? If they are all low, then it may not be the teacher's problem, right? It's inconvenient to attend class, far from the subway bus stop? Parents' educational awareness is not strong? There are competing organizations upstairs ... and so on. If you analyze it like this, you will find that your thinking is no longer at the most basic level. Therefore, this is the real benefit brought by "big data", not just a fantasy.
②KPI setting. Every operator has both pain and hate for KPI. Just like the book "We have worked hard, we are worthy", how many operators have expressed their heartfelt wishes. If we learn to use big data and establish a corresponding evaluation system, then the setting of KPI is no longer an invalid operation. Market data (including geographical environment, number of students, competitive institutions, consumption level, emphasis on education, etc. ), teacher data (continuation rate, sense of responsibility, communication frequency with parents, teachers' opinions, etc. ), and systematically evaluate these data, and draw appropriate conclusions to guide us to set KPI, not blindly, otherwise the executor will do it under such a bad comprehensive link.
3 marketing cross-border. Everyone knows that hot spots are a good opportunity to spread word of mouth quickly and bring traffic. But when we found the "big data" of hot spots, it was a little late, and the cost of rubbing hot spots was too high, because many institutions rubbed it again, and the benefits were not as great as expected. Then we can find another way, rub small hot spots, rub small demand, and make marketing cross-border and no longer limited to the education industry. For a simple example, in summer, one of my favorite activities is swimming, but it is very unlikely that the swimming pool will provide shampoo and shower gel. Shampoo, shower gel and waterproof bag are all needed by swimmers. This is the small data we need, and this is the hot spot we need. If you negotiate with the swimming venue with the scheme of "waterproof bag" and "small bag shampoo and shower gel", you can exchange the poster display opportunity of special class and the telephone information of swimming pool members, and even print our organization logo and the information we need to print on the waterproof bag. Win-win results can not only improve customer comfort, but also bring us exposure. This should be a small cross-border marketing that I think is feasible.
Cherish "big data", use the right method, stop going into misunderstandings, think more and think more, and your operation will eventually be smoother than it is now.
Where's Feng? Guangzhou new oriental school youneng high school business division