At the end of another year, companies are making work plans for 2021. Mentioning the work plan of data analysis, many students scratched their heads tangledly. How should the data analysis work plan be written? Today we will explain the system.
1. Common pits in work plans
If you let other departments write work plans, the probability is the following style:
- Sales: Create 2 billion revenue for the company throughout the year, and the planned monthly target is XXX
- Operation: Organize 10 events throughout the year, double 11 sales hit 500 million!
- Supply: guarantee the supply of 2 billion revenue sources, reduce the loss rate to
- Development: 10 events are guaranteed to go online, and the system runs stably for more than 300 days
So how to write data analysis?
- Writing method 1: Write 2000 lines of SQL every day, complete 500,000 lines in 250 working days throughout the year
- Writing method 2: Establish 20 prediction models to achieve a prediction accuracy of 99.99999%
- Writing 3: Establish 10 data systems to promote the company's digital transformation
- Writing method 4: Provide 10 activity reports and provide data accuracy of 100%
Question: Which one of the four words above is OK?
It is about the performance in 2021, readers must stay and think for a minute
think
test
One
Minute
bell
answer:
Sales and operations are related to the company's efficiency and financing progress, and are directly related to everyone's wallet!
Although supply and development do not directly make money, they can't make a penny without them, which is rigid support.
What data analysis alone does is neither rigid nor profitable, and is optional.
Among the above four wordings, 1, 2, and 3 are serious failures. Because 1, 2, and 3 are all about data analysis, that is, it has nothing to do with performance or income. But who is driven, how much is driven, and how to measure whether it is driven or not, I didn't make a single sentence clearly, and the business department didn't know whether it recognized it or not. It was just empty talk.
Only wording 4 is barely passing.
1. At least put yourself in the position of the support department, clearly positioning.
2. At least tie your work to the company's major projects, not dispensable.
3. At least the results of the work are quantifiable (output 10 times), and the major projects are online, and you must look at the data.
Although it is still difficult to measure performance in this way, at least it pulls itself and development to the same level.
This is the starting point for the data analysis work plan
2. Basic writing of work plan
Three iron laws of data analysis work plan
- Bundle the work of other departments of the company.
- Output content, new "optimization" guarantee.
- Quantify in a way that other departments can feel.
For example:
After such optimization, it can reflect the value of the data to a large extent, and it is easier to use than writing by yourself: I did XXX. You know: most people in other departments (including the bosses of most departments) don’t understand the principles of data. There are counts/no counts, to help you make money/saving costs, stability and error-free work is the work that most people understand better Results.
When making a plan, the task goal is locked, and subsequent performance evaluation is easy. Therefore, fundamentally, avoid the question: "What's the use of what you do!!!"
Of course, this is just a basic way of writing. There are better ways to investigate the nature of data analysis in depth.
Third, the advanced approach of the work plan
In essence, there are four steps to a data from generation to use.
These four steps correspond to the three important tasks of data analysis (such as the small picture)
The following parts are the key points:
Infrastructure construction will not be reflected as credit in any case.
You should do a good job, and you can't do it well. This is the true status of infrastructure. Therefore, if you are involved in infrastructure construction activities, such as burying points, designing/maintaining large tables for business departments, building indicators, and checking calibers. Be sure to bundle it with the company's key projects, major policies, multi-departmental linkage and other major events! In this way, the resistance is small when executed, and it is easy to settle accounts when discussing merits.
For example, to strengthen the quality of buried points, write:
l The company's key growth project sub-projects in 2021
l Fill in the gaps in the original WeChat fission channel data lack
l Total monitoring channels increased from 15 to 20
l Total user tags increased from 100 to 120
Only in this way can it be quantified, so that everyone can feel the workload, and it can be easy to talk about during the evaluation.
If you are not familiar with this way of writing, just look at how the opening of the railway was broadcast in the news network for a few days:
l National key projects in the 13th Five-Year Plan...
l Total traffic mileage reached 1500 kilometers
l The original 5 hours travel time was shortened to 2 hours
l Filling the gap of no high-speed rail from XX area to XX area
That's it!
The core of data production is tooling.
Without instrumentalization, whether people have looked at it or how much they have read, they don’t know, and it is impossible to quantify output. In all instrumentalization, new construction is always easier to reflect merits than optimization, so the new goals must be listed first.
Work output, the more conspicuous the better:
l Large data screens are more conspicuous than data billboards
l Data billboards are more conspicuous than data reports
l Data reports are more conspicuous than Excel reports
l Excel is more conspicuous than emails
Therefore, when making work plans, communicate more with the business department. If there are major events and major projects this year, please collect them first. Then try to push out the conspicuous data products, other complex needs should be pushed, and people should be recruited to fill in holes. This will make it easier to reflect the results.
The core of data usage is the scene.
The more specific the scene, the more likely it is to use it. Open a report for sales. It is estimated that less than 1% of the results will be seen. You can send task reminders directly on the company's WeChat. If you don't click, you will have to click. The reading rate will go straight up.
If you have to make a prediction model for the supply chain, if you have to predict 100% accurately, the gods will not be able to deal with it. If the scenario is specific to: reduce the cost of wrong selection, it is estimated that there is a chance to single out the obvious selection;
To make user portraits for operations, if you have to exhaust user characteristics, it is probably useless to die. But if it is specific to select high-potential users and eliminate the wool party, several characteristics may be confused;
So if you want the data to be useful, the scene has to be very detailed. The finer the scene, the better. It is best to use four or five application scenarios for a set of data to maximize benefits. The plan is detailed, and there are a lot of things to write about when evaluating performance.
Fourth, the core difficulty of data analysis planning
Data analysis plan is difficult, where is the core of the difficulty in assessment?
Everyone said verbally:
l Digital transformation is really important
l Data analysis is useful
l Data-driven business
But when it comes to evaluating performance, you will ask:
l The company's double eleven is 5 billion yuan, how many yuan is made by your model, and how many yuan is made by others?
l You can write sql, so can the development brother, and can also write in the operation. What special contribution do you have?
l How much money can you earn if you look at your report and don’t look at your report?
l There are too many digitization things, and you can digitize when you count?
This is the root of all the difficulties in data analysis! So when making a plan, you have to carefully sort out the scene and choose the direction so that you can go smoothly during the evaluation. Otherwise, the direction is selected and the plan is set into a running account. Naturally there is no good result. It is estimated that after reading it, the students will find that the difficulty is still: in the use of data, how to lock the scene so that the business department can really use it, and admit that this is the credit of data analysis. If you are interested, follow the public account [Grounding Qi School], let’s share it in the next article. Stay tuned.
The down-to-earth teacher Chen, master of mathematics science, business management from 985 universities, data science management expert, has 11 years of rich experience as data director and senior consultant. Served large enterprises such as Ping An Bank, China Guangfa Bank, Tencent, and created online and offline integrated digital transformation solutions for traditional enterprises such as Vinda Paper, Overseas Chinese Town, Guangqi Honda, and World Union Real Estate.
The creator of the public account [Grounding Qi Academy] has independently launched a series of data analysis courses, which has more than 20,000 students and is well received.