From SQL Boy to Manager: Five Stages of Data Analyst Growth

This article mainly introduces the five major development stages of data analysts from junior to mature, namely data acquisition stage, demand stage, analysis stage, value stage and management stage, and elaborates on the key capabilities and skills required for each stage.

Students often ask: Is there a trajectory for data analysts to grow? From my own experience, having served a large number of companies, and having coached thousands of students, I can see that there is a trajectory for data analysts to grow, but different companies have different growth ceilings for data analysts, so everyone's experience is obviously different.

Generally speaking, it can be divided into five stages.

Phase 1: Data acquisition

SQL Boy is a necessary stage for data analysts. The saying "Generals are born in the army, and prime ministers are born in the county" is exactly this reason. Because in real work, getting data is not as simple as writing SQL against a cleaned large wide table.

In order to improve data quality and ensure that data is collected correctly, there are many tedious tasks to be done: understanding the data caliber, understanding the business system and business process that generates data, understanding database design, designing reasonable payment requirements, checking data quality, and understanding the real reasons behind manual reporting errors and omissions... tedious, entangled, and complicated are all normal work.

These boring tasks are something that product managers and operators will never experience, and are also the unique experience of data analysts. It is in this difficult process that data analysts can accumulate business knowledge, the ability to see the business through data, and the ability to operate data, thus laying a solid foundation for subsequent improvement.

Of course, SQL Boy itself has also been criticized a lot, and everyone hates being a "human data extraction machine." However, this is not the fault of the data extraction itself, but many companies only let data analysis stay at this stage, without planning a longer-term development path, and changing companies can solve the problem.

Phase 2: Requirements

“Give me a number, before I leave work, and hurry up” – this is a request.

"We are launching a new activity this month and need to monitor the results" - this is the demand.

Meeting the needs of the business rather than the requirements is an important step for data analysts to gain recognition and work independently. When many companies recruit "senior data analysts", the word "senior" mainly depends on whether they can handle the needs of the business side. If you can handle the needs, you don't need to wait for the leader to arrange everything; you don't need to let the business repeatedly raise numbers and still be dissatisfied; you have the opportunity to think ahead of the business and find more cooperation opportunities.

From requirement to demand, there is only one word difference, but it means the improvement of many related capabilities. You can't just work hard and do whatever the business says, otherwise you will always be led by the business. You can't wait for the leader to teach you one by one. You need to know what indicators and analysis dimensions are needed for common sales, operations, products, supply, and production.

You have to take the initiative to communicate, propose solutions to guide business, and merge scattered needs into reports that can be monitored regularly. Only in this way can you achieve the goal of meeting needs and reflecting work performance. This requires improvement in data cognition, business cognition, and communication skills.

In some companies where the relationship between departments is harmonious and the data and business are good, this step is easy to achieve. But in some companies, it is difficult. The company lacks a cooperative atmosphere and only treats data analysts as human data extraction machines. If you are trapped in such an environment, students who work in data must have the courage to break through and change to a better company. Otherwise, it is easy to sulk for years without making any progress.

Phase 3: Analysis Phase

Note! Not all business needs require analysis. Many needs are simply "monitoring data." However, the needs that can reflect the value of data analysts must be analytical needs. By analyzing problems and making leaders feel that data analysis is useful, there will be opportunities for further promotion.

Anyone who has experienced building a data department from 0 to 1, or expanding a data department from 1 to 10, will feel deeply touched by this. When leaders are thinking about problems, they think of inviting the data team leader to come over for a chat, and they think of asking for data advice, which can lead to promotions and salary increases faster than any KPI/OKR assessment.

At this stage, we often face conflicts between theory and practice, ideal and reality, such as:

Even though very little data is collected, the bosses want “in-depth analysis”.

There was clearly no scientific sampling, but the bosses wanted to “make a reasonable assessment.”

Even though there are obvious problems, the bosses still hope that "the results will be better."

In order to meet the needs, the leaders/managers of the analysis team often need to jump back and forth. They often first try to learn statistics, algorithms and other knowledge, try to collect the practices of their peers and peers, and then find ways to deal with the shabby data of their own company, and then "understand" the boss's will and find ways to meet the boss's reliable/unreliable needs.

Many idealists will fail at this stage, complaining that their bosses are unreliable, that their superiors are just fooling around, and that they feel that they have no place to display their real talents and knowledge. Attention! Job hopping cannot solve the problems at this stage. Even if you change companies, it is difficult to guarantee that the data quality is 100% good and that the boss 100% understands and abides by the rules of statistics and algorithms. You must first enlighten yourself in this stage to survive.

Stage 4: Value Stage

Almost all data department leaders in companies have faced this kind of soul-searching question:

"What's the use of your analysis?"

“How is your work performance evaluated?”

"What have you contributed to the development of the entire company?"

At first, I was like a dog looking for numbers, and I thought the reports were ugly. This is a very real working state. As a data analyst, if you want to gain further recognition from the business, it is not enough to rely on PPT and Excel output. You must create several fixed value points and have your own product output.

This stage is a test of the project operation ability (pie-drawing ability) of the data department leaders. Find hot topics, guide the bosses to express their needs, rub data output towards hot topics such as "digital transformation", "data empowerment", and "data-based management", build good relationships with business departments, receive more letters of praise, and praise each other to show value. These are all basic operations.

At this stage, "how to package data products" and "how to improve customer experience" are two key topics. To package data products, we can combine hot topics in different time periods. For example, while the concept of the middle platform is popular, we can engage in CDP+MA digital operations. For example, while everyone is calling for "digital transformation", we can first launch BI projects such as "battle situation observation room" and "management cockpit". For example, while data empowerment is popular, we can launch mobile terminal reports. Use tools to replace temporary data collection, so that performance can be guaranteed later.

To improve customer experience, we must first clearly understand who our target customers are. The big boss, business department leaders, and frontline staff are three completely different groups. Often, we need to talk more about advanced concepts in the industry and paint a rosy picture for the boss. For the business and frontline staff, we need to find the topics they care about most and exchange interests: data proves that the business is doing well, and the business proves that the data product is useful. There are many detailed operations to talk about here. If you are interested, I can write a separate article later.

Phase 5: Management Phase

After going through the first four stages, you can actually become a data department leader in the company. However, the size of data teams in different companies varies, so the jurisdiction will be different. As a department manager, in addition to professional ability, you must also have a certain grasp of conventional management methods. This topic is a long one, so I won’t talk about it here.

However, not all data department leaders are promoted from the data analysis line. Among the clients I have worked with, there is a type that came from the business line, such as the strategic development department/marketing department, or simply parachuted in. This type of leader is often very good at managing upward and meeting the needs of the boss, but is weak in serving other departments and implementing data products.

The other type is those who came from the IT industry, such as those who started out working on data warehouses or business systems. These people have strong implementation capabilities, but their ability to manage upwards and paint a rosy picture needs to be improved. That is why I have a business to do, hehe.

Therefore, as a data practitioner, you always need to keep pace with technology and business. From collecting data to doing projects, from doing projects to making products, from making products to improving management, constantly creating value for the company and enhancing the experience of bosses, you can constantly make yourself progress and strive for more opportunities for promotion and salary increase.