Structuring Analytics culture and Data Careers in your organization
Data Science. Algorithms. Predictive analysis. Programming. Error 404 in HR.
Since the immensity of the universe called digital transformation has taken over organizations, congresses and events, HR areas have been trying to be on top of this trend. Cultural change, Programs focused on the development of digital skills and innovation are part of what should be the path to analytical culture.
The great truth is that this digital transformation boom came without a manual with the step by step for the company to set up.
I participated in many congresses and events in the latter in which pompous HRs went to talk about digital transformation and "People Analytics", the version of HR data science. Slides and more slides on fantastic things that HR can (remember that word, CAN) do. Take a test: when you finish a lecture like this, go ask the speaker if he / she can present a real case of your company about that mountéu of beautiful word spoken during the lecture.
Most likely you will hear "well, ...". Most likely, what you will see is a dashboard. A global survey conducted by the International Institute for Analytics has shown that most industries and organizations have medium to low data maturity.
But what does data maturity or analytical maturity mean?
CULTURE. It means having a culture in which the most important decision-making processes are done through data.
Analytical maturity is a process of constant evolution.
Ricardo Cappra, from the Cappra Institute, defines the analytical maturity structure on 5 different levels, in line with Gartner's analytical model:
Level 1 - Reporting: companies capable of creating timely reports and that use data in a more homely and unstructured way. International Institute for Analytics research shows that many companies are at this level;
Level 2 - Analysis: here it is already possible to make some decisions based on data and companies already professionals who dedicate their total or partial time in the organization, to work with data;
Level 3 - Monitoring: many companies that reach the level of having colorful and functional dashboards think they are alreadyon the top of the mountain. Hence they will present these same dashboards calling them data science and predictive analysis at congresses and events. Be careful not to be that company;
Level 4 - Automation: at this level, there is already a proactive demand by the organization for new methods and tools that enhance the data culture and, consequently, the level of analytical maturity of these companies. Strategy aligned with decision-making based on data and integrated systems start to take shape and allow the automation of analytical processes. Here there is already a clear knowledge of HR about data careers;
Level 5 - Artificial Intelligence: the company starts to use information as a strategic asset and to use the analytical culture at its maximum power as a way to reach the organization's objectives. With a mature and solid data structure, data scientists are able to work and deliver value through predictive algorithms and analysis.
Justa a moment. What about all that buzz around People Analytics in HR?
According to Cappra, "analytical culture does not refer to creating tools, it refers to making business decisions based on data. The basis of Analytical Culture cannot be tools, it must be people. Tools must be used to solve well-specified problems while culture must run in the corridors and ensure that business questions are being answered with the support of information and strategic analysis. That is data-driven".
If a large part of the companies are at the level 1, 2 or 3 of the analytical culture, how is it possible that HR is making such progress, according to all these events, lectures and congresses that we see out there?
Good question. Ask the speaker the next time you hear the thousand wonders of People Analytics. If we have very few companies with mature analytical culture at levels 4 and 5, probably the HRs that understand analytical culture are in these companies. The other HRs have no idea what they are talking about.
And remembering that this will not be a rule. Often, the analytical culture is an initiative that comes from business and not from HR. Therefore, even in companies with good data maturity, you can find human resources areas that are not only far from the business, but far from understanding this culture.
Therefore, my dear reader, the first thing that HR must do is to seek knowledge on the subject data, before everything. This is a very complex world. I also used to spoke many of these beautiful words of digital transformation when I was in HR and I only discovered the size of this monster called data when I made my move to one of the company's businesses that already had good analytical maturity.
That's when I realized that I didn't know anything about it. And today, 2 years later, I can say that I know 5% (and I'm not being modest).
I discovered this when I realized that in the organization, even though we had a good maturity, between levels 3 and 4, we were having a hard time hiring and retaining data scientists.
I started to analyze, ask and interview these professionals. Many of them had come from native digital companies and found a different, more hierarchical environment, with a lower analytical maturity, which made them to work with organization, structuring and cleaning data. I confess that after talking to them, I still didn't know how to deal with the problem. I had a lot of experience in HR with the creation of youth programs - internship, trainee and apprentice and the first idea I had was to create an internship program in data science, so that we could develop a pipeline in the database for data professionals, in addition to increase retention by offering a structured career acceleration program for these young professionals.
And in January, 2020, Elissa Suzuki was hired to help the organization accelerate digital transformation and data maturity. Since then, Elissa has not had a peaceful day, lol. On your first day, I remember we were having coffee and talking about the analytical challenges we had. I commented on the internship program I was thinking of creating and there, I got slapped: "Caio, I thought the idea was great, but let's talk about this data science thing".
It was there I realized I knew almost nothing about the subject. Not only did Elissa gave me a masterclass, she also became a valuable partner in advancing our analytical culture.
How did Elissa get into data? How was the career change for this training designer?
"My work with data started in 2016, when I was a trainee at Banco Santander. At the time, I had the opportunity to choose some areas of the bank to perform a job rotation. As it was a unique opportunity, I wanted to challenge myself and get out of the box, so I ended up choosing completely different areas of my expertise and previous experiences. In this challenge, I fell in love with data. 😊"
"I usually joke that every day is a career change in this trajectory, I learn a lot every day and always go after it. Speaking specifically of the moment when I decided to work in this area, the first of all was to admit that I didn't know much and then to go for it: I took courses online, the first ones were Big Data at Coursera and Python at Udacity and 2018 I started an MBA in the area. And always ask, don't be afraid to ask." Elissa Suzuki
Given this big and necessary introduction, let's talk about data careers
Where to start when structuring data careers?
According to Elissa, always understanding where the company is at. She adds:
There is no point in hiring a data scientist by road if there is no data for them to work and if managers do not have at least this analytical culture to manage these teams. Therefore, understanding where the company is at and then building what the needs are in that context is a good first step. 😉
She also comments that it is always good to involve business, HR and IT in this structuring. Business bringing the needs, IT bringing the technical look and HR taking care of career, people and training.
Elissa built a slide that helped us a lot in the storytelling of the change we needed to make in the organization:
There are many possibilities and possible chairs in a data career. For simplicity, based on a Mckinsey material, Elissa divides 3 major roles that an organization can start with, if it already has at least a level 3 analytical culture:
The data engineer is the person responsible for structuring the data, making it accessible. This role is the basis of data science, so there is no point in hiring data scientist, as she well points out, if your organization is at level 1 of maturity and has disorganized data.
The engineer is necessary for the creation of the pipeline that transforms the raw data that are in the most varied formats, from transactional databases to text files, in a format that allows the Data Scientist to begin his work. It is also up to the Data Engineer to keep this pipeline running so that the data can be collected at the right time, with the level of security required by the company. The work of the Data Engineer is as important as the work of the Data Scientist, but they tend to have less visibility, since they are more distant from the final product resulting from the analysis process, which is produced by the Data Scientist .
This issue of visibility is also a change management to be done in the organization. The data engineer usually has a more introspective profile, is the famous person who sits ain front of a computer for hours on and his job is a backstage job. This person will not be presenting a beautiful dashboard to the board, but without his work there will be no solid basis for the rest to occur.
This role aims to work on the solid basis made by the Data Engineer, analyze past situations and reports to create data visualization through dashboards and reports.
In many organizations with low data maturity level this role is mixed with Data Engineer, but note that the profiles are different, which emphasizes that the competencies are also different.
Data Analysts examine a large amount of data and extract insights and patterns from it. They are responsible for collecting, organizing and obtaining statistical summaries. These explanations are provided through visualizations and reports, so that companies can make strategic decisions with them. They need a basic understanding of programming, mathematics, statistics and it is preferable that they have excellent communication skills and considerable knowledge of computer science. R and Python languages are widely used in data analysis and provide an arsenal of tools for collecting, cleaning, transforming, processing and interpreting data. 
The Data Scientist is the filet mignon that every company wants now. But do you realize the importance of the Engineer and the Analyst before having this role?
The Scientist will look at past and present events to perform predictive analysis, that is, to bring possible future scenarios through algorithms and artificial intelligence.
A Data Scientist has all the skills of a Data Analyst, with deep knowledge of modeling, statistics, mathematics. machine learning, predictive analysis and computer science. Therefore, they can create complex predictive models that can provide valid recommendations based on historical data.
The point here, however, is the robust business acumen that differentiates the Data Scientist from the Analyst.
It is necessary that a Data Scientist can transmit the findings in the form of a story to IT professionals and business managers, so that they can take calculated risks and make viable decisions based on the information provided by the Data Scientist .
Which is most important to my company first? The Scientist, Analyst or Data Engineer?
Help us here, Elissa: What a controversial question, lol! I would say that the most important thing is to bring value to the business and the customer using data. Always remembering that for this to happen there are several data careers behind, including scientist and data engineer. An appropriate team must be set up depending on the analytical maturity stage the company is in and also the company's investment in data. In this way, the company will be closer to balance (Ps. It seemed like Master Yoda speaking, lol).
It is not a matter of hiring only scientists or just engineers, but finding a balance between careers within the context in which it is inserted. Elissa Suzuki
But, if you're going to start, you can't hire a Data Scientist if there's no structure for this professional, so go to the base first. This is what we did, after so many conversations, when creating the Bayer Data & Analytics Internship Program, one of the many actions that our organization does in favor of the analytical culture.
A Business Internship Program
Returning to our initial problem of hiring and retaining data scientists, we completely changed the course and objective of the internship program that would be to create a pipeline for data careers, focusing on data engineering and analysis.
The pilot of the program with 14 interns was launched in June 2020, 5 months after the initial conversations that Elissa and I had for the Product Supply LATAM area of Bayer CropScience.
The program was built in partnership with several professionals from different areas, including HR and IT.
How was it to create the Data & Analytics internship program at Bayer?
Elissa replies: not only was it, it has been amazing! Creating a technicsl learning path in partnership with Mastertech, connected to the reality we have at Bayer CropScience for interns from all over Brazil, to see their growth, development and the results of their projects reflected in the business, is really very rewarding!
The program lasts for 2 years and has a data learning path, provided by Mastertech, of 120 hours. The fact is that data professionals study A LOT and in the program we replicate that. Interns do not work on our short fridays. They have classes from 9:00 am to 12:00 pm, dedicating one day of the week just for study.
On the learning path, they have a programming base, statistics, Scrum (agile), modeling and data analysis. In addition, each intern has a strategic project linked to business needs and some of them work in a squad format, using Scrum.
Interns also rely on internal mentoring by our data professionals, in addition to mentoring provided by Mastertech.
It has been fantastic to see the results of the program and how much it has contributed to further advancing the analytical culture (towards level 5, one day 😉).
At the end of 2020, the interns had the opportunity to participate in our version of shark tank, in which they presented challenging projects to Product Supply LATAM executives looking for "investment" in the fictional currency PS Bitcoin. The interns were able to exchange the amounts received for mentoring, change management plans for their projects and training.
Tips for companies that are still starting the analytical culture
Elissa is categorical:
I will leave some tips here, but the biggest tip I leave is always to ask yourself ‘what problem do I want to solve or what question do I want to answer with data?’ Before hiring scientists, buying tools and solutions.
She also adds that when we talk about the beginning of an analytical culture, first of all we need to understand in which level of data maturity is currently the company.
"This step is very important because, perhaps, in the beginning the company does not need an algorithm that uses machine learning to answer simple questions that will help the business. After making this diagnosis, start small, list some cases of data use, test hypotheses that will output analytical products that will support decision making. In addition to all of this, have a suitable team, available and that owns this movement to prove value with these use cases and spread this movement throughout the company", points out Elissa.
Whether you are an HR or business person, the journey we have gone through and described here, the path to the creation of an analytical culture is complex, it can take time and it needs a lot of investment: time, people and can reach monetary investments, mainly equipment and programs. If you are coming to the conclusion that the path in your company will also be arduous, start by:
Education: obtaining the necessary knowledge to effect cultural change will be very useful. Readings, courses, conversations with the leadership and with the data professionals help. Knowing the concepts of analytical culture and change management will be fundamental
Alignment of the organizational strategy to the analytical culture that will be created: Without the support and understanding of senior leadership, none of this will be possible. Leaders have the enormous potential to stop or slow down these actions. A structured cultural change program will also be critical
Start structuring data careers: job descriptions, benchmarking of companies that already have data careers, entry programs such as internship programs, internal training programs for data professionals, etc.
Caio Ianicelli Cruzeiro has more than 13 years of experience in Human Resources. He is also a speaker and coordinator of networking groups focused on HR and executives and is currently Manager of Capability Building in Product Supply LATAM for Bayer CropScience. He holds a degree in Business Administration, a postgraduation degree in Corporate Education from SENAC-SP and an MBA in Agribusiness from Esalq-USP. Prosci® Certified Change Management Practitioner and Agile Scrum Certified SFC ™
Author of the book "Talent Acquisition: the evolution of traditional recruitment & selection" that can be purchased here