10 lessons I learned being a Data Analyst at QuintoAndar
Here are what I consider some of the most important lessons I’ve learned in almost 3 years working with Data Analytics at QuintoAndar. The first seven topics relate to Analytics hard and soft skills, the last three topics about career.
1 — Start from the actionables
2 — Always separate the problem from the solution
3 — Storytelling and communication deliver your impact
4 — Data Analytics comes in many flavors
5 — Teamwork and networks open so many doors
6 — Curiosity and connecting dots are the drivers of innovation
7 — Strategy will leverage your efforts
8 — Environment can make the ordinary into extraordinary
1 — Start from the actionables
Data Analytics is all about decisions. Every great analysis will drive actions. We can measure the impact of our work according to the quality of decisions they help to drive. Some even would argue that an analysis without actionables is a lost effort.
A product discovery, an experiment design, an unified metrics or a data model, in the end of the day, add value when they support decision-making. Some examples of them would be a focused product strategy, an impactful straight experiment, or well-tailored operational actions. Analytics will help to make these efforts in the right directions.
Usually the way we think is what->how->why. A powerful way to craft good analysis is thinking backwards: What are the actionables (why) I need? Only when we find good actionables, we start to design lines of action (how), and lastly, the required data and tools (what). Carrying a good tool set (e.g. SQL, python, analytics methods) is relevant (when we just have a hammer, every problem looks like a nail), but the tools are just mediums. The actionables are the end-goal and the destination.
2 — Always separate the problem from the solution
It’s natural for most people to jump on to solve something before understanding what “something” is really well. This trap in data happens when people ask, let’s say a dash of metrics, just to find out when ready that they didn’t deliver much value at all.
This doesn’t happen because people aren’t smart or knowledgeable, but because usually they don’t know how the enabling analysis can solve the problems in ways they didn’t expect it was possible. And also because it’s hard to predict in advance what solutions will work to overcome the obstacles. Discovering the solution is part of the process.
The product lesson of falling in love with the problem fits well here. Especially in complex scenarios, the Data Analyst can ask questions and brainstorm together to discover and understand the problem. Then, usually it’s possible to reframe the problem, and deliver actionable solutions way simpler and more insightful than firstly imagined.
3 — Storytelling and communication deliver your impact
The better you communicate, the better outcomes you will achieve. Behavioral economics taught us that humans are not mainly moved by data and reason. We are moved by stories, relationships, intuition. That’s how our evolutionary brains were wired. We also love to know why — Why is this relevant? Why should I care?
In order to move decisions and actions, the best analysis tell stories: Why the problem is important, how we approached to solve, what are the insights and actionables. This why-how-what structure flows naturally in speeches and presentations.
Apply different communication styles for different people (or groups of people). Simplify the messages as much as you can. Use data visualization techniques. Assume that in communication very few is actually understood and we forget quickly, so, remember Aristotle — “tell people what you are going to tell them, tell them, and tell them what you’ve told them”.
Write things down. Documenting in presentations has the side effects of being easy to share, build knowledge, promote data culture, and instigate conversations. If you want to go further, two great books on data storytelling are Factfulness and Storytelling with Data.
4 — Data Analytics comes in many flavors
Though each analysis is unique in its own way, over the time I categorized much of the Data Analytics work into five categories. Each one has its own abilities set and focused impacts on the business.
- 💡 Discoveries — These will throw a light on unclear problem or intuitions on the business. They are more risky, but when successful will help to assess opportunities size, present opportunities that were not seen (sometimes a huge payoff) and drive results way faster.
- 📈 Metrics & Analysis — Analysis related mostly to quantify specific product, operations and business aspects into metrics/dashes. When we measure properly, it’s way easier to manage, assess and drive the right decisions and actions.
- 🧪 Product Experiments — Analysis related to the end-to-end product lifecycle, from discovery to delivery. They will be around discovering opportunities to direct efforts, creating scenarios and thresholds for experiments, measuring impacts in A/Bs or other experiments to deploy with confidence and deliver impact.
- 🎲 Models and Governance — Models are useful when there is new data on the business, known processes to automate, or to operate better at scale. These incremental changes impact teams already working with data, reducing work and increasing efficiency.
- 🎓 Data Culture and Education — Promoting data across the company comes in many forms, such as empowering people to leverage existing data and analysis, teach needed skills such as SQL and actionable metrics, onboard new hires, give informal data mentorships and so on. These increases over time the team autonomy and productivity.
5 — Teamwork and networks open so many doors
Teamwork has a major importance to Data Analytics. Acting proactively as advisors enable us to drive way better decisions than only reacting to business demands.
Companies are made by people. The networks of trust we build slowly over time are a great source of inspiration and learning (e.g. a Product data analyst working with operations and strategy teams; or a Data analyst in multiple cross functional squads of a product manager, designers, software engineers). These established networks also help to open the doors we need (e.g. finding tools, resources, knowledge, contacts, opportunities) to deliver the impact we want.
One of the coolest projects I joined, about remodeling our demand conversion funnel (driven by product changes post-COVID, as talk to an agent and offers without visit) delivered its impact only because we had many areas working together (e.g. Prod, Ops, Mkt, Planning, Strat) and could unify a vision that helped to move on and drive changes in each one of these contextes.
6— Curiosity and connecting dots are the drivers of innovation
Some of the most interesting analyses are unexpected. They connect seemingly unrelated concepts to bring unique and new perspectives. For example, we could bring an income distribution coefficient to measure a housing marketplace dynamics, some behavioral economic learnings to design a conversion product experiment, or a social science metric to quantify some real estate agents challenging scenarios.
Large companies like QuintoAndar have a great flow of information in so many things happening, teams, projects, initiatives. Acquiring the needed domain and business knowledge will be faster. The key is staying curious. Some people (myself included) will even advocate for you to have multiple careers paths. When you put together different perspectives, you exponentially increase the number of ideas and value you generate.
The diversity of people, bringing their personal and professional stories along with diverse backgrounds (in the data chapter we have statisticians, economists, computer scientists, social scientists, business majors, engineers from civil, mechanical, chemistry, electrical, aerospatial areas and so on) enrich every discussion around how to approach a problem.
We can’t connect the dots looking forward; we can only connect them looking backwards. So we have trust that, if we learn, things will connect in the future. Though we don’t know exactly how this happens, it just happens. One of my favorite books, Linked talks about Network Science and how almost everything in our business, science and everyday lives are connected.
Einstein said “imagination is even more important than knowledge, because knowledge is limited, whilst imagination embraces the whole world”. Asking “What if’’ will eventually lead to some of the most interesting analysis and insights.
7 — Strategy will leverage your efforts
Strategy is about overcoming obstacles leveraging resources. The kernel of good strategy was applied in wars, governments, and business. We can apply it on Data Analytics projects as well:
- A diagnosis will define the challenge we are solving. A good diagnosis simplifies the complexity of reality and the problem we want to solve, in its essential and critical aspects. Let’s say we diagnose we have a high churn of listings.
- The guiding policy is the approach we choose to overcome the obstacles in the diagnostic. For example, we could proceed on a product discovery to find where specifically this churn is happening.
- Coherent actions are designed to deliver the guiding policy. In our example of product discovery, we could have steps of exploring the listings characteristics, segmenting depublication patterns, and assessing the opportunity size to resolve each segment. At the end, we would iterate the entire process again, on a next step to build specific experiments.
Besides that, there are many leverages we can apply on analytics projects, for example proximate objectives (e.g. an analysis that opens others), focus (e.g. narrow the problem, pick your battles) or design (e.g. one approach to solve multiple problems).
While doing things right (e.g. SQL, communication, team-work) is a foundation, strategy helps to find the right things (what needs to be done), reduce the amount of work and leverage the impact. I highly recommend Good Strategy Bad Strategy.
8 — Environment can make the ordinary into extraordinary
Before joining QuintoAndar, I never had realized how big is the role the environment plays in ourselves. Here we find many world-class product practices, such as the ones described in the excellent and straightforward book Inspired (for example, agile team organization, quarterly plannings with OKRs, strong product strategies) along with excellent peers and leaders.
This dramatically increases our productivity as Data Analysts. For example, in my early days learning data I would be stuck in analysis and spend hours on youtube figuring out some details. After joining QuintoAndar, I could just brainstorm analysis directions with my leader on the weekly 1:1’s, or ask my questions to an experienced peer at my side. These allow us to move on in minutes, instead of hours.
Over time, iterating again and again with diverse people and challenges leads to huge compound effects. These catalyzes our impact, learning and career progress. Some will refer to the “dog years”, when one year is equivalent to so many in traditional places. I’ve seen for many people, including myself, how the right environment transforms ordinary people into extraordinary teams.
9 — Empower & be a missionary
QuintoAndar is a prime example of a product-driven company. These companies solve hard problems in ways the customers love and work for the business. Product-driven companies rely on empowered teams, who are given problems to solve, instead of features to build. The impact then can be measured in outcomes rather than outputs. Working in empowered teams boosts our motivation. We have the freedom to be creative, the autonomy to grow, and super talented people around.
Being empowered is a gift, so we can pay it forward by empowering other people around us. Among the times I am the most proud of at work, are the moments when I teach something to someone (e.g. SQL to automate a process, storytelling on a presentation, problem-solving in a specific analysis) and later I realized that person is employing this knowledge and improving her job, or even teaching others. Openly sharing helps everyone, especially ourselves. For those who want to learn more about how to build empowered teams, I super recommended the book Empowered.
“Leadership is about recognizing that there’s a greatness in everyone, and your job is to create an environment where that greatness can emerge.” — Bill Campbell
We can be missionaries in Data Analysis by promoting data culture, focusing on outcomes, being driven by actonables and achievements. Lemann says that “Having a big dream requires the same amount of effort as having a small dream”, and I think the same goes for purpose.
Believing in what we do makes the life way easier and joyful. Giving a personal example, I am a product and tech enthusiast, an engineer who loves analysis, a student on real estate and cities. QuintoAndar is at many of these intersections and makes sense for me. Doing what we believe is the key to the sense of meaning and state of flow.
I know finding a job where we feel passionate about might be a luxury that many people can’t afford. Once, Richard Feynman wrote, “no problem is too small or too trivial if we can really do something about it” on the “Do not remain nameless to yourself” letter to a former student .
10 — Always strive for more
Data Analysis is so open and sometimes that can be bittersweet. To cope with the bad and the ugly side, you can delimit well the scope of analysis and never be a perfectionist (that usually leads to analysis paralysis) and just finish the thing. Be gentle with yourself.
The good part is that this is a super-creative job. Every day will be a brand new day. Whenever we desire to, we can open a problem and ask “what if”. There will always be insights to be revealed, and something waiting to be discovered. I leave you with the remarkable quote from Steve Jobs: “Stay hungry, stay foolish”. Also, remember to have fun during the journey.
—
These were some of the lessons about actionables, problem-solving, storytelling, analytics, teamwork, curiosity, strategy and career that I learned and made all the difference for me. Hope they can help you somehow too :)
If you want to be part of this amazing team, join us!
The books I referenced here which I super recommended:
- Good Strategy/Bad Strategy — Richard P. Rumelt
- Inspired — Marty Cagan
- Empowered — Marty Cagan
- Linked — Albert-László Barabási
- Factfulness — Hans Rosling
- Storytelling with Data — Cole Nussbaumer Knaflic
- Man’s Search for Meaning — Viktor Frankl
- Flow — Mihaly Csikszentmihalyi
