Three roles in Data & Insight at Oda — Part 3: The Data Analyst
In the initial article of this blog series, Nina Walberg introduced the three roles in Data & Insight in Oda: Data Analyst, Data Scientist, and Data Engineer. She gave a general description of how the roles are defined and what sets them apart. To give you a detailed overview of these roles, we have interviewed three people working at Oda. In this interview, Rohit Patil will give you a better understanding of his work as a Data Analyst at Oda.
What is your background and what attracted you to Oda?
My academic background is in electronic engineering and I have been working with analytics for over eleven years now, primarily within the marketing and growth domain. My previous job was with the largest grocery retail supplier in Norway, where I got the opportunity to understand the nuances of the grocery retail space and also heard a lot about Oda as it was disrupting the traditional brick-and-mortar trade.
I’ve been living in Norway since 2015, and one of the things I have found rather odd over the years are the high grocery prices (one of the highest in Europe) coupled with the high concentration of privately owned grocery stores. The value chain seemed somewhat inefficient, unsustainable and ripe for a change. This realization attracted me to Oda and its mission to build an efficient and sustainable retail system. I was fortunate enough to get the opportunity to work here as the company started expanding its operations in the midst of the pandemic.
What does your day-to-day work look like?
This depends on whether we are in a flex or focus period. During focus weeks, we’re more concerned with getting things done within a fixed time and with a variable scope. This means syncing with UX designers, product managers and software engineers on new features or experiences being rolled out, defining success metrics and setting up the required data pipelines for analysis. We operate within a data mesh structure where each cross-functional team owns their data end-to-end, with data engineering providing support through tooling and by sharing best practices. This approach reduces dependencies and lets embedded Data and Insight (D&I) personnel, such as myself, move quickly with complete control.
This does not mean Data Analysts operate in silos. We have regular syncs with other D&I-ers at an area (e.g. growth, logistics, commercial etc.) and discipline level where we discuss issues, incrementally improve our way of working, and collaborate on key initiatives.
During flex weeks, we pursue alignment across teams, which means we have a lot more meetings and try to scope work for the next focus period. It’s also the period where we have hackathons; this is where D&I-ers volunteer to find solutions to interesting data problems outside their usual area of work.
I also have management responsibilities for Data Analysts embedded in other teams within the growth area, which means I have weekly 1:1s to understand how folks are doing and how I can help. I’m often surprised by how technically and personally mature people are, and seeing them thrive is really rewarding.
Which tools and methods do you use?
Since Data Analysts work end-to-end, we end up using different tools across the data value chain:
- Data collection: various tag management and QA tools for verifying the quality of data being collected.
- Data ingestion: for most use cases, we use Fivetran for ingesting data from third-party APIs.
- Data transformation: this primarily happens in dbt (Data Built Tool) and Snowflake where we primarily use SQL and Jinja to create data models and data products. This is where I’ve been spending most of my time lately, but it tends to vary.
- Data analysis and visualization: Looker is our primary data visualization tool, but we also use Jupyter Notebooks and Amplitude for more on-the-fly analysis. The choice of tool depends on the question we’re trying to answer and the decision window, i.e. how urgent is the decision and the learning frequency. For building statistical models, most analysts use Python but we have a few that prefer R.
What are you working on right now?
I’m embedded in the Growth Activation team where we’re trying to improve the onboarding experience for new customers. We use qualitative and quantitative data to answer difficult questions and try to deploy new user experiences. Unfortunately, what we think should work logically might not always be empirically true. This means we usually try to validate each new experience or treatment through rigorous A/B testing. So lately I’ve been grappling with events data to validate our initial hypotheses and also monitor ongoing experiments. I’m also researching new approaches to significance testing which should allow us to conclude experiments more quickly while mitigating risk.
We also have some data models that I’m collaborating on that should help answer key questions around user behavior in our app and on the web. These data models ultimately become the input for Looker explores that are used not only by analysts but by business users as well. We have a self-serve approach to data exploration where we encourage business stakeholders to be self-sufficient. This helps everyone in the organization understand the data more intimately and ensures that drawing insights or building dashboards is not viewed as the sole domain of Data Analysts.
How do you collaborate with your team?
The Activation team consists of two software engineers, two UX designers, a Growth lead and myself. We work in two streams and try to develop two new experiences simultaneously for different parts of the product. We usually have a weekly planning session and a mid-week check-in. Decisions around direction and prioritization are made in a democratic way. During flex weeks, we have team retros where we look back at the last flow cycle and reflect on what worked well and what can be improved. We also reflect on how things are going with the team and in Oda in general.
Some of our team members work remotely, so meetings often happen over video calls and are usually well structured. We also have long coffee chats discussing technical problems in one another’s areas which is a great way to learn about other disciplines and develop a common language.
What was it like to join a scale-up in rapid growth?
I was not completely aware of Oda’s growth ambitions as I joined just before the unicorn funding round. We have been pretty busy establishing new teams and improving processes for onboarding new recruits. Speaking of which, my onboarding process was surprisingly smooth with a good mix of orientation sessions and hands-on technical onboarding workshops. I also had a lovely “buddy” who showed me the ropes outside the organized workshops. The onboarding process has only gotten better since my time and continues to be a shared responsibility across areas.
Oda’s decentralized way of working across cross-functional teams was relatively new to me and the fact that, as we have grown, we’ve maintained a close-knit community of D&I folk is quite remarkable. Personally, the intra-discipline and cross-discipline interactions have been extremely enriching and I’ve learned a lot over the last year or so.
How do you think the Data Analyst role in Oda will look in five years?
Interesting question. I believe there is room for domain specific specialization. We already see the need for analysts that have an in-depth understanding of the Growth domain, so it’s only natural for more specialized roles to emerge. However, I’m not completely sure if this is a good thing, since domain experts are more prone to tackle problems with strong priors. Perhaps we’ll see the role rebranded to better represent the outcomes, for example, “decision analysts“.
As we expand into new markets, we’re bound to see even more diversity in cultural backgrounds, which is super exciting. I would also like to see a gender ratio that’s less skewed than it is now.
I hope this article has given you some insight into what the working life of a Data Analyst at Oda looks like. If you’re considering joining our growing organization, take a look at our job postings. If you want to know more about some of the other roles within D&I, checkout some of these other posts:
- Introducing the three roles in Data & Insight
- The Data Scientist interview
- The Data Engineer interview
Thanks to Kate Kuzmina, Hilde Dybdahl Johannessen, Hanna Heggheim Lee and Siri Bruskeland for their editorial inputs.
