Over the last 3 year, Level alumni have moved on to work in a number of data driven roles, in data analytics and science, research, marketing and programming, to name a few. Since graduating Level, alum Josh Chang has been enjoying his career at HubSpot as a Senior Marketing Manager, using his data analysis skills to find “single sources of truth” to determine the right metrics to measure their efforts by. We got together with Josh to talk about how data analytics has been a key tool for his success. 

I was initially exposed to data analysis was when I was taking an Excel-heavy supply chain and operations management course during my senior year. We were doing pretty bland problems, but this is when I realized how powerful Excel and data analysis could be. Unfortunately, being in college and nearly done with my marketing and international business dual-degree, I didn’t pursue it much further after I graduated and started work. However, I kept finding new ways to leverage data analysis in my professional and personal life.

I thought it was a blast (I’m easily entertained) to figure out how to use Excel and Sheets to automate tasks, like splitting vacation costs with my friends, tracking statistics for our slow-pitch softball league, and determining if we’d become overnight millionaires when friends and I pooled money to buy 150 lottery tickets together. Once I started doing paid search and social for a boutique ad agency in Boston, I hit the ground running.

Every day at HubSpot is different, and it can be ridiculously fast-paced. Our core responsibilities consist of managing HubSpot’s paid advertising strategy for freemium acquisition and lead generation across the globe, day-to-day budget management, and optimization, and developing reporting and analytics to help our team and the company understand the impact on the business. On top of that, we work with the product team on HubSpot’s in-house paid advertising features, communicate learnings and results to internal stakeholders, or work with customers on their own paid ad strategies.

There are a number of analytics tools used at HubSpot. On the marketing team, we are heavy users of our own software, which helps us measure the entire marketing funnel from prospect to customer. We lean on Google Analytics for web analytics to see paths-to-purchase on our web properties. To gain even deeper insight we use Looker, which is a powerful business intelligence and data analytics platform built on SQL that helps the company merge dozens of data sources across teams into a data visualization user interface that enables real-time business intelligence.

We have whole teams dedicated to building and managing the company’s data infrastructure, but one of our core responsibilities is figuring out how to leverage all of this data into actionable insights our team can use to make decisions and improve our campaigns.

Lastly, we also use Amplitude, which gives us detailed analytics and usage data for our actual product. It is really important for us as marketers to understand how the product is being used so we know what actions our most valuable users are taking and where there are gaps in the product to improve.

One of my favorite projects so far has been an ongoing, in-depth analysis of our campaign-data to figure out what campaigns, ad groups, keywords, and ad creatives are the most and least valuable from a Lifetime Value and return on ad spend perspective. As a SaaS business, return on ad spend is much more complex than it is for an e-commerce company or a business model with a shorter customer journey. Figuring out how to measure from the top of the funnel all the way to purchase and assign different “credit” to different lead/customer touchpoints has been really challenging and interesting and is a constantly evolving process.

One of the most difficult parts of working as a data analyst, especially as part of a marketing team, is establishing “single sources of truth” in your analytics infrastructure. It is challenging to understand which metrics are the right ones to look at, or which ones are the leading indicators of success. Oftentimes marketers might get hung up at higher-level metrics like click-through-rate or cost-per-click when it is much more important to look deeper at metrics that are more indicative of impact to the business and the bottom line.

The skills I learned at Level have been instrumental in my success at HubSpot. From a technical standpoint, Level gave me a great base to continue building my SQL skills, which have been key to learning powerful BI platforms like Looker. While the whole point of Looker is so that users do not have to write code, having a solid understanding of SQL and LookML (Looker’s data modeling language which is based on SQL), has helped me become a much stronger user of the platform.

My advice for getting started with data would be to get REALLY GOOD at Excel or Google Sheets. It is such an accessible way to start understanding the power of data analytics, and doing that first helped me understand that this was something I was passionate about and wanted to continue to pursue technical skills in the data analytics world. Building on that, regardless of what technical skills you need, it is really helpful to have a real-world problem or project, either at work or on the side, to apply data analytics skills to.

Classes and case studies are great, but you really start to understand and learn how to apply data analytics when you are working on something that you are truly invested in. At the end of the day, if you come up with a cool data analysis or analytics project, chances are there’s a way to do it, you just have to figure it out.

If you are looking for a career change, like Josh, it is never too early to get started playing around with analytics on your own. Starting with Excel is great advice, and when you are ready to go deeper, a bootcamp program like Level can help you hone your analytical prowess and prepare you to excel in a new role.

level data analytics


This site uses Akismet to reduce spam. Learn how your comment data is processed.