Back to past episodes

How to Operationalize Data In Marketing Ops

With Jason Bilog, Data Operations Manager at Patreon

Jason Bilog, guest speaker "How to Operationalize Data In Marketing Ops"

Overview

Jason Bilog transitioned to Data Ops from Marketing Ops. Hear about how to best harness data, what middleware makes your life easier in the product pipeline, what readiness of operationalization of data means, and how to build a business case for new tech.

Meet Jason

Jason Bilog is a ​​PMM - Marketing Automation, Scaled Communications at Google. At the time of recording he was the Data Operations Manager at Patreon. Jason is a data-driven tactical marketer with a focus on marketing analytics, marketing automation and Salesforce CRM. He has experience working in the education, healthcare, hospitality and SaaS sectors. 

Top Takeaways

Moving from Marketing Ops to Data Ops

What led you to want to move from Marketing Ops to Data Ops?

In other organizations I've been at, I feel like a lot of the data that go-to-market teams use really is just within your CRM. However, at Patreon we have a lot of other information beyond just what we have in our CRM. I started to think about where the data lives. What can we use? How is this information relevant for our marketing and sales, what we call creator partnerships teams? 

I'd worked with my colleagues in data science to really help allow me to understand what data we have available. About six months ago we got a tool called Workado to help us do some ETL data pipelines to help connect a lot of data into Marketo, Salesforce, Pendo. As I worked on this you might say I started to lean more towards data science, data, even engineering. Yet I was doing that within my role of go-to-market. I finally asked, “Hey, should we have these responsibilities fall under a more traditional data team?”

That’s when the switch was made for me to be Data Ops.

Middleware

How do you actually solve having some of that product aggregated information pushed back into the CRM through middleware?

We use Amplitude for our product analytics. All this information within Amplitude, where we stood up a whole process last summer around getting this information into our Redshift, which was an incredibly big project.

Now within Redshift we have a bunch of tables on product utilization. We can look up anything from just how much a CRA's making on the platform to what kind of questions they might have.

We use Mailgun for our transactional product-focused emails on Marketo. We use Salesforce and Pendo, plus a bunch more. The bottom line is that we have as much as possible of everything going into Redshift.

From Redshift, we work on how we can get data out of Redshift and into the hands of go-to-market.

Readiness for Operationalization of Data

What does readiness for operationalization of data mean? How do you ensure that the data you have is operational and ready for use?

Data is used for analytics and tracking versus operationalization of data. In operationalization of data, there's a much higher degree of real time availability of information. 

For instance, what is the user doing with the product? Is there anything new that's happened? Whereas with general analytics, it's much more, “Hey, like we need this information to build a model to do some sort of analysis around like what's happening.”

Timeliness of information is important. When we think about how you're moving data from a production MySQL table to a table being manipulated within another schema, that timeliness is huge because, you know, you don't wanna be querying something that's in production, it might tax your system.

It goes back to, “Hey, go-to–market: What exactly do you actually need? What is the business case here? Most importantly, how quickly do you need this information?”

We use Datafold, which helps with data observability. It sits on top of our Redshift instance. It provides this metadata of what exactly is happening within your data warehouse.

It's really helpful to understand the prominence of different data. Look at how is the table created and what columns are available. What's the primary key? As a marketing operations person, you can work with someone within data science to understand, “Hey, this is what we're trying to do. These are the segmentations that we're trying to create. This is our use case. What do we have available?”

From a frequency perspective it's very easy to understand how many times a table is updated. Like this is what we have here. Then you can operationalize the data and take action on it.

Building a Business Case for New Tech

How do you win the case for new tech for Marketing Ops?

While Marketing generally understands that Marketing ops is more of a technical role, other parts of the organization may not understand that. You need to build a defined business case for new tech in Marketing Ops.

A big part of building that business case is going through receipts. If you have tickets that have been created over a year ago and put on the back burner, that's a great example of showing how something is a problem that continues to be a problem.

The second part of the equation is working within your infrastructure. Reach out to the teams affected by the lack of the tech you’re making a business case for. Help them to understand the cost benefit to them of the new tech Marketing Ops can use to make their lives easier.