Dwolla: Finding a scalable solution with Tableau and AWS

Dwolla is a U.S. based payment company that offers real-time payment solutions. When Dwolla adopted Amazon Web Services (AWS) to scale their infrastructure, they needed an analytics tool to leverage their data. The requirements? The tool had to connect directly to their data sources and quickly deliver insights. More importantly, it had to be able keep up with the growing company’s data needs—without breaking the bank.

Dwolla is an e-commerce company that provides online payment solutions with the click of a button. The company mission is to help members move money quickly while keeping costs low.

Dwolla adopted Tableau as a low-maintenance way to interpret data and provide immediate insights.

In video 1, Fred Galoso, Software Developer, talks about how Tableau can connect directly to multiple data sources, like Amazon Redshift, Amazon Elastic MapReduce (EMR), Microsoft SQL Server, MySQL, and PostgreSQL. Tableau’s ability to quickly pull data allows them to move from time-consuming spreadsheets to powerful dashboards.

In video 2, Fred describes how report creation was once an IT-intensive process that required a lot of engineering resources. With Tableau, they can author dashboards and spot trends—in a couple of hours instead of weeks. Today, teams at Dwolla use Tableau to analyze customer demographics, manage product development, and detect fraudulent activity.

Tableau: What attracted you to Tableau?
Fred Galoso, Software Developer: Tableau's ability to connect with Amazon Redshift, to connect with Amazon RDS, to be able to tap into our elastic MapReduce services, that's a big part of why we picked Tableau— is its ability to connect to all those different data sources.

Tableau: What is the benefit of using a cloud computing service like AWS?
Fred: Our move to Amazon Web Services was to keep up with demand and scale up our infrastructure. About a year after, we came to the realization that we have, as part of that demand, all this data now as a side effect that we need to process. So we've been leveraging a lot of Amazon's products to store, process and have that data at least available at scale. Tableau came in after that to be something to surface that and bring those insights to the forefront.

We’re a small, fast-moving company. And it's all about reducing time to insights for us. So both using Amazon Web Services to be able to scale up very quickly and on-demand, but also to respond to all the different data questions that we may have, and really being able to answer what-if as quickly as possible.

Tableau: Was cost a factor when choosing a data solution?
Fred: Total cost of ownership was one of the main reasons we moved to AWS, and in conjunction with that, Tableau. So with AWS specifically, we're able to store data and also process that data at scale for a fraction of the cost of some other commercial providers that are out there. And then Tableau, of course, is a natural fit and saves us a lot of time, effort and engineering resources to be able to create those insights.

Tableau: So time to insights was particularly important to you?
Fred: If we look at our mission, for example, of moving money as quickly and as low a cost as possible, both AWS provides us the speed in terms of our infrastructure, and from an organizational standpoint the flexibility to change. With Tableau we also get that speed. It's really performant, but it's also really flexible and really easy to adapt anything that we need to change to.

It was really easy to be able to connect Tableau with AWS. There are already built-in connectors for the different databases that we have. For example, we have data in SQL server, MySQL, Postgres, Redshift. So we're able to connect to all that out of the box with Tableau.

Tableau: Did you find it easy to connect Tableau with AWS?
Fred: It was really easy to be able to connect Tableau with AWS. There are already built-in connectors for the different databases that we have. For example, we have data in SQL server, MySQL, Postgres, Redshift. So we're able to connect to all that out of the box with Tableau.

Tableau: How do you see Tableau complementing AWS in your company?
Fred: With Tableau I often run into cases where I thought I had an idea of how I wanted to visualize something, and then I'll just be sitting side-by-side with say someone from marketing, a marketing manager or a product manager. And we're just working together trying different combinations.

I really think the combination of the two is really exciting, AWS and Tableau. Essentially, you don't have to think about, am I going to be able to answer the questions I was today, even when my company is ten times, a thousand times, however many times it is. And there's all of that however much data you need to be able to look at and process.

Powering the payments network with data

Tableau: How do you use Tableau?
Fred: For market research and customer demographics or from a sales funnel perspective, we use Tableau to get an understanding of who's using Dwolla, why are they using it, and what products are they using to be so successful with Dwolla.

Tableau: What kind of data are you analyzing?
Fred: We're looking at any sort of data that we need to power our payments network. So be it customer demographics, looking at our products and managing our products, or things like fraud and compliance, it's a big part of making sure we're protecting our members and keeping the cost of moving money in the Dwolla network as low as possible.

So let's say a fraud analyst is looking at registration, for example, or has noticed new accounts. They've been studying and looking at where some of the hot spots are for high-intensity financial crime areas. So they're able to understand and be at the forefront of those trends. And from a visual standpoint, they're able to just pan and zoom and see all the different types and zones that there are in a map without having to just pull down a data sheet.

Tableau: What were some of the challenges you were facing before Tableau?
Fred: A lot of the challenges we had with the existing BI platform we had were engineering or IT-intensive stuff. So just keeping everything up, getting data to that existing platform was just a lot of time and management that we had to do.

In addition, creating new reports or dashboards, if we wanted any sort of flexibility, required a lot of developer or engineering resources.

Tableau has essentially allowed us to go from weeks to author a dashboard into hours and days.

Tableau: How has Tableau allowed you to streamline your analytics process?
Fred: We’re able to essentially plot everything and look at clusters and colors, and trend lines over time. Being able to see those trends in Tableau from a glance is a lot easier than looking at a spreadsheet.

Tableau has essentially allowed us to go from days to even weeks to author a dashboard into hours and to days.

Tableau: So what are some next steps?
Fred: Really, what we're hearing from our colleagues is, “we want more.” We've been able to tap in or migrate our dashboards. In less than a year, we already have hundreds of workbooks and data sources. We just need to get more of it in front of our colleagues.

Tableau: What features of Tableau have you found to be the most helpful?
Fred: One of the favorite vizzes that I've built is a map that actually shows various different areas of different kinds of customers we have. So a lot of the mapping tools in Tableau are a lot of the capabilities that we didn't have before. So, for example, being able to show high-intensity financial crime areas and where we have hot spots that we need to be on the lookout for.

You might also be interested in...