In this edition of Off the Charts, an interview series where we chat with individuals doing industry-changing work in the data space, we spoke with Jake Stein, Yaniv Leven and Dave Fowler about the current state and future of data technologies.
Jake Stein is the the CEO and co-Founder of Stitch, an ETL tool that inspires and empowers data-driven people by connecting and consolidating data sources.
Yaniv Leven is the CEO and co-Founder of Panoply.io, which provides an end-to-end data management-as-a-service platform.
All three of you have extensive experience with both traditional and modern data stacks, how have you viewed their transformation over the past few years?
Jake: One of the things I’ve seen change over the past few years is that the modern stack requires fewer things to be planned up front. If you had traditional tools and wanted to analyze data, you would have had to do the prep work weeks or months in advance with an ETL process set up, a data warehouse provisioned and a data analytics tool in place. And again, all of this had to go through your IT or technical team. Asking questions of your data was a long and slow process.
I’m excited that now not only can you ask questions and get answers quickly, but ask questions that haven’t been asked before. With tools like Stitch, Panoply and Chartio, you can get all the insights you need without being a Data Scientist or needing your Engineering team to build reports for you.
Modern tools allow companies to be more agile and eliminate the latency of waiting on IT to answer your question. Now, you can build new reports in minutes or seconds, which leads to more question asking and better insights.
Yaniv: All the companies I’ve worked for have been around the evolution of SaaS technologies in the data space. Ten years ago, I worked at a startup that was just starting to contemplate outsourcing parts of our analytics solution. At this time, no one had thought about going outside the company for an analytics solution, let alone going to a third-party ETL tool. These were the days where I also shared an AC unit with our physical servers. Nine years later, we’re in a much different scenario.
Now, we’re able to automate our data ingestion processes and can choose from a variety of vendors to fulfill every part of the data stack. No one starting a company today thinks about having a server stack in their office.
Dave: One big transformation I’ve seen is the release of Amazon Redshift, which came out five years ago. Before, if you wanted a BI product, it was bundled with ETL and a data warehouse. Amazon Redshift broke up the entire process of buying BI products. All of a sudden you could buy parts of the stack separately and it resulted in some really focused companies creating great best of breed components.
Now one of the new trends I’ve seen is the volume of data organizations aren’t storing themselves but have in their cloud applications like Salesforce, Hubspot, etc. There’s so much data and you don’t have just one database anymore. With all that data, there’s a desire for a single solution again, and there’s an opportunity for those best of breed components (ETL, data warehouse and BI) to partner together and giving people the ultimate experience.
One thing we’ve noticed is how more business users are becoming involved in setting up, using and buying data solutions. Are you seeing this trend with your customers?
Yaniv: One of the main reasons we built Panoply was to enable business users to harness technologies that were previously monopolized by IT. And one of the main reasons we partnered with Stitch and Chartio is because when we talked about the vision of our companies, in our specific domains, we had the same vision of streamlining, simplifying and making the data process as frictionless as possible. With that, most of the buyers of Panoply today are not Engineering or IT, but business users.
Jake: For Stitch, when we started out, most of our users were developers or in IT, and we’ve more recently seen more business users as well. We’ve seen that developers and IT are reluctant gatekeepers—they don’t necessarily want to play that role, and business users don’t want that either In the old model of analytics, it was not desirable, but it was necessary.
Now, if we can abstract away a lot of the things that IT has to do like scaling hardware, notifications, credential management, etc. and pare down the experience to just the essentials, that allows a wider group of people to use data to be more effective at their jobs. That’s our goal with Stitch, and we have carved out straightforward and focused use cases that target business users and eliminated the need for gatekeepers.
Second, we focus on integrating well with other tools, which is one of the reasons why I’m excited about this partnership. Business users care about the end result, and we need to provide them with the best joint user experience so that they can take the right action at the right time with minimal effort.
Dave: I often ask people, “who at your company knows your data the most?” And people usually point to someone in IT or a data analyst. And I disagree (it’s a trap question) because while those people certainly know where and how the data is stored it’s impossible for them to also keep up with all of the context around the data. The people that really know about the data are those generating and relying on it, for example, people in Marketing running the AD campaigns. The only thing holding marketers back from “knowing it” per se is access to data and the ability to explore it. That’s what Chartio is enabling, and so we’ve seen a large part of our user adoption around business users.
And as Jake said, the biggest bottleneck for this enablement of the business user is that IT is seen as the gatekeeper of data, even though they don’t necessarily want to be and people don’t want to be gate kept. That’s why Chartio, Stitch and Panoply are partnering here. We’re all here to help organizations get past that hurdle.
How does Chartio, Panoply and Stitch Data fit into this new narrative for end-to-end analytics and BI, in regards to the customer experience that would enable access to data by all?
Jake: We’re very conscious to the fact that Stitch is not useful on its own, and that’s intentional. We’re pulling data from data sources, sending it to a data warehouse like Panoply, and then the data is analyzed with a tool like Chartio. If you just used Stitch without all of those complementary products, you wouldn’t get much mileage out of it. So our key focus is on being a really great unit in this modern analytics stack.
For context, I previously was a co-founder of a company that tried to do the entire stack, end-to-end, and it was difficult. We found it hard to focus on everything and spent only a small amount of time on each part of the stack. So with Stitch, we felt that there are already world-class tools out there to do the other parts of the stack, so we could focus on ETL and have great integrations.
Yaniv: Panoply is essentially middleware - you don’t see us. We have data connecting to us on one side coming from an ETL and BI or data analytics on the other end. We understand that the end-to-end data stack is a three part process:
- Data Warehouse: smart management of storage and data technology
- ETL: data ingestion to different data analytics tools
- Data Analytics/BI: visualizing data for insights
There have been companies that have tried to focus on all three, but again each of these areas are a world of their own. Now, the whole industry is going through a paradigm shift and moving from static, traditional technologies to more adaptive tools for end users.
Dave: There are two ways that customers approach putting together the stack that Yaniv described.
- Some start with a data warehouse, then an ETL and finally a BI tool. That’s how a Data Scientist or Data Engineer thinks about the data stack, often beginning the conversation with understanding where they’re going to store the data. This definitely works and will continue to work.
- But most people looking to work with data start by looking for a tool to explore and analyze their data (the BI tool) - they haven’t really started thinking about an ETL or data warehouse yet. This second market has been largely underserved, but this is where we are going.
Along the same lines, what’s the future of analytics and how do you see your companies aligning for it?
Yaniv: When we talk about the future of analytics, for Panoply, that’s the world of augmented analytics and machine learning automation. What we try to do is make data warehousing more agile, smarter and work for you. So for Panoply, we’ve incorporated machine learning algorithms into our warehouse making it learn and self optimize for you.
Jake: Our mission at Stitch is to inspire and empower data-driven people. We believe that the companies that use data to make decisions tend to grow faster and are more effective at reaching their goals. The way that Stitch plays a part in this growth is by enabling people to access their data and get it consolidated and in a useful format. We are continuously working on adding more integrations to more tools to our product.
Today, we integrate with 60+ data sources and we have received over 600 requests for others from our customers and prospects. For us, the future of analytics is creating more data integrations and also working with our partners to create a seamless user experience in each part of the end-to-end data analytics stack.
Dave: My prediction of the future is that the process gets continuously easier with a lot less friction which enables more and more people to explore and understand their data. And as things get easier at every layer of the data stack, the gains are exponential for everyone. Through Panoply, managing and scaling Amazon Redshift will be seamless, users can rely on Stitch for ETL without having to write a bunch of code and Chartio enables easier data exploration. With the combination of these three, it’ll just make everything easier and people can focus less on tooling and more on getting insights.
Do you have any advice for companies looking to move from traditional data analytics to a modern data-for-all analytics model?
Jake: Do it! Don’t let perfect become the enemy of better. These new tools are so agile that getting up and running is easier than ever. People should approach modern, end-to-end data analytics with a different mindset where you can start with one data source going through Stitch into Panoply and then analyzed in a single dashboard in Chartio.
You can get start getting value today. You won’t migrate your entire company or business onto this new data stack today, but you can validate that it works and you can show the other members of your team its benefit. I’d recommend starting small.
Dave: Jake is spot on. One of the benefits of moving to a modern system is that it’s agile. So just embrace that benefit early and realize that you don’t have to do it all at once. You can start small and just build. So do that, start small (and now) and let it grow.