Mobilizing Data to Accelerate Cloud Data Warehouse Success

Dayle Hall:

Hi, and welcome to our podcast, Automating the Enterprise. I’m your host Dayle Hall, the SnapLogic CMO. This podcast is designed to give organizations the insights, best practices, and ideas on how to integrate, automate, and transform their enterprise.

Today, we’re excited to have a very special guest from that very small company you may have heard of, Google. It’s actually a rather large organization, but we’re going to focus on a very specific area. And we’re delighted today to have Jobin George, who is the solutions architect of data and analytics partnerships at Google. Jobin, welcome.

Jobin George:

Thank you, Dayle. Good morning, good afternoon, and good evening, everyone. First of all, thank you, Dayle, for having me. Really excited to be part of this podcast series.

Dayle Hall:

Of course, of course. We’re very excited to have you. I know you’re doing a lot of work with Google, specifically around cloud data warehousing. But let’s kick off with getting to know you a little bit. Obviously, we’ve all heard of the company name. How did you get into this current role? And how did you become what you are today, which is focusing very much on data analytics partnerships there?

Jobin George:

Hi, again, everyone. My name is Jobin George. I’m a staff solutions architect at Google focusing on analytics partnership, as Dayle mentioned. My primary responsibility is to work with partners like SnapLogic and build integrations they do at Google Cloud and create joint solutions to offer to our mutual customers.

Prior to working at Google, I held a similar role at Amazon Web Services, where I was, of course, fortunate enough to work with SnapLogic in the partner solutions architect organization. Prior to that at Netgear and Hortonworks, I held roles similar to what I am having now, and most of my career was around data and analytics. And I enjoy partnerships, partnering with people, and I just know that’s the best, what I can do a lot with partners like SnapLogic for making our customers’ life easy.

Dayle Hall:

Well, we like that. We definitely like our lives easy. And of course, the key to a partnership is making your life easy too. The fact that you’ve worked with us twice tells me that we’re not screwing up this relationship job. And so I’m happy with that. That’s a good start.

Jobin George:

Likewise.

Dayle Hall:

Okay. Look, I see this everywhere. We talk about this at SnapLogic. I know you talk about this at Google. But obviously, this massive market trend, this big topic around cloud data warehousing, Google are all-in along with a number of other vendors.

But let’s start with this. Why now? Why have you seen so many projects, big cloud data warehouse projects? Why are they growing so quickly and becoming such a part of what enterprises are looking to do?

Jobin George:

Like I said, in recent years, we have seen a lot of growth when it comes to cloud adoption across industries and especially enterprise. And recent studies show that cloud data shortage accounts for more than 40 percentage of all the enterprise data. And the number is still growing. If you double click, you would be able to see that we live in a digital world. Every moving metal has a sensor attached to it. Everything you do has some digital significance. Your watch, your phone, your car, everything is emitting data.

And for businesses to understand that and make use of that, capture that and grow their business, they have to look into the data. The more digitizing the world is, that is going to grow. And unlike old times, nobody has time to wait for their infrastructure to come. People are looking at agile methods to grow.

In the digital world, there is no surprise that cloud storage and cloud data warehousing is growing. To answer your question, I feel like in recent years, that style, how we anticipate and see things have changed in the digital world and it’s exploding, and we only see that trend going up.

Dayle Hall:

For sure. I’m actually surprised. I think you said the amount of storage is about 40% and growing. That’s what we hear the most in terms of people moving to the cloud. There obviously are some organizations that want to still keep some things potentially on-premise, but it becomes less of a priority, I believe. And everyone we’ve talked to on this podcast, we talked to someone yesterday, that’s focused on people analytics, Aramark, and has just moved to Freeman Group. And what he was talking about is that cloud has enabled so many opportunities for organizations to grow, to give data in the right space but also that enterprise-focus, which is obviously how can we cut costs, which is obviously a big concern as always. 

Jobin George:

Totally. Many companies, they don’t even have the resources to run their data centers so that it’s easy for them to run the show like they used to in the past. Nobody is able to scale as they like. Even though they have some presence in on-premises, it’s not easy to keep the show running forever because it doesn’t scale. And if you start using cloud and you still have something running in your data centers, it’s not easy to compare. and you probably will get rid of it as soon as possible.

Dayle Hall:

For sure. And we see that too. Are there any trends that you’re seeing from specific industries? Is it specific use cases or lines of business that are driving this? What are you seeing when people approach Google for these kinds of things? What generally is driving that? What’s their original goal? Do they have use cases, business goals they’re trying to solve? How are they approaching you?

Jobin George:

Cloud data warehouses could be at a petabyte scale today. And then there are small and medium customers to enterprise customers who have different, different needs. And it’s not practical for everybody to have their own data center and grow it as they need. And then, of course, the cost concerns you mentioned, and everybody wants to make sure that you don’t break your bank.

How do you do that? You need something which is agile and scalable and easy to manage. We have customers starting from different verticals like healthcare, or automotive, and retail, insurance with different, different use cases, starting with predictive analytics to forecasting, log analytics, anomaly detection, AI/ML use cases. Five to six years back, seven years back, there were not many AI/ML use cases which were around. And then right now, you can’t find any customers who don’t have at least one use case of AI/ML. And then traditional data warehouses will not be able to satisfy that need.

And then, if you look at customers who are in this space, they have more and more streaming capabilities which are needed. In order to accommodate those, you will have to have something which is quickly adopting and changing, which traditional data warehouses cannot run along. It’s easier. 

Dayle Hall:

What I love about specifically the opportunities around cloud data, SnapLogic is an integration company, integration, automation. And what we usually tell customers is start with a use case, start with the business goal that you’re trying to solve, and then we’ll show you how to solve it. What I think cloud data warehousing and the opportunities there are driving is there’s so many opportunities. It can impact every single business case.

You just mentioned this yourself, which I think is key. There’ll be people out there looking at I don’t care if you’ve got one AI/ML use case or 15. By leveraging what’s available around cloud data warehousing, you can actually get a lot more value from that. Because let’s face it, it needs the data, fast access to the data, more data points to actually get the most out of AI and ML. There are two trends that are helping each other to keep growing and growing. More data in the cloud are available, more use cases with AI and ML, and they help each other to keep building. It’s an exciting time for this technology and this trend.

Jobin George:

Totally. To SnapLogic, low-code/no-code, I don’t have to explain how easy you make it for the customers. Similar to that, customers using data warehouses, if they don’t want to know or they don’t have time to know about how ML works in the background, there are different types of users who want to use it. And the new generation of cloud data warehouses, BigQuery, makes it easy. You just call a function and then you can get it in fingertips out of that. It comes with embedded integration, which is not readily available anywhere else.

It makes the life of a customer easy, and then they don’t have to spend time on writing codes or making things, creating something by themselves. It is a natural, organic way customers get used to it and then they demand more and then it goes. And both of us, like cloud data warehouses and the partners like SnapLogic, support it and everybody evolves and then you move forward.

Dayle Hall:

There’s obviously been a ton of other developments in and around technology that helps support cloud data warehouse initiatives. We talked about AI and ML. There’s this term that we’re all familiar with in the industry of ETL versus ELT. And I’m sure most people understand. But for those that don’t, extract, transform, load versus extract, load, transform.

I don’t want to get into a more technical discussion. But between that discussion or other technologies that you’ve seen, is there something that you could point out that says this was a key factor in these initiatives becoming more prevalent or more successful? Is there something specific around the technology? I know you’re going to talk about the BigQuery program, which I’m excited to talk about. But is there something specific? You said this was a key. All of a sudden, this will open the floodgates.

Jobin George:

Like you can imagine, even the cloud data warehouses a couple of years ago compared to today have gone through a lot of changes in recent years. Newer customer challenges, newer use cases, it just evolved and you get to adapt to it. Like you mentioned, in AI/ML integration and in all open source and storage compute segregation, all of these things are key to how customers would look at a problem and then try to solve it.

When it comes to data movement, I’ve used SnapLogic myself, and then it gives you the flexibility. Customers want the flexibility. If they wanted to push down the query rather than spending compute cycles elsewhere, they wanted to make use of the data warehouse itself in order to transform their data rather than doing it elsewhere and then spending more cycles and then coming and then spending more time within the data warehouse.

It just evolved. And then customers, what I have seen in recent years is they’re adapting more and then they’re moving to the way they can achieve the same thing in a better cost-efficient way, which I’m sure partners like you support a lot and then make the customer challenges a little bit easy when they move forward. So yeah, I’ve seen a lot of changes in how they use cloud data warehouses in conjunction with products like SnapLogic.

Dayle Hall:

Yeah. And I feel like we’re still in the infancy of this. There’s a lot more room to grow. I think we’re going to move to the point where obviously more and more data is in the cloud. In your organization, with the partners that you work with, how did the pandemic help to either accelerate or create new opportunities?

I’m going to give you an example that we heard. It was about six months into the pandemic, it was one of the analysts. You know the analysts, they generally know everything about everything. But this was actually a great insight that I actually really enjoyed. One of the analysts said at one of our customer advisory boards, that when the pandemic happened, the organizations that thought that they had all these integrations and the automations and everything was working fine, the second people were out of the office and someone wasn’t sitting next to someone else, they realized that there were so many manual pieces of process that they’d thought they’d automated, that they thought worked, but it was still someone being prompted in an office, it’s someone being around and moving something. And when everyone was suddenly remote, a lot of those things broke down.

Now that’s around automations and certain types of integrations. And obviously, those can be addressed. But did Google or did you and working with other partners, did you see something that said what happened during the pandemic helped to accelerate? Was there something specific you thought — again, no one’s looking to take advantage of a global pandemic. But let’s face it, we’re here to help customers. And if we can help them be successful during this tough time, that’s our job.

Jobin George:

In my opinion, and what I’ve seen in the field, the pandemic played a substantial role in accelerating the cloud consumption, particularly cloud data lake and data warehouses. When people are at home, they’re not moving. Internet traffic increases and the traffic means more data, consumer behavior changes, usage pattern changes, e-commerce soared. And then companies adopted remote work use, as you mentioned, and then hybrid working and then schools closed.

But the internet consumption, it just soared and people are attending classes from home. They’re attending sessions from home. The events which we used to participate in, everything went virtual. And then it is actually driving or creating different patterns. Now after 2 ½ years, I attended an in-person event and then I saw how it is to connect with these people. When the trend is moving in a different direction, behaviors change. There are companies who thrive because of so and so reason, and then they need to capture all this data, which is a byproduct of these changing trends.

And if they don’t analyze it, they’re losing out on the benefit, make use of the data and provide it back to the business so that they can make their business thrive. They decided, okay, you know what, I’m going to capture all this data in my data warehouse, in my data lake, and then analyze it so that I can make more out of this. Just as a personal example, I don’t want to speak too soon, but the pandemic is almost behind us.

Dayle Hall:

We’re all praying for that, Jobin. We’re all praying for that.

Jobin George:

Exactly. From my own experience, I was never an online shopping guy. I have to go to the store. I have to see what I was buying. I have to feel what I was buying. A little more than 2 ½ years into the pandemic, I’m thinking if I can order something that I want sitting on my couch, I can get it the same day, next day, in two days, why should I get up from my couch and do that? There are millions like me who contribute heavily towards this huge data creation. The trends are changing. And then in order to adapt to that, our mutual customers will have to have cloud data warehouses, which can scale and deliver what it really takes to get their business going.

Dayle Hall:

There’s usually some shifts in our environment. Pandemic is obviously something that no one really saw coming, but things change and we adapt. And like you said, some behaviors change and it generates different types of data. And here we are now about to come out of that, like you said, hopefully, or at least not have to go through it again as seriously. And then behavior’s changing again.

All of a sudden, I don’t think it’s going to go back to what it was before, but then it’s new patterns and other new behaviors. So many things that can impact it and it keeps things fresh and ready. And these are the technology trends that I think are just exciting to be part of. I’m excited about that. Obviously, Google is investing heavily in this area. You have something that you just want launched called BigQuery. Talk to me a little bit about what that means, why Google is investing in it right now, and how you’re using that with some of your customers today.

Jobin George:

Google BigQuery is one of our flagship data warehousing solutions, which was around for some time. And then our investment in that more and more, it’s getting a little bit more prominent. And then we see all these trends and growing customer base and the demands from customers are growing and it is adding features left and right. Every quarter, you see a new feature coming out, multiple features coming out. And then customers are still there asking for more.

We understand their patterns and behaviors are changing. And then from the partnership perspective, we are also trying to put a new spin on that, creating programs which go well with it, working with partners like SnapLogic. We are trying to build a vibrant ecosystem of partners so that we can cater towards each and every need of customers, especially BigQuery invests a lot into building a network of strong partners so that we can jointly help them.

We recently launched a program or initiative called Google Cloud Ready – BigQuery with respect to that. And then we have so many partners participating, including SnapLogic, who was a launch partner who worked with us in the prior case of that program. And as you might have seen Thomas Kurian talking about doubling down on our partnerships and channels. What better time for BigQuery partners to help our customers.

Dayle Hall:

Cloud Ready – BigQuery partner program, and obviously we’re part of that. And I did hear how Thomas talks about that, and I came from Cisco and Oracle and Aruba Networks, and we worked a ton with channels. Obviously, I’m a big channel advocate. I love working with partners. It actually is probably the only way that real enterprises can scale. How do the partners work with you?

Let’s say I’m a customer out there and I’m trying to figure out how to work with Google, potentially another partner. What do I need to know about how the solutions are validated between us and you as part of the programs? If anything, I know that’s been de-risked because we’ve already worked together to make sure these technologies work.

Jobin George:

Our BigQuery validation initiative has three phases, evaluate, enhance, and enable, namely. During the evaluate phase, we have a dedicated partner engineering team who run a series of data integration tests and in a sandbox production environment and compare results with the best practices and performance benchmark our product team have. 

Then we take the results of the evaluation phase and work with partners to fill in any gaps we found during this phase. We call that as an enhancement phase. Partner team collaborates with both Google partner engineering and the product team during this phase. And we work towards adopting newer features as both the products evolve.

The last phase we have is an enablement phase during which, jointly with our partners, we refine existing documentation so that our mutual customers have everything they need to successfully implement both the solutions together when they’re designing theirs. We also provide go-to market aspect of the initiative and we support the partners out there as well. All of this, just to give a better experience and benefits to our mutual customers.

Dayle Hall:

Again, if I’m a customer out there and I’m thinking about who to work with, knowing that there’s that rigor and support during how the technologies, how the companies work together, we know that enables, like I said, customers to potentially de-risk. So when all the benefits come, they get faster time to value, they get better ROI, and they actually can get the insights and leverage these technologies a lot faster and get a better return.

Let me ask you a question specifically around who’s using these technologies. And I’ll give you our example, what we see from SnapLogic. Obviously, IT is still a big stakeholder, a big buyer of technology. They want to enable the lines of business. But we’re also seeing more requests from lines of business, more people from different functions in the organization that want access to different data. They don’t necessarily sit in their organization and say, hey, I’m trying to build my own cloud data warehouse. 

But what they’re going to the IT team is I need better insights on my top-of-funnel data or how it’s moving through the lead flow and what’s happening when it goes to sales, of which people like us and people like you as part of this program can actually help surface that data in working with IT. When you talk to customers specifically or when you have that opportunity, is it still IT that is driving this? Or are you seeing more lines of business, more functional people coming in and they’re sharing what they’re trying to do and what data they need access to?

Jobin George:

It is not IT always. Now we have different functions, different sub-orgs within an organization trying to tap into unified data location. Unlike the old times, you are not going to keep a separate copy of the entire data set for different, different functions within an organization. We work with different types of partners who would cater towards different, different siloed customers, if it is a data scientist within an organization trying to do something, if it’s an IT team trying to do something, if it’s a marketing team trying to do something. They will have their own specific needs, which would require a certain set of partner solutions to work with the cloud native services which we offer.

It could be as simple as a BI, business intelligence, tool. It could be a machine learning tool. It could be an integration tool like SnapLogic. It could be a simple connector like Tableau or CData. Different users within our organization will have different needs to work with different types of partners, along with when they try to build their solution on Google Cloud. We work with them and the demands from different sub-organizations within a company.

Dayle Hall:

Right. We have more conversations now that include lines of business, which is one of the reasons why you mentioned this earlier, the whole low-code/no-code is really important to make sure that different functions get access to that data. I’ve heard this for years. The CMO is going to be the biggest buyer of technology in the next five years.

I feel like I’ve been told that for the last 10 years. And whilst it’s true, the best organizations, the best enterprises and some people that have actually been on this podcast today, they talk about a good process between functions, between IT. And I think that’s really when the benefit will come. Because again, I wouldn’t anticipate a sales ops team really knowing how to work with Cloud Ready – BigQuery program. But knowing that that’s out there and then when they come to us or they come to you, that this is available to them, that again enables them to get better value from their investment, which is why we exist essentially.

Jobin George:

Exactly. A company cannot succeed if only one function within that company is successful. It has to be across the board. Otherwise it’s just going to have a natural close-up.

Dayle Hall:

For sure. One of the things that I like to do on podcasting is I listen to other people’s podcasts. It’s always important that I hear something and I think I could do that. Or if I’m listening to a podcast when I’m driving, then I can think, oh, I can do that today or I can implement something like that.

Let’s talk about something specifically around a customer case or something that you’ve seen with a customer be successful. And think of this in the context of if someone’s on the start of this journey and they’re looking at which partners to work with, what technology to implement, and they get to their office, if they’re still commuting in, or maybe they’re doing it on the Peloton. I’ve heard that happens a lot too. I’m a Peloton user. I’m not ashamed to admit it. But outside of that, what are the things that customers should be thinking about? It doesn’t have to be a customer with SnapLogic but some examples of customer success that you’ve seen.

Jobin George:

One of the iconic customers, possibly like a BigQuery customer success story, which I found very interesting, was Home Depot, which I’m a big fan of. I feel like I’m a five-year old kid in a candy store when I’m at Home Depot. It imbibes its associates with Google Cloud BigQuery by providing timely data to keep 50,000-plus items stocked over 2,000 locations to ensure their website is available. They provide relevant information throughout the call centers, and everything is online. Next time you log into or after hearing that story at Google Next ones, when I log into their website, or when I go into the store, I’m thinking, this is how it is tracking, this is how it functions in the backend.

Home Depot, when they started looking at other solutions, they had an on-premises data warehouse at the time, which was under a lot of stress since the demand was really growing, and then it was not able to keep up with the expectations and requests they’ve had from the internal use cases. After careful consideration, Home Depot chose Google Cloud BigQuery as its cloud enterprise data warehouse. And their legacy data warehouse had nearly 450 terabytes at that time. But the BigQuery data warehouse had over 15 terabytes of data after a few years. And then their specific workload which got migrated into BigQuery, like sales analytics, store performance, supply chain use case, etc., so a huge reduction of overall execution time ranging from 83 to 99 percentage, which means hours or days of execution time reduced to minutes and seconds. 

This particular project, what I’m talking about, is one among 600 projects Home Depot is using Google Cloud for, delivering meaningful business results every day. And whenever I visit that store or homedepot.com, I order something, which I did two days ago actually. I’m still expecting the shipment. I’m thinking, you know what, this is Google Cloud technology in the play. I would say after hearing the story, it was really interesting to me, which is great. They’re actually a marquee customer for us.

Dayle Hall:

Yeah. Actually, Home Depot, I think, well, yes, I’m very excited when I go in the store. It’s a little bit like going into Target. I always see myself reaching for many things I don’t particularly need. But I’ve worked at other companies like Home Depot, from their IT and their technology perspective, they’ve always been ahead of the curve. They’ve always been very innovative. They’ve always been looking to take on new technologies. That’s a great story.

As we wrap up here, you’ve been around obviously cloud data warehousing for a number of years, and we just talked about one of those successes. But I’m sure there’s also been a few failures. You probably have a couple of key thoughts or suggestions out there. And again, they’re thinking about how do I engage? What technologies are at the right time? How do I build this if I’m not currently doing it today or potentially I want to add different functionalities around a data lake, lake house, whatever? What are the tips that you would give them to say, look, these are two or three things that I see people make mistakes with? Or really think about this first, don’t just charge off and pick a vendor. What are the couple of pieces of advice you would give to those people as they start this journey?

Jobin George:

I can give from my experience and from my perspective how I think, if I’m making a decision or when I’m making a move. Make sure you set your objectives right and goals defined even before you start. It’s very easy to get lost in the marketing hype.

Dayle Hall:

I think as a marketer, I know exactly what you’re talking about.

Jobin George:

Exactly. Make sure you know what use cases you are targeting. Because when you are not really decided on what use cases, you probably would be making a poor judgment on things you are thinking and then later you figure out, oh, you know what, this is not something which I want. I wanted more. And then, that’s not possible here. Based on your resource availability, decide what architecture you’re looking at. Is it a completely managed version of what data lake or data warehouse you need? When you’re starting off, start small. Do a POC, if possible, onboard one use case at a time so that it’s easy for you to scale and if things fly. 

I’ve heard success stories in my past life as well, figuring out POC at the right time at the right moment within seven days, cutting down your sales cycle, getting things moving in a very short period of time. I’m happy to say that, that story which I’ve heard was with SnapLogic. And then within seven days, you are getting POC sorted out and moving into the next phase. When you’re migrating from an on-premises, compatibility is always going to be something which you will have to keep in mind. 

And always make sure you have a clear migration path. After you start out, you’ll figure out, oops, I should have thought about this earlier. There were so many people who promised who couldn’t deliver, and then you would be in a tricky situation. Make sure that different data sources’ compatibility with that is already taken care of and then total cost of ownership. When you talk about compatibility and things, you don’t have to worry about that a lot because there are awesome partners when you’re choosing your cloud provider versus when you’re choosing your technology around it. Partners like SnapLogic can help you through that always. Google Cloud partners with the right technology to help our native customers.

As you move forward, slowly you will understand how much friction you will find and then you will find the way to get that. And then you don’t have to be ashamed of getting help, getting the right partner to help you through this. Like you said, until analysts, people left the office, you wouldn’t have known automation is pinging the guy next to you and getting something done. You have to have the right thing in place. The right partner can make your journey really, really easy.

Dayle Hall:

It sounds elementary. But it’s actually really important, what you just said, which is to know your use cases, start with the POC, get the use case up and running, and then expand. And I think one of the things that we hear a lot is, and I’m sure you’ve read these too, around failure of digital transformation initiatives, but that is mainly because they’re not tied to specific use cases. They come in and say we’re going to just digitally transform and we’re going to use the cloud. But start with those use cases, get that up and use a POC. 

And like you said, there are partners to help make this more successful. You don’t have to worry as much before around compatibility of data sources because people are here to help with that kind of thing. But there’s a really vital point, which is if you’ve got use cases and you know what you’re trying to achieve, then partners like you, like SnapLogic, can help get those up and running actually much quicker than I think a lot of people realize.

Jobin George:

Totally. Yes, you’re right. You’re spot on there. I’ve seen success stories out of failures when that came to us, re-digitizing their evolution, what they were almost about to give up. But I’ve seen success stories like that too.

Dayle Hall:

This has been an invaluable session for me. Hopefully, for people listening to the podcast, I can spend another hour. I could also spend another hour asking you how to get SnapLogic higher on Google Search results, but that will not be something we’ll discuss today. As we leave our session today, if people are interested to know more about your Cloud Ready – BigQuery partner program or other programs, where should they go? What would you advise them to do next?

Jobin George:

We have a lot of information regarding Google Cloud Ready – BigQuery initiative on our webpage. The easy way to do that is to just search for Google Cloud Ready – BigQuery in Google Search. And then, boom, you will get the first or second link.

Dayle Hall:

I hope so.

Jobin George:

Of course, it does. I tested it a few days ago. If you have any questions or if you are a partner hearing us and want to be like SnapLogic, who is part of this, you can always email us at [email protected], and all the information is available in our documentation and our webpage. It’s easy to find.

Dayle Hall:

Excellent. Jobin, thank you very much for your time. It was a great conversation. Potentially, let’s do another one sometime soon. But thank you very much for your time.

Jobin George:

Definitely. Thank you, Dayle. Thank you again for SnapLogic and, Dayle, for this opportunity. Thank you all for listening. Really appreciate your time.

Dayle Hall:

Great. To everyone else that’s listening, thank you for listening to our podcast today, Automating the Enterprise. Look out for the next one, and we’ll see you then.