This episode features an interview with Jon Hyman, Co-Founder and CTO of Braze. In this interview, Jon talks about migrating data to the cloud, why data-driven companies outperform competitors, improving email marketing with data, and much more.
This episode features an interview with Jon Hyman, Co-Founder, and CTO of Braze. In this interview, Jon talks about migrating data to the cloud, why data-driven companies outperform competitors, improving email marketing with data, and much more.
--------
How you approach data will define what’s possible for your organization. Data engineers, data scientists, application developers, and a host of other data professionals who depend on the Snowflake Data Cloud continue to thrive thanks to a decade of technology breakthroughs. But that journey is only the beginning.
Attend Snowflake Summit 2023 in Las Vegas June 26-29 to learn how to access, build, and monetize data, tools, models, and applications in ways that were previously unimaginable. Enable seamless alignment and collaboration across these crucial functions in the Data Cloud to transform nearly every aspect of your organization.
Learn more and register at www.snowflake.com/summit
Steve: [00:00:00] Hi, John. , welcome to the show. Glad to have you here is good to talk to you again. I know we've had a couple of good conversations in the past.
Jon: [00:00:08] Yeah. Thanks Steve. Appreciate it. You have me here.
So, you know, we talk about the data cloud and we define that as the great migration of data and data analytics to the cloud. How is that phenomenon transforming your industry.
The data cloud is something that I think has just an enormous potential, , to really change how industries operate because it allows sharing data inside an organization. And importantly. Having a company use the same data set across their different business units. And what I think it allows businesses to do is ultimately become more data-driven.
for example, if you look at a lot of research, you see that data driven companies tend to perform very, very well. We actually, I was looking at this study that McKinsey did last summer and they were looking at characteristics of top performing [00:01:00] companies against their peers. And what they saw is that across every dimension.
That the top performing companies actually used data more so like the best companies. We use multiple sources of data to assess any of their unmet needs, and they would do that weekly. They would dedicate time to learn about new technologies. They would share learnings that they found from testing, and they would do that on average about monthly.
I mean, essentially it means that if you want to innovate, then you need to iterate and putting your data into the cloud and letting your business see that data. Always accelerating that.
So what about your industry in particular? Consumer marketing and advertising?
. So we help consumer brands power relevant and memorable experiences with their consumers, and we really want to help provide context and underpin that with every interaction that a consumer is getting from a brand to ultimately foster a human connection.
And so understanding the impact of a [00:02:00] brand's engagement strategy is essentially just critical in informing like how they iterate and optimize their messaging to their users. And so the role that we play is that we are essentially tracking a wide array of different types of event data that's going on.
That actions that consumers are taking, how they're responding and interacting with those engagement campaigns, making that available for analysis for retargeting or any other use cases, and we're helping our brands essentially. Learn from that data and essentially decrease the amount of time it takes to get value from it.
Cause if you think about data overall, there's really kind of two challenges when it comes to getting value out of it. One is that data decays over time. So as soon as it's collected essentially is not as valuable. , if you kind of sit on it versus acting on it immediately, I mean really speed. It's just very crucial to generating value.
You even, you think about personalization. If you are a retail store and you're trying to send customers emails about [00:03:00] items that they shopped and were looking at online, it's just simply better in most cases, to send them messaging around items that they've viewed more recently than items maybe they viewed three or four months ago.
And you know, then even if you had that data and you could like operate on it. Another challenge is then just being able to get insight from it. , because you know, you may have the data somewhere, but in order to see report or put that into some kind of the machine learning algorithm that may need to require you to build a data pipeline, you might need veterinary resources.
You really need to, uh, you know, kind of like go from data collection to. Actioning off of the insights that could take a while to actually do so. We've, part of the role that we play in that is that we're helping brands make their business intelligence a little more turnkey with how we are exposing their data.
We see this big trend that a lot of brands are trying to really kind of take their data that they have that is across their different [00:04:00] SAS providers and they want to collect it themselves. I mean, we'd see that a lot of companies don't want their data to be siloed with any individual provider. And so what we've done is we built a product a few years ago when we kind of saw this future coming that is called currents.
And what currents is, is a continuous real time data export of all of this interaction data that we collect. So if we send you a personalized email. And you open that. We now know that you opened that email and were able to then share that with the brand and stream it to them in real time. We serve you in in that message and you see it and you click on something where you dismiss it.
If you look at a push notification, click on that. If we send you an email and it bounces, we get a lot of this type of engagement data on how you're interacting with the messaging, and we can then want to give that back out to our clients so they can then use it across their businesses. And we've seen some really exciting capabilities, , and things that our clients have done that I believe in made possible by this kind of [00:05:00] phenomenon.
Yeah. Hey, give me an example of that. That sounds intriguing.
So, yeah, there's, there's a couple of things here. , one is that we help our customers, uh, as mentioned, we give them this data, , in real time with, with currents, and we've seen that data that export it from grays can like power. A lot of great custom reporting tools. We've had brands that are like really optimizing their life cycle engagement strategy all off of this data.
In fact. , I was looking at this one really great example just this morning where one of our customers are a very large, very well known public lifestyle company. They created their own email dashboard that was powered by this data that we send them so that they could look at their account signups, their unsubscribes by a number of different dimensions.
So, for example, and because we're sending emails to their users and we're streaming that information of how they're opening those emails, clicking links on subscribing. In signing up for their products and whatnot. Um, back out to them, they're able to build [00:06:00] dashboards of things where they could break down signups and unsubscribes by what market they were in.
So that could be like, you know, the U S Canada, Europe, Australia, and whatnot. They were breaking it down by the product that they were advertising for so they could really see it. Does the sign up rate increase or the unsubscribe rate decrease. Based on the content or the product that's being served, or they were also, we're just looking at different audiences, different age groups, different, interests, , or just kind of different descriptions they had assigned to their user audiences.
And so that was a really kind of exciting output that they put together. But typically we see things like brands using our data source to tie other data sources together. So looking at like, if they wanted to do a retention analysis, they could take raises information on the engagement funnels that they have from the messaging that we're sending, and then they could tie that down to some of, tie that back with some of the other data that they have.
And, um, you know, really kind of come up with a good insight there.
. [00:07:00] Now when we spoke before, I think it, you said it's a big kind of grand and inspiring, at least to me. You said our focus is on helping to humanize the communications between brands and consumers on a global scale. , I want to know about two things. Is the humanizing, is that about personalizing and then what do you mean by global scale?
There.
Yeah. Sure. So when we think about about being human, essentially, if you look at the traits and attributes that consumers want out of brands that they're loyal to, we actually were, we ran a study with Forester at the last two years to kind of look at this, and we see that, that consumers are looking for brands that are thoughtful, that are friendly, are personal or helpful.
Really, ultimately, these are, these are human attributes. You know, we, we want to be empathetic to our clients. Um, at the same time, , we think that brands should, , , really the principals as well. , and so when we think about like humanizing, what [00:08:00] we want is we want to give.
Our customers are the marketing teams at the brands that use us, like the power to take their creative ideas and then build them into like well orchestrated and personalized conversations that are delivered to people at the right time or through the relevant channels that really ultimately leads to a strong relationship with their customers.
Because it is, you know, it is the brand being thoughtful or helpful or friendly or relevant and timely. And on a global scale is just essentially a byproduct of the fact that we're living in this digital age now where we really have been able to kind of democratize, you know, distribution. Well, the fact that a brand can be everywhere because of the mobile phone.
And because of things like mobile apps or just just the web here where, , you know, consumer brands could really be interacting with people. All over the world and they could be operating with, , tens or hundreds of millions of users. We work with some very large, [00:09:00] , customers of bridge, very big brands that have hundreds of millions of monthly active users for their products and services here.
And that's something that you know, is, is really kind of new. I mean, you know, you go back 20 years ago to have. Yeah. Hundreds of millions of people using your product. You likely had to be like a Coca Cola or you know, a sound papers or something like that. But, but now there's so many businesses that just can have that reach just by the fact though, that it's very easy to acquire their product.
It's on a phone or it's on the web, and you know, doing it at that scale is very, very challenging.
And I'm going to go back to something you mentioned at the top about data sharing being one of the, core things here. I wonder if you could kind of get into the nuts and bolts of, that's a little bit, I mean, in the past, sharing even within an organization was very difficult, but now it seems like with the cloud.
, organizations are able to share better internally, but also between them and their partners. So what's the how, [00:10:00] and I know you guys are our snowflake customers and use the data cloud platform. If you can, you could just go into a little detail on what, how does that enable the sharing and why is that valuable.
So snowflake data sharing really helps make our partnership with our customers a little more turnkey when it comes to us providing them this engagement data. So I mentioned that we do have this product called currents that is continuously exporting data back to our clients and typically, um, before snowflake and before this partnership.
, what we would see happen a lot is that our customers would export our data to one of the clouds that they're using. Maybe it is on Amazon or Google or Microsoft Azure. And then they would take the raw data there and then they would extract, transform, and load it into their own data warehouse. And this is one of those things that really decreases that time to value.
You know? Because even if for example, in praise is able to export what's [00:11:00] happening and what your consumers are doing to immediately, like say you open up an email and then we share that information with a brand instantly. Okay. Their process on their side involves like a nightly and a batch upload that takes that data and downloads it and extract it in.
Then at best, the marketer who's responding with that data, it's still only using data that's, you know, maybe a day or two old. So it's not just about how fast can braise deliver data to our clients, but how fast can our clients as business process take advantage of that? And we want it to make this a lot better.
And snowflake is something where we kind of looked across our customers and we saw that a lot of businesses are starting to work with snowflake. It's a very important technology. We also have seen that ourselves, as you mentioned, where snowflake customers, and you've been shifting, um, a tremendous amount of our stack, and kind of our data storage over to snowflake.
product that you mentioned with data sharing. , this is a really [00:12:00] interesting and exciting prospect for us. So for folks who may not necessarily be familiar with Snowflake's architecture, they all essentially decoupled the storage of data in the data warehouse from the compute and processing of the data and the data warehouse.
And so what that means is that essentially you can have a dataset. That can be accessed by some other computing resource in the snowflake ecosystem. So really what Snowflake's data sharing functionality allows us to do is it allows braise to basically give our brands secure access to their data, like through our snowflake portal.
And they don't have to worry about any of this. Friction or slow down, , by like an ETL process. They don't have to worry about any failures or no kind of maintenance or unnecessary costs that come with kind of maintaining their own data pipeline. So basically what this means is that as soon as braise gets that data and we put it into our [00:13:00] snowflake, for instance, which is essentially seamless and instantaneous, then all of that engagement and campaign data is immediately accessible and queryable as soon as it arrives to braise by the customer.
So we don't have to copy any data to our clients. We don't have to move any data, they don't have to own any pipelines. I'm going essentially, as soon as braise ingest that data. And it goes into our snowflake account, then we can share that with our clients and they can instantly access this, and we've seen this be be a huge boon for our customers when we have one client of ours that comes to mind.
Very large media company that owns thousands of different television stations and other digital publications. And they use snowflake data sharing internally for them to share large data sets across the organization with different business units. So if they have one TV group, they can actually kind of share similar sets with them and embrace as a part of that.
And so they essentially are [00:14:00] sharing, well, braise is sharing data from our snowflake instance with their core data team. And then that's able to be used across their whole company for everyone to essentially have instantaneous and identical access to the same dataset, which I think, you know, we'll just say like having different business units use the same source of truth for data is, , really kind of a game changer in the sense that you, you know, you, you don't have any kind of problems that people work with the stale data and it's songs and it gets rid of a lot of unnecessary processes and, um, it's a great way to improve efficiency.
Yeah, it seems like that's one of the Holy grails of the data world. Being able to make sure that you have one set of data and it's, everybody is sharing the same information. Everybody has the same knowledge, so that's really cool. , I know you share, you have masses of data that you collect on behalf of each of your individual clients and you share it with them.
but you also , , you take some data that you share with all of them, that kind of, it cuts [00:15:00] across the industry that they can learn from. Talk about that a little bit.
so. Essentially, when we look at what, , you know, is, is kind of really exciting about braise from, from an industry standpoint, is that we work with a tremendous number of different industries. Whether that's media is, I was just talking about with retail, costumer, with gaming, with lifestyle onto man.
And really what is on a lot of folks is minds when they're looking at their engagement programs as they're wondering, like how they're stacking up against their peers or their competitors. And what is, you know, has, has been a really great service that we've provided to our customers, is able to give them those kinds of industry benchmarks around, you know, how they're doing, just to make sure there's, there's no reason to kind of look at your engagement blind and see if your open rate, you know, is, is good or bad if you don't have anything to compare it against.
It's very hard for you to iterate and optimize and kind of know where you are. And so we've always provided our own customers. There are benchmarks from what we see. So if we're [00:16:00] working with a retail company, we can say, Hey, here's your email open rate. Here's what we're seeing across the media space, or the retail space, sorry.
And we can tell them, you know, maybe you're ahead of the curve or here's some areas in which you can improve on because we can compare it to the two benchmark. And we wanted to make that something that any anyone could take advantage of, not just brace customers. So what we did is we took, kind of an anonymous aggregated dataset across the mobile web and email engagement data from more than 600 of our brands that are in the EU and the U S and we looked at things like monthly and quarterly and yearly engagement rates by industry.
And we essentially took millions of campaigns across more than a dozen industries to look at. Again, like things like email open rates. Push notification clicks, , email, click rates and all of those kinds of ratios. And then we created their own dataset called the braise benchmarks. And this is a dataset that is [00:17:00] freely accessible to anyone who has a snowflake account where they can just pull this data directly from us.
But what I think is really novel about this is that this data is, is. Essentially served directly from braises instance, like we're not copying the data and dumping it over to snowflake. Our data exists in snowflake already and it's updating continuously, so we send tens of billions of messages every single month.
And as we were sending those messages, this benchmark data is essentially being continuously updated. And so you can kind of get access to this benchmark data just right directly in your snowflake account. And what that means then is you can input it in your own BI tools. And even if you're not a brace customer, you can see how you compare in the industry.
You know, if you think about how you would get benchmark data today. You know, you can probably basically just Google for, you know, email, open rates by industry. You know, at best you [00:18:00] find, you know, maybe a PDF from last quarter or last year is open rates and it's very hard to, to really kind of respond or interact with that yourself.
But this is extremely interactive on, you know, you have access to the raw data and it's constantly updated. I mean, I think it's really one of the kind of game changers for folks who are in marketing to be able to see how they stack up.
that's a great service you're doing for your industry, and I'm sure they appreciate it. Hey, you know, um, it's funny thing is it's a personal story, but . I'm recently switching from one email system to another, and it really has made me think a lot about email marketing, email advertising, because I kind of have to see how Hotmail did it and now how Gmail does it , and work it out.
And you know. You have an incredibly challenging job and your clients have a challenging job to take. Something that for many people, Oh, we use the internet and use their computers is. It could be a [00:19:00] nuisance, you know, we got spam, they're unwelcome, , you know, approaches all the time. So you have to take what might be normally seen as an unwelcome approach or a news nuisance and turn it into.
A welcome interaction. It's almost magical in my view. I mean, if you can get to the essence, and maybe you, you know, we've already talked about some of the big concepts here, but what's the essence of turning something from a nuisance into a welcoming experience.
Sure. So one thing I'll just kind of preface this on with is, is when done well and personalization is something that. And there's a huge benefit to them. Consumers. I mean, I think that if you just look across, you know, our digital lives, we really have very high expectations around the amounts of personalization that we get from anything.
Whether we're online shopping or we are watching Netflix or something. We're consuming any kind of media. , you know, to [00:20:00] our listening to music. I think that now you see that brands that do this really well create this self reinforcing loop. That affects all industries, and it really raises the bar for everyone to deliver a great personalized experience.
So there's point where, you know, it's not a new sense of, it's personal. , what's, what's bothersome is when you get a message that is completely not personalized for you, you know, when you get a message from your credit card company reminding you to download the mobile app to pay your bills.
, when in fact you've been using the mobile app to pay your statement off every month for the last year. Right. And you see, those kinds of bad experiences happen from brands that are siloed. If they have an email team that is separate from their web team that's separate from there, . From their mobile app team.
And they're all using different data sets and using different strategies. And so the things I would say is that if you want to turn this into something successful, it was a couple of different things. , but they, they really, the pattern relies on. having great data and focusing on that. , you know, thinking about how your teams operate and your [00:21:00] technology.
So, I mean, I would kind of say the takeaway points for me would be just , really kind of like focus on your customers and integrate multiple different data sources. You know, really try to understand what are your customers essentially doing. You want to think about like, what happened.
I'm like, what do they use my product? What do they do with my product? Well, that's kind of a backward looking question with your data, but essentially you want to understand why did a consumer take a certain action so that you can ultimately kind of, you mean a mature your own kind of thoughts on this to be able to predict what your consumers want and how you can, you know, kind of help make something happen for your customers.
And so, you know, being focused on, on their customer and integrating multiple data sources, there is a huge benefit. I'll also say then teams are very important, is investing in talent. It's super critical. And we have now, you know, a lot of different folks that play a part of the engagement strategy. It's not just the marketing team.
I mean, consumers don't really care if it's the marketing team or the email team or [00:22:00] the product team that is sending them an email. They don't even know. Ultimately, they're just having an experience. , and so you need kind of best in class folks and data science engineering, product marketing. Really to kind of be productive here.
And also then like empower them to be independent and to make good technology decisions. So that leads me to my last bit, which is that, you know, you need to now build on a modern architecture and essentially have something that is able to interact with your clients and respond to what they're doing in real time.
And that's where having a solution like braise is extremely important. It's where something like snowflake is very important because it allows you to be very nimble, to have all your data, to share it across your whole organization. and really, you know, cause I mentioned like have a very fast on to insight.
Steve: [00:22:47] Yeah. No, I think those are really good
points. You know, John, we've had a really good discussion of, you know, how your technology works, how your industry works, how to make email and all kinds of digital [00:23:00] marketing better. But the context for our conversation today is really in the middle of this phenomenal.
Coronavirus pandemic that is really causing chaos in society and kind of be in businesses. So I wanted to check in with you. What is the immediate impact of the pandemic on your clients and on your business?
Sure., , . So as you know, Steve, I mean, really all industries have been affected by the coronavirus pandemic and, and some of them have been affected more than others. And what we do at praise is that we provide our customers with the ability to communicate with their consumers.
And this is more important now than ever. , we've been fortunate, . , I mean, have been able to prepare for these types of scenarios. We're very diversified and across our clients, we work with over 800 customers from over 50 countries across multiple industries.
, you know, as you can imagine though, um, some of our customers have been greatly affected by the coronavirus. Uh, we've had seen some of them were, , , there's been increased [00:24:00] usage. Some things like the work from home, that's been driving more, kind of traffic to them. , but it's really kind of been a mix.
Jon: [00:24:07] So, you know, during this crisis, obviously there's, there's lots of pain, there's lots of concern, and , brands have to be very careful about how they present themselves. But they also have this great opportunity if they take it to really reach out to people and to help them get through it.
And I, I think you've had a couple of examples of, of clients of yours that have really used their, their marketing tools, their advertising tools to really engage with people in new and very positive ways. So if you could talk about one of those, that'd be great.
Sure. One thing that's become a little more clear is, is how important it is for brands to communicate directly with their customers. This was already a trend that we had been seeing, but it's really accelerated now, with kind of the current situation. And, you know, there were some businesses [00:25:00] that I see that, that don't do this super well.
I mean, there's a fast food, . Shane that has been sending me the same push notification every day, the exact same time for me to come on by and purchase an item from them. And, you know, they're not really kind of innovating or testing or doing anything different with their messaging. Um, but then we have a customer of ours who is also in the fast food industry and they've been doing something which you think is, has really shown their human side.
And shown that they understand kind of what's going on and they want to be helpful and friendly. And one of the things that they've been doing is. Got it. Been playing on top of the fact that you have so many students who are at home doing remote learning, and there's a lot of focus on education and teaching, and they've been creating campaigns that are all around quizzes and trivia for different subjects.
So they'll send you a push notification one day that has a math problem on it, and the push notification says, are you good at math? If so, [00:26:00] click here and solve a puzzle and get a free item. it's something like that, that's not the exact quote, but you click on the push notification, then you get served.
And in that message that that braise is serving to you, that has, you know, really fun branded kind of look. On the math question and then it has a problem for you to solve, and if you solve it, then there's a box to put in the coupon code. And if you get it right and put it in there, then you get a free item.
And they've been doing this with a couple of different subjects. There's been math, there's been geography, computer science. I think that's just something where they're really trying to connect with their consumers because they know that folks are at home, they're having to teach their kids. , you know, maybe they can teach their kids some math and also get a free item, , from, you know, a really great fast food restaurant at the same time.
And so I think that's been a great example that I've seen come out of this, that highlights the creativity and the ingenuity of people even in this trying time.
Steve: [00:26:54] Yeah, that is a great example. You know, just the idea that there, the kids are kind of trapped at home. [00:27:00] Maybe school isn't as much fun there and they're making education fun to kind of bridging the gap. That's, that's really cool.
, yeah, and I think that at a time of great stress, that a time of crisis. You know, leaders of government, of industry, business leaders, brands , the ones that step up and show kind of a sensitivity to the situation and, uh, authenticity. That really does create a lasting impression. I mean, I think it's, in a way it's opportunistic, but it's also just a smart way of dealing.
I think we do see some of the brands doing that.
you know, the last question I would asked you is, you know, this crisis will end. We don't, nobody knows when and how, but we know it'll end. So I want you to look ahead for a few years beyond this, the big picture, how do you see data and [00:28:00] these new uses of data changing your industry and even changing business in general?
Jon: [00:28:06] Earlier in our conversation, I mentioned that we see a trend that businesses want to have access to their raw data and that they don't want it to be siloed in their SAS providers. And so one of the things that I think we're going to see is that trend will continue and you will have an increase in data op interoperability.
And I think that it's not just , business demand that is going to drive that, but also the consumer demand that's driven by regulations like GDPR. for example, under GDPR for data privacy, the essentially have rule and save that data. Controllers have to no guarantee that any personal data they have is transmitted in like a structured.
Commonly used format and can be interoperable. Um, and I think that, you know, you're going to see more of that kind of definitions on, on data take place. I think Snowflake's gonna be an interesting spot in this because [00:29:00] with data sharing, you know, essentially allow brands to be able to. Unlock the data that they have in their own data warehouses and share that with clients and share it in a format that is easily accessible and digestible.
And I think you'll start to see a larger ecosystem be built out on top of data warehouses and products like snowflake, where you're going to kind of be able to, to extend and expand upon SAS businesses by using their data for other things. Other additional things. So, um, I really see , that trend of kind of data, openness, spawning new businesses and new use cases.
, Um, and also then, , being ultimately, again, at boon for businesses, which ultimately ends up being a boon for consumers.
. It's like a big data mashup in the sky. So that's a very positive outlook. I like, I like to end on a positive note cause we did explore the, the Corona virus pandemic Abed and that's clearly a very serious and somber matter. But, , John, I want to thank you for your time today.
[00:30:00] I think you've really brought a lot of insight to this that maybe people hadn't thought about before or weren't aware of before. And I think you're really in the middle of this, you know, data marketing, data communications, which is so essential, you know, especially today. So I want to thank you very much for your
Yeah. Thank you so much for having me, Steve. Talk soon.
So John is, was there, was there an initial pilot thing that you'd launched just to kind of detest it and, and see what you can get out of it?
We did start with a pilot and the team that we started it with was our business intelligence team. In business intelligence at braise is run under our growth organization. And what growth organization's mission is, is that they essentially work to help braise and our customers. . We essentially be able to scale a little bit better.
And one of the things that we were looking at supporting was just the business intelligence that we use for our executive business reviews. So I was mentioning before about benchmarks and being [00:31:00] able to provide that kind of data to our customers. There was a whole amount of business analysts and data analysts kind of work that goes into producing that.
And so we thought about could we. You snowflake just to kind of power those types of use cases ourselves. And so the initial customer wasn't our engineering organization. It wasn't even at any of our product side. It was kind of on that business side of the house. We wanted to take data that we already had and then essentially extract it and load it into snowflake and then transform it so that they could.
They could be more efficient themselves. Um, and so we didn't have to really build anything new on engineering, which was very nice because we just allowed the business intelligence team. To essentially take data we already had bring it in and then pilot it, um, before we even started connecting up a real time flow into it there.
So it was something that, you know, we, we started with, we saw a lot of promise on, and then we thought a little bit bigger picture, , not just for the growth organization, but then how we can [00:32:00] use snowflake across our customer, across our business. And then across also, our engineering organization kind of went from there.
Steve: [00:32:08] Okay. Hey, you mentioned efficiency was the goal there , is there any way you can quantify? Is there a metric you can, you can put to that kind of results you got out of that initial pilot.
Jon: [00:32:18] So I don't know the exact metric in terms of where we are now, but we did look at how long it used to take us in order to provide support or run queries and things that I know that things that we used to take hours, we're able to get down to just the minutes. And then we've been able to build dashboards that have truly been repeatable, where now we don't have any.
Central human intervention. Oh, we can essentially just kind of let people self-serve off that data. And so from an efficiency standpoint, we've certainly met the goal for some context on what we were before. I'm using snowflake cause we had this giant Hadoop cluster and um, had a tremendous amount of data on it.
And it really [00:33:00] took probably that 20 to 30 hours a week from my dev ops organization in terms of just supporting it and the queries with sometimes runs slow or they would time out where there would be a system failure. And it may be, it would, it would cause some problems there. , that our BI team is having some challenges with.
And, um, at least now with snowflake in the picture, Hadoop is, is gone. We no longer have any engineering maintenance costs. That LLS team got all that time back and then the queries are returned much faster. And so, um, you know, ultimately on a, essentially a human capital perspective, it's been extremely valuable for us.
Steve: [00:33:37] , and how long did that project run? The pilot? I was just a matter of weeks or months or.
Jon: [00:33:42] So we could do over the course of a few months. , they were, you know, kind of different pilots at different times. So I'm kind of lumping these all together in my head, but, , you know, our, growth team. might've spent two or three months looking at it. in a product organization before we, I've done, we're with snowflake.
We like to do our own thorough testing to [00:34:00] make sure that, you know, we kind of understand the technology and it's going to work well for a use case to be cost effective. And that method for us took about four months. Um, but all in all, , no is relatively fast to get up and running. There's just a lot of validation we want it to do.
Yeah. Because the clouds were able to get into the application immediately that sounds great. A couple of months and you have results. Good.
Yeah.