More data? More growth!

Speaker A: What if your MMP data was just a start? Hello and welcome to Growth Masterminds, the podcast for growth marketers. My name is John Katir. What if you could put app store data, all of it. I'm talking about ratings, reviews, sales, click through rates, conversion rates and organic, social and even more granular ad network metrics and much, much more from many different sources all together right beside all of your MMP data. On this episode of Growth Masterminds, we're talking about Extract. It's a new next gen ELT and reverse ETL platform, uh, that puts all your data where you need it. So much more data, in fact, than you might have right now. It's relatively new from Singular and we're going to chat about how marketers are using it to chat. Our guest today is Mayan Shore, the product manager for Extract. Super pumped to have you on the show today, Mayan. Let's start off right here.

Speaker B: What is Extract is our new generation ELT and reverse CTL platform. Basically, it's all about moving data where you need it. This is what it's all about. On the ELT side, you can pull extract raw data from really wide range of sources, anything from ad networks and app stores and tools like Salesforce or HubSpot, and really anything you can think about that have an API and load it straight into your own database for analysis. Just the raw data as it is.

Speaker A: So this is Growth Masterminds and we're usually talking to mobile marketers and many mobile marketers using mmp, probably Singular, hopefully Singular, uh, if they're using an mmp, if they're using Singular already, why do they need an ELT tool?

Speaker B: So the idea in Extract is to make it super simple to get data in and out of your system so you can really use it to bring the value. And actually what you just asked me, this is a question I've been, uh, hearing a lot from customers telling me, you already give me so much data in the mmp. Why would I need more? I mean, why would I need more data? There is so much data and the truth is that the MMP data is just a foundation, but the ELT fills in all the missing context. And I think this is the key here to fill the missing context. The MLP already does a great job showing paid traditional campaign performance, but there is a lot more context that marketers want and need to really and need to get in order to really understand what's driving results. And with Extract, they can easily pull maybe more granular ad network data. Uh, that's usually available only via the API, as well as data from other places and sources like, I mean outside of paid marketing, like app stores, organic, social or on channel panels. And when they bring all of this together, you can either blend this with your MMP data to get a richer and more granular view, or you can use it to analyze the teddy as its own anyway to give them. Either way they choose, it gives them way more flexibility and depth in how they measure and optimize their activity.

Speaker A: Cool. So some people who are new to mobile marketing and maybe m just got an MMP for the first time going like, whoa, I have so much data already. What are you talking about? Just the foundation. There's more data, I need more. So we're going to get into all those use cases, what data, where it's coming from, what it looks like, what it does, what it helps, all that stuff. Let's start with what extra data you can get. Maybe let's start with app stores.

Speaker B: Yeah, I must say that the, the use cases are getting really creative and we really learn a lot from the customers and the users who currently use extract on a daily basis. Uh, and they're very. I can maybe narrow it down to a few more. A few use cases which are the strongest that I see. And the first one is around that data types that they actually have access to. Even if you're a really small company who uh, only started as an mmp, so you have access to it, but you don't really realize how much performance data is available or that, or that you're putting it in a very manual and very limited way. And two big examples here, as you said, are uh, the App Store data and the organic social data, which is also a big one with App Store data. Uh, yeah, sorry, no worries.

Speaker A: Just on App Store data. What are you talking about specifically?

Speaker B: Yeah, so with appstradata you can track downloads, deletions, rating reviews, maybe subscriptions, crashes and even acquisition sources. All of this data is available in the App Store. Some customers use the revenue data as the source of truth in order to understand their real income. And so they need to be able to access this data programmatically and all the engagement reviews, acquisition source, this is really, that's really goal for understanding the real revenue and your stability, user sentiment and to understand the impact of any store changes that you do on user acquisition.

Speaker A: Super interesting because of course you won't see that most of that in terms of your MMP data. Right. Like reviews, ratings, that sort of thing.

Speaker B: Right.

Speaker A: And so that's super useful if you see a downturn in installs but you're sending the same amount of traffic to the App Store to Google Play. You know, you can look here and say oh that's what's happening or something is going on there. We're winning less of those conversions when they get there. Very cool. Any other use cases from App Store?

Speaker B: Data Social Organic data becomes a thing among our customers where you can get insights uh, on ah, activities on your uh, social Pages like your TikTok page, TikTok account or uh, Facebook page or Instagram account, uh, anything that you do which is not related to a paid campaign. So you get insight on the post, engagement, the followers, comments and from what we see it helps them, the marketers to understand which contents drive the most engagement and then use it. When they optimize the paid campaign they can profile their organic audience because you have all this data on where all my followers are cutting from, uh, and collect feedback from the reactions, the comments and et cetera and it can really give them like information to optimize both messaging and paid campaigns.

Speaker A: That's super cool as well because we know that some companies are super dependent on virality, right? Some companies are super dependent on what's happening on organic social, whether that's TikTok or Instagram or even Meta, uh, Facebook, sorry in some cases or other social networks. Super interesting as well. There's also some value to an ELT tool like extract with ad, uh, networks which is kind of funny if you think about it, right? Because you think your MMP gets everything and you're all good but sometimes like for instance on meta, you can't grab all the data in one fell swoop. You have to grab several different slices of the data because Meta just doesn't allow you to export it. Talk about that a little bit.

Speaker B: So I think right now, I mean if I'm trying to continue this direction, most of the teams are only like scratching the surface. I think that they're utilizing maybe just 10 or 20% from what's really available when you pull it from the API and they're missing the rest. So yeah, they can get a more granular data from Meta, uh, or Facebook. But I think what's most interesting here is to see how customers like use uh, ELT and in a way that it becomes less about more data just for the sake of having data, but also to fill in blind spots and to get more context that makes the decision smarter. And beyond just getting a lot of additional metadata on your creatives and tons of metrics that's available. The more interesting use case is the creativity that we see them use in the way they pull the data. Uh, for example in extract you can really control on how often and in what uh, extract and load mode will you pull the data. So not going in, don't want to be too much technical here but for example we have customers who pull bid and budget data in a higher frequency like almost near time. So they can use it to optimize and react faster to changes in performance, uh, or pulling the campaign changes and create some kind of change log in or I mean they have this view of all the changes that were done in all of their campaign in all of their sources and then they make sure that no important detail is being missed from their eyes. There are no blind spot and then they can really analyze and understand what was wrong. So I think in my opinion this one, I mean the, the usage of like the, of the flexibility that extract offers really gives them like the way to create like additional layers of context to what they already know.

Speaker A: That's almost mind blowing. Hey, to pull that near real time and see what's going on there and just have that really, really tight sense of understanding what's happening with my campaigns, my prices, my win rates on, on, on, on ad auctions, all that sort of stuff and how that flows. I mean, wow, that's very sophisticated.

Speaker B: Yeah, it is.

Speaker A: We've talked about app stores and I'm sure there's multiple other things you can get from the app stores that we didn't mention but that is possible. So we've talked about ad networks a little bit organic, social. Is there any other type of data or other data sources that you're seeing marketers pull with Extract?

Speaker B: We're seeing some architects use. So uh, tools, App store presentation tools. Yeah. Which also give the granularities and data around the keyword and categories ranking and tons of data which related to app store optimization. So this is also something that we see marketers use and actually I mean we do have customers that already trust us that they are starting to use like data which is not related just to marketing but for example HubSpot and stuff that they're using in their business in general, not just marketing related services. But I think these are the main one. The App Store Organic. Uh yeah.

Speaker A: Okay, cool. So Singular has had like an ETL capability for a long time that's still around. How does it work with Extract, an ELT tool? What's the difference?

Speaker B: So Singular ETL is also a Data transfer tool in a way. But in singular ETL we have schemas, we have predefined schemas which already consists like a predefined set of metrics and dimensions. We aggregated bring this, all of this together into one standardized table. And this is the table that the customer is eventually getting in the destination. It is on a database. The extract platform really gives you just the raw data, uh, not normalized, unprocessed, from different sources. So when you use extracts you will have different tables in your database from different sources and you can use it uh, either to uh, blend it together with your MMP data to join it or you can use this in standalone views. For example. In most cases app store data will be standalone views and will not be joined together with the uh, uh, MMP data. But in the changelog use case for example, uh, you can see data that is associated with campaign, for example, when was the last time that my creative was replaced in this campaign? And then they can add it to the schema that they already get from the mmp. So in either way the user, what they see, the value, it quickly becomes part of their production reports if it's is standalone or together alongside MMP data.

Speaker A: So it basically just gives you flexibility using how you want, where you want, when you want and move it where you need to move it.

Speaker B: Exactly.

Speaker A: Okay, cool. Now you've just integrated singular and extract, is that correct?

Speaker B: Very soon, very soon, all um, of the user level data destinations. Basically the user level ETL feature that we have and now we have two, we have the aggregated ETL which is more about network data and we have the user level ETL which is just user level data. And this feature will be supported via Extract. We will have a new source called singular user level ETL in extract and all customers currently using the features and the one to follow, we automatically get access to the new platform which is very exciting and, and they don't have

Speaker A: to sign up or anything like that. Their same account will work, is that correct?

Speaker B: Yeah, it's going to be very smooth and seamless. Everything you can think about from a single sign on and we will create everything for them. They will be able to move between singular and extract naturally and to manage the data flows there. And um, get all the values that comes with using extract.

Speaker A: Cool, cool. M the benefits, who do you see using Extract? Mostly? I mean I'm talking not about people who just come in and use extract from wherever, but I'm talking about singular customers, people who are using an MMP and they come in and they use Extract. Is it typically super high end mobile marketers, super high end publishers who have a full BI team and all that. Is it a mix?

Speaker B: I think it's a mix but you do need to have like a BI system in place. Like you need to have some sort of database although you can also use uh Extract and just pull reports to your Google spreadsheet or something if you don't really have need to have like a database like Snowflake or Bigquery or whatever. But most of the customers have already some BI department or BI engineers uh, uh, which are uh, part of it in order to be able to process all this data after they sync it from the, from extract. But the marketers are the leaders here. I mean usually they know they need to understand the value. They stand and ask for different types of data and together with the engineers we make sure that they have like the most the setup that's right for them.

Speaker A: Cool, cool, cool. Super interesting. I, I would have assumed it was primarily super high end teams with whole BI teams and everything like that. But it is also interesting to just be able to pull it to a Google spreadsheet or something, something like that. So theoretically you could simply pull a certain subset of data from the app store, download it into a Google spreadsheet and you have instant charts for certain people in your organization who just need to know this piece of information. How many installs or what our sales are or revenue, whatever it might be. Correct.

Speaker B: Yeah. And you can automate it. The whole thing is that you decide when and how do you want to put it, what fields are most interesting to you. It doesn't have to be like this scary raw Data reports with 2000, with two hundreds of fields. You can take just what right for you and automate it so you will uh, get this in the frequency that you need it.

Speaker A: Cool. Awesome. Thank you so much. Really appreciate this. I think it's free to try if somebody isn't using it right now.

Speaker B: Yeah, everyone can sign up regardless if you're our uh, customer or not. You can go to Extract and sign in and get 1 million records per month free per month to join a free program and try other sources that you want.

Speaker A: Yeah, cool, cool. Thank you so much.

Speaker B: Thank you John.

Speaker A: Sam.