Note: This is Part 1 of our two-part 7investing interview with Muji. To see Part 2, please click here.

If there’s one thing you can count on with technology, it’s that it can completely change the way we do things.

Music that used to be played on vinyl records is now being listened to through Bluetooth earbuds as digital mp3. Film reels in movie theaters were replaced by VCRs, which then got replaced by Smart TV streaming. Stick-shift cars eventually became fully-autonomous, while libraries evolved into digital assistants in our living room. There’s no industry that’s immune to technological innovation. Even our coffee is now being made and served by robot baristas!

This pace of innovation continues to accelerate further, as advancements in cloud computing and artificial intelligence are unlocking new opportunities for organizations of all sizes. Forward-thinking businesses are increasing sales by learning more about their customers, while those with complex supply chains are becoming more efficient by taking costs out of their operations.

All of this innovation is excellent news for growth-style investors. By understanding the fundamental technologies that are shaping the future, we can identify the most promising companies who are capitalizing upon them.

So what are those important trends that are brewing today? And what are the specific investment opportunities that are arising?

[su_button url=”https://7investing.com/subscribe/?marketing_id=18024″ target=”blank” style=”flat” background=”#96C832″ color=”#000000″ size=”6″ center=”yes” radius=”0″ icon=”https://7investing.com/wp-content/uploads/2021/04/7Investing-3.png” icon_color=”#000000″]Sign up with 7investing today to get access to our 7 top stock market recommendations every month![/su_button]

To answer those questions, we’ve brought in a technical expert. Matthew Eash — better known by his online moniker “Muji” — is one of the world’s most forward-thinking technologists. After spending decades as a data architect, he’s now founded his hhhypergrowth newsletter to focus on the technologies powering today’s most innovative companies.

In this exclusive conversation, Muji speaks with 7investing CEO Simon Erickson about the important trends taking shape in the tech world. Muji discusses why the recent Confluent (Nasdaq: CFLT) IPO is so interesting to him and why several companies built upon open-source platforms often end up butting heads with the Cloud Titans. The two discuss usage-based pricing and the metrics investors should be evaluating in the new era of cloud-native software companies. They describe why Snowflake (Nasdaq: SNOW) is such a unique opportunity, as well as a variety of other topics that range from Kubernetes to edge networks.

Finally, Muji takes a closer look at several important events that have taken place in the cybersecurity industry during the past year, including the SolarWinds data breach and FireEye (Nasdaq: FEYE) spinning off its products division. He also describes what impact the White House’s recent Executive Order will have on Zero Trust technology.

Muji and Simon share several of their favorite investment ideas throughout the conversation. This is a must-listen for any investors who want to capitalize on the biggest trends taking shape in technology!

Publicly-traded companies mentioned in this interview include Alphabet (Nasdaq: GOOGL), Amazon (Nasdaq: AMZN), Cloudera (NYSE: CLDR), Confluent (Nasdaq: CFLT), CrowdStrike (Nasdaq: CRWD), Elastic (NYSE: ESTC), Fastly (Nasdaq: FSLY), FireEye (Nasdaq: FEYE), Fortinet (Nasdaq: FTNT), Microsoft (Nasdaq: MSFT), MongoDB (Nasdaq: MDB), Okta (Nasdaq: OKTA), Palo Alto Networks (NYSE: PANW), Snowflake (Nasdaq: SNOW), Splunk (Nasdaq: SPLK), Twilio (NYSE: TWLO), and ZScaler (Nasdaq: ZS). 7investing’s advisors or its guests may have positions in the companies mentioned.

[su_button url="/subscribe/" style="flat" background="#84c136" color="#ffffff" size="6" center="yes" radius="0" icon="" icon_color="#ffffff" desc="Get full access to our 7 best ideas in the stock market for only $49 a month."]Sign Up Today! [/su_button]

Timestamps

00:00 –  Introduction: How did you come upon the name “Muji”?

03:04 –  Confluent’s Initial Public Offering

13:54 –  Open Sources Platforms Competing with the Cloud Titans

25:43 – Usage-Based Pricing and New Investing Metrics to Watch

(Part 2) – “Nice to Have” Versus “Must Have” Companies in Developer Tooling

(Part 2) – Microservices and the Role of Kubernetes

(Part 2) – Cloud Vendor Ecosystem Lock-in and Switching Costs

(Part 2) – A Closer Look at Snowflake

(Part 2) – A Deeper Dive into Cybersecurity

Transcript

Simon Erickson  0:00

Hello, everyone, and welcome to this episode of our 7investing podcast. I’m 7investing founder and CEO Simon Erickson and there is a digital transformation underway, companies across the globe are moving to the cloud. And they are embracing big data and analytics to get more insights about their operations and their underlying business. And who better to talk about some of the biggest trends taking place in technology, then Matthew Eash, better known to most as Muji, he has been a data architect for several decades. He’s also now the founder of his own site, hhhypergrowth, which you can see at hhhypergrowth.com that’s with three H’s. Matthew/Muji it’s such a pleasure to have you on our 7investing podcast this morning.

Muji  0:44

Hello, it’s been a spell since our last one. I think I talked to you about a year ago.

Simon Erickson  0:50

That’s right, I always like to catch up with you Matt. I always think I always think of you as kind of at the forefront of technology, you’ve got such a thorough understanding of the trends. It’s always really neat to catch up with you at least every year or so.

Muji  1:01

Yeah, exactly. Yeah. I mean, you mentioned I was a data architect, and software developer for decades through the rise of, you know, the Windows platform through the rise of the Internet, and so into web development, but I’ve always been at a database focus. And so it’s something I finally pivoted towards the market, you know, instead of just trusting services, I decided to dive in deeper into these platforms and understand, what was driving them. And I found I could explain things fairly well from my prior history as a data architect explaining complex architectures to management and stakeholders, but also to customers. And so, you know, I got somewhere along the line got really good at explaining complex systems. And that’s kind of what I’ve translated to the blog.

Simon Erickson  1:49

Absolutely. And I think that’s really important for investors to to have a deeper understanding of what’s actually taking place out there. So let’s deep dive into several of those topics. I do want to talk today about an upcoming IPO that’s scheduled for this week, I want to talk a little bit about open source platforms. I want to talk about cybersecurity. But perhaps I might start with a question I think is on everybody’s mind. Before we get into the tech side of this conversation, how did you come up with the nickname Muji?

Muji  2:15

It is a nickname from high school, it was an odd word that got thrown out by a friend, another friend latched on to it. And when I went to college, I went to CU Boulder in Colorado. I met all my friends through this mutual friend that loved that nickname. And so I became Muji at the time, and you know what’s really taken hold when your mother starts using your nickname instead of your actual name. And so I’ve been Muji ever since. I will say My apologies to Nepalese listeners. I’ve learned what Mooji means in Nepalese and you can just call me, Matt, if you’re from Nepal.

Simon Erickson  2:54

Fair enough. We do have some listeners from Nepal. So good to have the disclaimer out there. mooji. let’s let’s let’s kick this off by actually talking about some relevant news of today. Tomorrow, a new company called Confluent (NASDAQ: CFLT) is scheduled to IPO, it’s a data streaming platform that’s built upon an open source platform called Apache Kafka. And it’s one that you’ve been following quite a bit lately. I know you’re excited about this IPO. But what can you tell us about Confluent as a business, and why are you so interested in this one?

Muji  3:23

Yeah, absolutely. So yeah, I’ve been recovering extensively have a premium service on my blog, as well as a free side. And so I’ve been covering it extensively in my in my premium side, covering what Apache Kafka is as an open source platform, and then taking a look at Confluent through the lens of their S-1. And I’ve actually used their platform, I’ve been a Confluent customer in a prior job as a data architect, because there’s a key need, which is moving your data around between the applications and systems that have to interconnect within your software stack. And so it’s a very common need in on the data side of things, which I kind of cover and follow extensively, you know, data and analytics tools. And so, Apache Kafka, as you said, is a data streaming, or pipelining tool.

It’s basically a middleware, all your applications will connect to Kafka. Regardless of where they live, it can help transport data between systems, and whether that is applications having to talk to each other, whether that is remote sensors, collecting data and passing it inward into a database. Kafka can serve as that centralized river of data, and then everything taps into it streams from there. So all the producers tap into produce and are kind of the on ramps of data and then the off ramps are the things collecting and following the individual topics that are being published to Kafka, so very vital need I was really surprised at how well It’s done as an open source package, it is absolutely one of the top open source packages in data. And I think from the Confluent S-1 and from the Apache Kafka site, 80% of Fortune 100 companies use it. 70% of Fortune 500 companies use it internally.

It is vital infrastructure for data movement, especially as your enterprise is larger and larger, you’ve got more and more segments that you have to exchange data with. You’re acquiring new companies through M&A that you need to onboard into your into your data streams. You’ve got a global presence, where you’re in Asia Pacific, you’ve got factories located other places, you’ve got retail stores scattered about the globe. This is perfect for collecting data across the globe, in a centralized manner. And then what Confluent adds on top of the open source, the original authors of Apache Kafka at LinkedIn formed Confluent, an enterprise company, exactly as akin to like a Elastic’s (NASDAQ: ESTC). story, they formed an enterprise company to provide more enterprise features on the open source package around security and governance in particular, and the reliability the platform, as well as enterprise support.

So that’s the entry point for Confluent is it is an open source company, akin to Databricks, akin to Elastic akin to MongoDB (NASDAQ: MDB), that is built around supporting that open source package for enterprise use. Then, like those other companies, they have a cloud based SaaS component where they have managed hosting of that service as well. And to me, as a hyper growth investor, with a very condensed portfolio, I’m most interested in that cloud aspect of their business, because that’s where the company can scale more. And so I want to see, you know, extensive growth on the cloud platform. And that was what led me to my interest in MongoDB, in Elastic, I haven’t owned them for more than a year now. But you know, back in 2019, I was heavily invested in those companies and following them closely.

Simon Erickson  7:17

It’s really interesting to see the story on Confluent. You know, when you see the founders describe their business, they call it data in motion, just like you said, we’ve built kind of these data silos, or we’re tied to things, thermostats, cars, whatever they might be that were collecting data. But really more interesting is is kind of how does that impact in your organization? So the events, the logs, you know, what does that data mean? the bigger picture, they’ve kind of built, as you said, the middleware to kind of compile all of those and make it more relevant. You mentioned the managed hosting section of this a cloud based side of it, which means, and let me let me double click on something you said there with MongoDB and Elastic, we have done similar things to we know that Atlas, you know, MongoDB database, as a service, which is managed within the cloud has been growing incredibly quickly. How important is the Confluent Cloud to the future of this business?

Muji  8:05

vital, exactly as with MongoDB, I expect a similar arc with confluent, that MongoDB and elastic hat with their cloud solutions, which will be Atlas and elastic cloud. And so, you know, I want them to drive more into that usage based pricing and get their enterprise customers away from self managed clusters. And they can do it for them. And they started adding a component in particular, had a internal project called Project metamorphosis over the last year. And so you know, at the same time the pandemic hit, and they work, restricting, hiring, you know, kind of letting the, the foot off the gas. With hiring and sales initiatives, they did at least take the year to reinvest in the development of their enterprise platform, and specifically to enhance some features of the cloud based service. Or if you are self managed, your own instance in the cloud as well, you can take benefit of some of these things. And in particular, they added tiered storage, which decouples storage from the Kafka brokers, which is the server side of Kafka. That’s vital as a Kafka user. You know, Kafka is be as a middleware, it’s asynchronous as all the individual pieces parts are interconnecting and exchanging data through it, whether it’s producing or consuming that data. They’re coming and going as they need. It can be batch processes that are happening hourly, it could be real time connections that are getting every single record as it appears. It retains the data on the platform, as long as you need. It’s configurable. And this tiered storage allows a decrease coupling of storage from the inner exchange of data. And I see a lot of potential with that, in particular around machine learning, where you can have a historical data set being stored in an underlying cloud storage, which would be s3 as your blob. What tiered storage allows is storing using that as a long term archive that it can tap into. And it can playback data from that archive, or can playback the real time stream, as it always is done. And with machine learning, you know, you can immediately see the benefit there is that you can train on the historical data that’s stored in the underlying object store, and then turn around that machine learning model, and be doing analytics over the real time streaming from there. And so it really does improve the analytical capabilities of confluent. And then they also added some what’s called cluster linking, which is global replication of your data across different data sets. This is going to help them across in multi cloud and on prem. So hybrid cloud environments where you need to exchange data from an on prem cluster to a cloud one, or you need to exchange data between AWS and Azure, because there’s two different segments of your company use those different products, or you’re using different features across those, or you’re acquired a company that uses Azure and you’re an AWS shop, you could be in or exchanging data across your clusters that way, so they really improved the enterprise features over this past year, brought it up to snuff with a competing open source platform that’s starting to gain a little traction called Apache pulsar. And but but, you know, by far, Apache Kafka is the winner confluent will be a winner, it’s just I, you know, I need to see them focus on that cloud service exactly as MongoDB did with Atlas.

Simon Erickson  11:49

Perfect. And so it sounds with this managed hosting and kind of this transition, it’s more of a service than our product, right? Like for people that don’t have either a huge IT budget to build it themselves, or they don’t have the talent of data architects like yourself, Mooji to build these things. I mean, they want to hand this off to somebody like confluent to do it for him. And then they say, Okay, let’s, let’s charge us, you know, on a usage base on a monthly basis to get up and running, what it is that we want to do.

Muji  12:16

You’re actually butting up against where I’d like to see them go. Because with all these open source companies, they’re providing infrastructure. And so with their managed service, they’re providing managed infrastructure for you, I can host your Kafka, it Kafka, confluent, cloud hosts, the Kafka cluster, wherever you need it in whatever cloud provider you need it in. And even on prem, you know, then that’s their, you know, self managed side. But they can give you tools to install a cluster with Kubernetes, for instance, and ease the pain of setting things up. However, I would like to see them go into that managed service versus managed infrastructure where it’s no longer the customer. It’s a little more turnkey, from the customer’s perspective, I don’t need to be fiddling with configuration options, I just need to say I need this cluster, it needs to have this capacity in this cloud. And then you just hook all the agents up into it. And so I think that’s the ultimate path for confluent. Unlike I mean, elastic starting to do that, I guess I’d say unlike MongoDB. So you know, ease of use less IT staff to not only install it, which you save with confluent Cloud, there’s no Installation, but you don’t have to manage it to you don’t have to tune it. And it is a very high maintenance software over time. So I’d like to just see them erase all that make it a managed service versus managed infrastructure, ultimately.

Simon Erickson  13:53

So transition that to the next topic I want to ask you about is it seems like anything that is added as a service platform as a service software as a service, or whatever, as a service, like we were just discussing, is still built upon, kind of this oligopoly of the cloud service providers, right, you’re probably working in one of three or four different ecosystems that everyone’s familiar with, which is great, it makes you more profitable, you don’t have to worry about building the infrastructure yourself. But on the other hand, we’ve seen a lot of these types of companies that we just mentioned, competing against those cloud Titans later on. You mentioned Elastic Search, which put out a kind of a report that they weren’t so happy about one of the things that that one of those cloud service providers was copying. That way we know that MongoDB with their document, databases was competing against one of them, who is wanting to launch their own document database. How do you think about the relationship with confluent or any of these companies? It’s built upon cloud infrastructure with those cloud service providers themselves. Yeah,

Muji  14:54

you just reminded me about a blog post. I think it was by elastics. CEO titled AWS not okay, exactly. Yeah, it’s I don’t, I don’t understand AWS stance here. Clearly Microsoft and Google are not copying that stance, they are more likely to partner with those. x experts have open source, and Whoa, and allow, you know, and better integrate with their cloud solutions. You know, so Microsoft and Google are both partnering with elastic, both partnering with confluent. In order to stand those up. Now, they also have competing kind of messaging as a service platforms, you’ve got pub sub, which is publish, subscribe, the whole underlying kind of paradigm of Kafka, you’ve got competing services on those native platforms, where any company like Kafka benefits is that they’re cloud agnostic, you know, so they can, if you’re going to use kinesis, on a part of your application, to talk between AWS components, such as a lambda function and a hosted dynamodb. You know, you can use kinesis to inter communicate between your Amazon portions, but you can’t use it outside of Amazon. And so, you know, the whole value prop of Kafka is that it’s, it’s works across all those things, it’s the river, across clouds. And so, you know, there’s a lot of value to that much less of the experts and built the software. The Amazon is just, it likes to, it wants to eat the margin, you know, it wins. Either way, it wins, the more elastic cloud is successful, the more Atlas is successful, the more confluent Cloud is successful, but they want to eat those margin points and provide a managed hosted, hosting for those and did it with all three of them, it had a managed MongoDB service, it had a managed elastic service, it had a managed Kafka service, in retaliation to AWS is kind of aggression on that front, meaning it’s basically just taking open source software and hosting it as a service and profiting on it without re contributing back into the open source software itself. And with elastic, they took a very drastic turn with their, their I forget what was called Open, open searcher, open elastic initiative. But they, you know, after they stood up, it was a matter of weeks, and after Amazon stood up managed Kafka service, and hosting, that confluent changed its licensing. And so all of them back in, you know, I’m gonna say 2018 changed their licensing scheme, specifically in to retaliate against AWS. And they weren’t so vocal about which particular company at the time. And I’m sure there were some Chinese clubs doing it as well. But, you know, in specific retaliation to AWS, creating these services, and so they all change their licensing on the open source model to prevent, they branched from what was a patchy version to licensing, which is, you know, open source for all any new features that these companies created, from that point, were under there, either community or enterprise licenses from there. And so they all started community licenses, which is basically we’re going to be Apache, like, open source, but we’re going to prevent you from standing this up as a managed SAS service in direct competition with us, the founders of this company. And so I, I, I love open source, I was extensive open source user in my role as data architect, I’m not surprised by this move, it was it was more than rude to basically try to erase their businesses, as it’s done with countless, you know, folks on the e commerce site. And so, you know, it’s just a, an aggressive company that way. And so it’s, you know, they really had no choice they in order for their business to remain afloat and for this open source package to continue to be relevant from here and be supported from here and have new features added from here, besides community added features. Their company had to remain profitable. And so I you know, there’s a lot of purists out there with with open source I, I’m not that

I think they made the right move with their licensing and as an investor in these companies, absolutely. They made the right move in this company because it keeps their companies relevant, and keeps the open source project alive.

Simon Erickson  20:07

It’s really interesting Muji, I’ve gotten definitely more of an appreciation for these community licenses through reading your blog, you know, it’s something I don’t think there’s really appreciated, but then you kind of see the impact of hortonworks, who built upon the Hadoop open source platform. And then you ended up merging with Cloudera, because they needed enterprise support, they need to make some money eventually, from something like this. You see, MongoDB, who built Atlas SAV is a managed service, you know, of an open source. But then, you know, we saw the resignation of Eliot Horowitz, your CTO just last year, because of kind of direct competition from some of the cloud providers, this is changing the relationship in an interesting way between kind of these building blocks, right?

Muji  20:46

It is, and you know, I can’t fault the the cloud providers for providing their own solution, it was taking the raw open source and just straight up, we’re just going to manage this for you, you can still maintain all the infrastructure parts of it, and the configuration and such but, and their services, Amazon’s in particular weren’t that good, it’s managed elastic service and managed Kafka service was just, it’s, we’re gonna throw up a cluster for you, it basically just automates using EC two, and standing it up yourself. And so the services were not really anything special. Cloud era is a little bit of a different case, you know, work cloud era, I think, faltered you know, brilliant move with Hadoop, in general, it was a, you know, game changing, paradigm changing way to do analytics, which was separation of storage and compute. So, you know, you had storage as HDFS, and then the compute was all kinds of that ecosystem of products that existed on Hadoop, where cloud era, the company was slow was the cloud. It was, again, an extensive Lee complex system to stand up and then manage going forward from there, you had to manage user access rights, you had to manage, you know, at all your kind of commodity systems that you’re adding into the pool of systems that that Hadoop was running on. It was, it was a nightmare to run. And so required an extensive IT team to manage and set up much less the underlying costs of the hardware and things like that. They were slow to adopt the cloud model of where they could just, you could run Hadoop in the cloud as a managed service. And Amazon, I think, really ate into their entire existence, because of the managed to do, service that they provided. And so you know, they were just slow to go to cloud from the on prem mentality, I think that’s where all the other open source kind of data related systems like confluent, like Mongo, were paying attention and realize that, you know, folks don’t want to run this themselves, they just want to stand it up and immediately use it, and not have it, you know, have it be as frictionless as possible, and are willing to pay for that. Because guess what, it’s pretty much the same costs, and maybe even lower than standing up all the hardware, you always have to overbuy the hardware, because you want to have as much capacity as your peak usage. And so, you know, you had a lot of idle hardware sitting around because you’re not running at peak most of the time. But then you had to manage it going forward. That was, you know, the, the hidden costs of the total cost of ownership of these systems was the fact that you had to do, you had to patch the underlying operating system, you had to patch the, the, the the database itself, you had to have ongoing maintenance with some of these things internally, within the database systems like elastic, you had to run optimization techniques, you had to archive old data. So it required a lot of hand holding, in general, all of these systems. And so I think that’s, that’s kind of why I mentioned with confluent, I’d like to see them go to a managed service versus managed infrastructure, the more levers and knobs that they can take away from from users, it makes it more and more frictionless to use these things. And then they adopted a pay as you go model and usage based billing on confluent cloud with some discounts for annual commitment. And I think that’s a great stance to take. And I think that’s ultimately where they’re gonna go. Elastic went even a step further, they, they not only created managed SAS services, much less their enterprise, you know, self managed licensing and support, but they started to spin out individual SAS SaaS services for a managed service where you didn’t, you didn’t care about the infrastructure at all. You didn’t even care that it was Elasticsearch under the hood. You could throw in and Hi in your entire enterprises website, your entire enterprises tool set and have searched over it, you could be searching over slack and GitHub, and emails per user, with each user having their own certain, you know, visibility rights over the content. Those managed services are were an exciting addition for elastic, I don’t know if they’ve made full use of that direction. But I think that’s a good direction for these open source companies to take is starting to build managed services, versus managed infrastructure.

Simon Erickson  25:32

And it’s certainly beneficial, not just for the cost, but also the ease of use, if somebody doesn’t have to configure everything, like you just mentioned themselves to say, Hey, take care of this for us, we want to get up and running. Let me use that as a transition to our next topic, which is usage based pricing. Because the cloud is now gaining adoption, you’ve got services like confluent, and others that we just mentioned, getting getting companies up and running in the cloud quickly, it seems that we’ve seen a transition how a lot of these businesses are charging their customers for those services used to be kind of the famous, you know, per seat model, right? How many users do you have times how many months you put in a license together, and then you just kind of get paid for the year, three years, however long the term runs, we’re seeing a lot of that much more, as you just mentioned, into usage based pricing, not just with confluent. But with other companies, too. Twilio is doing this. We know that Splunk has adopted this we know that fastly has adopted this snowflake is usage based, this seems to be the new metric to look at, if you’re investing in this space. How do you think about that? Is that a metric? You’re also interested in Mooji? Or is there other ways that we can get insight into how these businesses are doing out there?

Muji  26:39

Oh, yeah, I mean, it’s it’s definitely a separate class of SAS, you know, versus the, you know, per user, or you know, even like per system pricing of something like data, dog observability platforms. But typically, things are around per user for service like Okta, in cybersecurity is pretty heavily per user. What with usage based, you basically have this class of SaaS company, that is a toll booth to me. And so they built up something of value. And the more you use it, the more you pay. And like I mentioned, confluent is starting to go that pay as you go. method, it’s fairly new for them. But again, it kind of reduces the friction, you don’t have to worry about the underlying infrastructure so much. But companies like snowflake, as you mentioned, Twilio, fastly are all usage based companies that are marking up, you know, whatever it costs them to provide the underlying platform. Now, the key to understand these companies is that they pay what I call the infrastructure tax to the underlying architecture that they sit upon. And so with snowflake, that’s, you know, all three cloud vendors, with Twilio, that’s the, you know, the the inner communication network that they built with telcos and they have to pay telcos. You know, fastly, it’s providing bandwidth and marking up from there. So it’s, it’s, what I watch for with these companies is I want to see them as they scale up as a with success of their company, is that I want to see them paying fewer gross margin, or I want to see gross margin going up, I should say, I want to see them paying less infrastructure tax. So to me, that means the operating leverage of the company is increasing. So as they get more and more efficient, I want to see that reflected in the underlying financials of the company. So that’s the metric I watched the most with these companies is gross margin. And with snowflake, I want to see that slowly ticking up. And that’s exactly what it’s doing.

Simon Erickson  28:54

It certainly makes a lot of sense. A you want to see that that operating leverage. As you just mentioned, another common metric kind of look a bit about these software’s of service companies is net dollar base protection rates, right? How much are you getting from your customers or the sales that you’re getting from existing clients? This year versus last year? We’ve seen some of those. I mean, it used to be Mooji, that 110% was pretty good. A 10% growth was was now now we’re seeing companies like snowflakes hitting 150 160% database retention rates, that means that I think the cloud is is getting some pretty good adoption out there.

Muji  29:29

Yes, actually, I should have mentioned that as the other metric I look at is NRR or any our net retention rate and that expansion rate, whichever they’re reporting, you know, whether factors in churn or not is basically the main difference there. And you want to look at the notes in in the filing and make sure you understand what is exactly being reported. But in general, I look at net retention rate, and absolutely you want to see that, above 120% or more. Because again, this is a usage based dream. company. So besides gross margin going up, I want to see land and expand. So an RR is the expand part, existing customers are spending more and more as time goes on, but also want to see a lot of customer growth is this. Does this service have wide applicability in something like Twilio, you know, every app needs communication methods. And they’re serving as you know, the communication cloud now. And so you’re going to tie into them for communication services. And so their customer growth remains incredibly high. As more and more apps get developed, more and more developers appear. And then as those apps get more and more success, and grow as an app, or a service, that’s where the net retention rate should be seen to that you see usage rising over time.

Simon Erickson  30:51

The switching costs that appear after after amount of amount of time you’re getting more and more value as you’re using this. You’re paying more for that but you’re also deriving a lot more benefit of it as a client