The Data Stack Show - The PRQL: The Beauty of Commoditization

Episode Date: November 26, 2021

For this week's PRQL, Eric and Kostas preview their upcoming episode with Michel Tricot.

Welcome to the Data Stack Show prequel. This is where we talk about the upcoming episode right after we record it. So you get raw thoughts from me and Kostas on the conversation that we just had. Boy, was this an interesting conversation. We talked with one of the founders of Airbyte, who's built an amazing open source product in the data integration space. Costas, I'm interested to know your thoughts on commoditization. So this is something that came up towards the end of the discussion. There's this tension between sort of the inevitable commoditization. So examples like, okay, telephone companies used

to have a monopoly on sort of landlines and telephone service, right? And now in many ways, that's been commoditized, right? You can put a chip in your phone and sort of you buy minutes and all that sort of stuff. And the hardware has been commoditized too. In the data space, we think about warehouses a lot, right? I mean, five years ago, of course, warehouses existed. Right now, like a small company can spin up Snowflake, you know, BigQuery, Redshift, and be off to the races at an amazingly low cost for the amount of sort of storage and around the stack. There's a tension between commoditization and then also sort of the, and I'll use it maybe an unfair term here, scarcity that's needed to actually sort of run and scale a business, right? I mean, you need to have something that other companies will pay you to use. And so how do you think about that tension, right? Commoditization is inevitable because of technological change and advancement, but companies have to retain some level of scarcity in terms of their products to actually make money and

sort of create value in the marketplace. Yeah, that's a very good point, actually. But what I would say is I don't think that companies have to maintain scarcity. I think that for a have to maintain scarcity. Like I think that like for a technology company to do that, it's a mistake. And in the longterm, it's going to end up like with the company.

And it has to do like with the nature of like software and technology. Right. It's not inevitable. Like commoditization is not inevitable. Right. Like we can decide as humans that we don't want like to commoditize something. Like we can do that. But the beauty of commoditization is that it enables innovation.

So a company that, let's say, like operates in the true spirit of like being like a tech company, what they should try to do is like commoditize and then build first on top of that the next innovation wave. And then you think about all the big companies. Think about Amazon. Amazon, online, e-commerce. And then suddenly you have AWS. Why to do that? Because they had to innovate. Microsoft, the same thing. Microsoft, the same thing.

Google, the same thing. Yeah, you have the main product, which is advertisement. And we commoditize online advertisement. Everyone can do it. You just create an account and start spending your money. But Google is not just that. It's like deep mind. It's like even, I don't know, like even quantum computing.

So I think this kind of like we need to create artificial scarcity it's doesn't make sense at the end i think it's more of an insecurity that like entrepreneurs might have but if you won't like to survive in the long term i think you have to move away from that and like what we are going to the person we are going to chat with I think it's a very good example of that because that's what is happening like you have like think about pipelines

and companies that they do that right we reach a point where we are like okay it's solved we have the winner there and suddenly you have our guest and their bite come and like start changing the game. Right. So you have to keep innovating and you have like to keep improving and like reinventing your product in a way. Yeah, I agree. I mean, I, my personal take

is that artificial scarcity, like you said, it may actually create sort of near-term success, but it ultimately limits your ability to create success in the long term. One interesting example of that is Snowflake talking about themselves as a cloud data platform, right? Which has sort of been language that they've started to use relatively recently instead of cloud data warehouse. Right. And it's, I think, I think it seems like they really understand great, like let the cloud data warehouse get commoditized. It's like the more that that's commoditized, actually the better it is for them because they're building componentry on top of it, that sort of makes the data and sort of the functionality of the warehouse even more valuable exactly and the best thing that can happen in a company is

actually create a platform where markets can be created on top of that like think about apple right like you have the iphone but you also have like the app store the app store exists because it was built like on top of like this technology and they opened the technology in order like to do that otherwise it would never happen like if they were like yeah we are only going like to create games we are only going to create like applications and software like for this what would be like today uh this product like it wouldn't reach like the size as a market that it is today. And I think that's what also

all these companies are trying to do. If you check Snowflake right now, they are trying to create network effects through data sharing. And they are trying

to use the data warehouse as the foundation to build the platform and on top of the platform build marketplaces and create ecosystems and all that stuff. Well, like build like marketplaces and like create ecosystems and all that stuff.

Well, I think that's going to be a really interesting thing. And actually, we'll give a little bit of a preview here. So Michelle worked at LiveRamp for a long time, and we didn't actually touch on this in the episode, but to bring our prequel full circle, I think one of the interesting things is going to be if we think about commoditization in the context of data pipelines in the data warehouse, situations where live ramp data actually becomes most valuable in the context of data directly accessible right next to your first-party data. And the use cases and products that are going to be built on top of that are going to be amazing.

All right, well, really interesting episode with the founder of Airbyte. Be sure to tune in because we will publish that here in a few days. And make sure to hit subscribe if you haven't already. As always, thanks for joining us on the Data Stack Show and we'll catch you on the next episode.