Thoughtworks Technology Podcast
Oct 1, 2026 · 37 min · 11 segments
The way we think about and build with data is changing. Data lakehouses have become an industry standard, and conversation is turning to semantic and governance layers as organizations seek to…
Lauren HartmanGuestUm, well, the definition I actually build against and I prefer to use is that it's the smallest valuable self-contained unit of data that a single team owns end to end, um, with a clear purpose, with its own life cycle, and also with a defined way to be consumed.
And the impor-important mental model for me is that a data product is an architectural quantum.
Um, it's the smallest unit you can deploy and operate, um, independently, but it bundles more than just the data, right? It, it's the transformation code.
Um, it's the access policies, and it's, it's the way you, um, input and, and output data from, from that product, so it's input/output ports as well.
Every definition around of data products and what I've just emphasized and stressed earlier, the self-contained, the owned, the consumable, it never specifies who or what the consumer is, and the consumer was never only, um, a, a human.
So output ports have always fed machines to an app calling an API or another data product reading a SQL endpoint or, or a file.
The real common thread is that every consumer we've designed for, human or machines, came through a predetermined structured interface.
So a person clicked a visual that someone else, an engineer, had already built or an app bound to a schema you'd, um, you'd agree on in, on advance.
Either way, the questions were decided up front, and the consumer was one that you can name and interact with.
Now, with, with, um, with the AI, we're seeing that what's genuinely new about these conversational agents is it's not the machine.
It's the f- it's the fact that it's the first consumer that formulates its own questions.
So the gap that we're here to talk about is this is a port, an output port with consumers who you didn't, um, get to design for.
I can finally talk to a data store without looking like a total, you know, buffoon." And so that's, that's always friendly.
I, I do like the fact that I can actually, like, type in, in sort of, you know, my broken English, "Here's what I'm trying to do," and then have it generate those kinds of things.
Where is that becoming useful for companies? Because I, I, I think some of the conversations I've had with clients, in fact, I just had one of these last month, you know, big mainline retailer, you know, lots of experience, and, you know, they're, they're somewhat concerned about what if this, this startup comes out of nowhere and they can create products? And, and sort of my response was, "Yeah, but you've got these decades and decades of data about your products and what works and how to price them and, and what that actually does and lead times, et cetera.
And if you can combine that with some of the agility that these, these smaller competitors have, like, you can get faster, faster than they can get decades of data," was my point to them.
I think a lot of organizations are really seeing the value within data products, especially when you can tie in that business context and really model out, um, through semantic modeling the business context.
And like Larisa was saying, as an outport, and now we're seeing it as data agents as the outport, whereas before it was just more concretely, um, reports and dashboards which were predefined questions that were being answered.
And now organizations are able to go and ask the data directly whatever question that might come to mind, and I think that's where it's becoming super important of the idea of garbage in, garbage out that you've probably heard a lot of.
Um, well, the definition I actually build against and I prefer to use is that it's the smallest valuable self-contained unit of data that a single team owns end to end, um, with a clear purpose, with its own life cycle, and also with a defined way to be consumed.
And the impor-important mental model for me is that a data product is an architectural quantum.
Um, it's the smallest unit you can deploy and operate, um, independently, but it bundles more than just the data, right? It, it's the transformation code.
Um, it's the access policies, and it's, it's the way you, um, input and, and output data from, from that product, so it's input/output ports as well.
Every definition around of data products and what I've just emphasized and stressed earlier, the self-contained, the owned, the consumable, it never specifies who or what the consumer is, and the consumer was never only, um, a, a human.
So output ports have always fed machines to an app calling an API or another data product reading a SQL endpoint or, or a file.
The real common thread is that every consumer we've designed for, human or machines, came through a predetermined structured interface.
So a person clicked a visual that someone else, an engineer, had already built or an app bound to a schema you'd, um, you'd agree on in, on advance.
Either way, the questions were decided up front, and the consumer was one that you can name and interact with.
Now, with, with, um, with the AI, we're seeing that what's genuinely new about these conversational agents is it's not the machine.
It's the f- it's the fact that it's the first consumer that formulates its own questions.
So the gap that we're here to talk about is this is a port, an output port with consumers who you didn't, um, get to design for.
I can finally talk to a data store without looking like a total, you know, buffoon." And so that's, that's always friendly.
I, I do like the fact that I can actually, like, type in, in sort of, you know, my broken English, "Here's what I'm trying to do," and then have it generate those kinds of things.
Where is that becoming useful for companies? Because I, I, I think some of the conversations I've had with clients, in fact, I just had one of these last month, you know, big mainline retailer, you know, lots of experience, and, you know, they're, they're somewhat concerned about what if this, this startup comes out of nowhere and they can create products? And, and sort of my response was, "Yeah, but you've got these decades and decades of data about your products and what works and how to price them and, and what that actually does and lead times, et cetera.
And if you can combine that with some of the agility that these, these smaller competitors have, like, you can get faster, faster than they can get decades of data," was my point to them.
I think a lot of organizations are really seeing the value within data products, especially when you can tie in that business context and really model out, um, through semantic modeling the business context.
And like Larisa was saying, as an outport, and now we're seeing it as data agents as the outport, whereas before it was just more concretely, um, reports and dashboards which were predefined questions that were being answered.
And now organizations are able to go and ask the data directly whatever question that might come to mind, and I think that's where it's becoming super important of the idea of garbage in, garbage out that you've probably heard a lot of.
The rest of this transcript — segmented and speaker-labeled, so you land on the exact moment something was said
Search every transcript — by keyword, by phrase, or by meaning, across every show Radar indexes
Trends — what is surging across podcasts, measured against its own baseline
Alerts — when a name you follow appears in a newly indexed episode
No account is needed to search Radar.
