Remote Data Sourcing Explainer Videos

"The simplest way I explain remote sourcing is this. Traditional CDPs work like old external hard drives. To use the data, you have to copy files across. Back and forth. So the moment something changes in the source system, the copy inside the platform is already out of date.

That’s how most CDPs still work. Extract it, transform it, load it, sync it, maintain it… Remote sourcing is completely different. There’s one live version of your data. The platform connects directly to it and queries it when needed. No copies. No pipelines. No drift. Your warehouse stays your warehouse. We just read from it."

"Here’s an example. A customer upgrades their account at nine in the morning, but the CDP synced overnight. So, at midday, the platform still thinks they’re on the old tier. Your campaign then fires, sending them an offer they no longer qualify for. That’s not a bug, that’s the architecture working exactly as it was designed. The messaging layer is real-time, sure, but the data behind it is hours old.

And, the more dynamic your business is, the more visible that gap becomes. For financial services, telco, retailers… customer states change constantly. Remote sourcing queries the live source at the moment the campaign runs, not last night’s snapshot."

"And when you put the two models side by side, the difference is stark. Traditional ingestion means you extract the data, transform and load it into the platform, schedule syncs, maintain pipelines, troubleshoot, manage a second governance layer… And after all the hard work, you’re still working with a copy that’s already going stale.

With Remote sourcing, you connect once! The platform queries your data live with no extraction or transformation, or duplication. It’s the same segmentation and activation. But the data never moves or becomes stale. And your governance stays exactly where it was.

This is more than a small improvement. It’s a fundamentally different architecture. One model creates a second data estate: the other activates the one you already have."

"Here’s a scenario every marketer has lived through. You want to target customers who own a specific product and recently raised a support ticket. Simple enough, right? The data exists. But it sits across three different systems. And your CDP doesn’t have it because nobody’s built the pipeline yet.

Basically, your campaign request becomes a development project. You’re waiting weeks. Or you just work with whatever’s already in the platform, even though you know it’s incomplete. Neither option is good.

With remote sourcing, the data team connects the source once. Maps the tables once. And from that point? Marketers self-serve. Live data. No tickets. No pipelines. No waiting. The dependency just… disappears."

"And it keeps compliance teams awake at night. Every time you copy customer data into another platform, you’re not just copying records. You’re creating another set of access controls. Another deletion workflow. Another audit trail. Another residency question.

Under GDPR, under any serious regulatory framework, every replicated dataset expands your compliance surface area. And there’s an irony here that’s hard to ignore. The platform that’s supposed to help you use customer data more effectively… is also making that data harder to control.

Remote sourcing removes the irony. The data never leaves. Governance stays where it is. Your security team doesn’t need to learn a new environment. They already own this one."

"Everyone’s talking about composable CDPs right now. Keep the warehouse as the source of truth. Let the activation layer work against it. Sounds great. But the question nobody asks: if the platform still copies your data before it can use it… is it composable? You’ve just moved the duplication problem one layer down. Instead of copying from the CRM, you’re copying from the warehouse. It’s the same problem, but a different pipe.

Remote sourcing is what makes composable architecture really work. The platform extends its data model into the warehouse. Queries it relationally. Treats it as part of one customer view. With no copies or secondary storage, your data can power real-time marketing from where it already lives."