Companies have been spending more on data every year, and getting less out of it. The platforms got better, the vendor relationships multiplied, and somewhere along the way the basic questions got lost: Where did this data come from? How old is it? Does it reflect what real people are actually doing right now? Instead, the industry organized itself around volume, cost per record, and whether the dashboard looked full.
Most commercial data purchased for targeting or audience intelligence is older than it looks, assembled through chains of intermediaries whose sourcing practices are hard to trace, and optimized for delivery rather than accuracy.
Integration Solved the Wrong Problem
Customer data platforms took off because the old multi-vendor setup was slow and inconsistent. The workflow did get better with consolidation. Data moves faster, and the friction between acquisition and activation dropped when both happened inside the same system. However, the quality of the underlying data did not improve at the same rate, and platform consolidation did nothing to make that data more accurate, more current, or clearer about where it came from.
Many organizations are now running a faster version of the same flawed operation. The pipeline is cleaner, but the data moving through it is still old, still sourced through opaque chains, and still carrying consent questions that no platform upgrade resolved.
The Sourcing Problem Underneath the Performance Problem
The performance problems with commercial data are well-documented, but they sit alongside a more basic one that the industry has been slow to address. Regulators in the United States and Europe have spent years examining how consumer data gets collected, packaged, and resold, and the findings have not been favorable. The data broker ecosystem that supplies much of what enterprises buy for targeting and analytics offers little transparency about consent, sourcing, or how far a signal has traveled before it reaches the organization using it.
For companies evaluating their data infrastructure, this matters beyond legal compliance. Data with unclear sourcing carries reputational and regulatory exposure that never shows up in a performance report. The growing push toward consent-based and first-party data sources is partly driven by regulators, but it is also a quality argument. Data collected transparently, with clear origins and known age, tends to be more accurate and more defensible than data assembled through intermediary chains no one can fully audit.
The Shelf Life Problem
A large share of what gets sold as data is better described as a record of the recent past. By the time it reaches the buyer, the signals it was built from have already aged, and the audience it describes has moved further along in a decision process that did not wait for them.
Intent has a shelf life. A logistics buyer evaluating vendors this quarter is in a different position six months from now, and a consumer researching a product today will not be in the same window in 90 days. Most data pricing models do not reflect this, which means organizations paying for volume are frequently paying the most for the portion that has already expired.
The more useful way to evaluate a data investment is not cost per record. It is the distance, measured in time, between when a signal is captured and when it reaches someone in a position to act on it.
Processing Is Where Most of the Value Is Made or Lost
Raw data arrives unstructured, inconsistent, and not yet connected to anything actionable. The processing step is where it either becomes usable intelligence or hardens into a liability that looks like an asset.
When processing is handled by a vendor with no visibility into how the data will actually be used, the fields prioritized, the noise filtered out, and the refinements applied all get decided without the context that would make them meaningful. The output is technically processed but may have little to do with what the receiving team needs.
Peter Kazan, founder of integrated data intelligence firm, Atlantic Tech, has built his company’s model around this gap. The timing dimension, specifically how quickly a verified signal can reach the person positioned to act on it, is the part most procurement conversations treat as a logistics detail rather than a quality variable.
Where Atlantic Tech’s Argument Actually Begins
Atlantic Tech does not primarily compete on workflow integration. That argument was already made and largely won by the CDP market. The distinction Kazan draws is upstream of the platform entirely, at the level of how data is sourced and verified before it enters any system. The company’s focus on intent-based acquisition and traceable data sourcing is a response to what platform consolidation left unsolved: the data flowing through a unified system is only as good as what went into it.
Kazan has described the distinction plainly: “The real advantage isn’t having more data. It’s knowing exactly what to do with it, at the moment it matters.” The timing component points to a sourcing discipline rather than a platform capability: whether the signal was captured close enough to the moment of intent to remain actionable by the time it informs a decision.
That framing connects the intelligence and execution question to the sourcing one. “Information is the most potent currency in the modern economy, but its value depends entirely on how it’s used. If you separate intelligence from execution, you lose precision. We built Atlantic Tech to keep those two things inseparable,” says Kazan. In practice, keeping intelligence and execution connected requires owning the data at the point of origin, not just at the point where it gets deployed.
Volume as a Proxy for Value
Volume is easy to measure and precision is not. A vendor can hand over a report showing records delivered and impressions generated. What that report cannot show is how many of those records reflected real, current intent. Procurement processes optimize for what can be counted, and data budgets grow around metrics that look clean on a dashboard even when those metrics are poor proxies for what the organization actually wants.
Atlantic Tech works across B2B targeting and B2C campaign execution, with concentration in logistics and commodity trading, industries where the difference between stale and current intelligence has direct consequences. In those sectors, precision over volume is a practical requirement, not a positioning statement.
What Sophisticated Buyers Are Asking Now
The organizations handling this well have changed the question they ask during vendor evaluation. Instead of focusing on how much data a vendor can supply and at what price, they are asking how that data moves from acquisition to deployment, who is accountable for its accuracy at each stage, and how quickly a signal captured today can inform a decision made this week.
Those questions surface accountability gaps that standard vendor evaluations miss. In a multi-party data arrangement, no single vendor owns what happens between their piece of the process and the next one, which means the organization doing the buying is typically the only party accountable for what the data actually produces.
A Structural Problem With a Structural Answer
The companies spending the most on data they cannot use are rarely doing so carelessly. They made reasonable decisions at each step of a procurement process that was not set up to surface this kind of problem. The vendor relationships look defensible. The dashboards show activity. The issue is in the structure connecting all of it, invisible until the results stop coming, and durable precisely because of that invisibility. The question worth asking first is not which vendor to use or how much data to buy. It is who owns the full lifecycle, from the point a signal is captured to the point it drives a decision.