Industry lingo creates easy shortcuts for insiders. It lets us say more with less, so we can make our points faster and get on with things. The trouble is, when we don’t use it correctly or if we start to build more into the definition than is really there, it gets us into trouble faster too.

Jake Moskowitz from the Emodo Institute breaks down the misuse, misperceptions and misunderstandings of media terms we use every day.

#1 Deterministic

Scientists use this term to describe data that they know for sure is true. When the marketing industry borrowed the word, we altered the meaning to data we collected in a raw form and didn’t change.

The problem is, we’ve lost the truth part, which, with data, is essential. Saying that we haven’t changed the data doesn’t mean the data was ever accurate to begin with. If we care about data quality, it starts with knowing that it’s true.

#2 First Party

While we can all agree that if I say I have first-party data it means I gathered it myself, the problem is everything else people read into it. Our industry continues to support an assumption that first-party data is superior to other data. That may be true from a privacy or usage rights perspective, but it’s definitely not the case generally.

First party refers only to who gathered it. That’s the only thing we can derive from it. Not accuracy. Just because we’re closer to the data source, doesn’t always make it better quality data.

#3 Transparency

The split-personality of how we use this term is emblematic of the larger split in our standards for media as compared to data. In a media discussion, transparency means proving that you get what you’re paying for, such as with IVT or viewability measurement. With data, transparency is about knowing where it came from.

When you buy a segment, you deserve to know not just where it comes from, but that the audience you paid for accurately targeted the audience you’re seeking. Let’s bring the same stringent media standards to data.

#4 Accuracy

The desire for accurate data may be fairly universal, but the definition is not. Accuracy is about a level of correctness, whether something is more right or wrong. In usage, however, people often use the word Accuracy to refer to precision instead.

We can be precise — incredibly specific about something — but not accurate. Let’s not let a high level of detail in our data distract us into thinking that data is correct.

#5 Scale and Quality

The challenge we find with scale and quality is not their definitions but their relationship. Speak however you like about each one on its own, but when they’re together, they’re negatively correlated.

It’s not possible to optimize for both scale and quality. The more stringent you are about a look-a-like model, for example, the fewer devices you’ll find. Any company that claims to give you accuracy without impacting scale needs a closer look.