This article was first published on the Digital Marketing Magazine here.
‘Second party data’ is data that comes from external partners that exposes lots of valuable information about consumers and their online activity. Marketers today are constantly inundated with data from many sources. However, to really make the most of our digital campaigns, marketers need to understand how to use data to get true value from it.
By its very definition, programmatic is the application of data to marketing; but while this is a highly accurate description, it’s not very helpful one. Very little modern marketing occurs in a data vacuum, so it's often more useful to think of programmatic as exposing the data marketers collect, to inform media buying decisions in real time.
The concept of ‘real time’ is an important distinction. Display ad slots are loaded by users millions of times a second and a completely fresh auction happens for each one, to see which advertiser shows an ad. In this environment, marketers need to know precisely when and where they want to show an ad, and data helps make these determinations. In an "if this, then that" approach, marketers look for key signals in the data to make four basic decisions: Should we show an ad to this user? What ad should we show? Where should we send anybody who clicks on it? And, how much should we bid?
If these decisions are based on data, then data becomes invaluable. Not in a nebulous "knowledge is power" way, but in a real-world measureable way. As an advertiser buying programmatically, the goal is to show as few ads as possible to users who aren't likely to be positively impacted and to push content, and budget, towards those consumers who have the potential to buy a brand’s products. By programming all these decisions in advance, the buying platforms (DSP, Demand Side Platform) can make these determinations in real time based on the data. With that in mind, every piece of data that filters out poor quality impressions has value.
Categorising data as first, second or third party
‘First party data’ is the data provided directly to marketers by users of their owned channels, typically from website analytics, app usage and customer relationship management (CRM) systems. This provides marketers with the email addresses of customers and offers access to cookies from their browser history. Whilst this data comes at a smaller cost, the value is usually low because it doesn't reveal more than what a brand will often already know.
Similarly, second party data forms the same basis but comes directly from a partner brand or agency’s platforms. For example, by paying an ad platform to run a campaign, the statistics about that campaign are second party data. Yet, despite being much cheaper than third party, this form of data is also the toughest to utilise.
Finally, third party data gives marketers the most flexibility. It is not limited to only users a brand has interacted with, but can access cookie pools from the wider marketplace. Unfortunately, this third party data comes with a much higher cost, often sold by cost per million impressions (CPM), but marketers need to balance this cost against the saving made in terms of both time and resources from not serving ads to the wrong audience.
Second party data is often overlooked
Of these data options, second party data is one of the toughest sources to use. It often comes in the form of an Excel spreadsheet, a marketing dashboard or even locked into a proprietary user interface. All of which makes it very hard to us to leverage. Unhelpfully, each digital channel also has at least one source of targeting information which is unavailable anywhere else.
The whole practice of search is based on keywords that indicate intent, for example, "cheapest car insurance" tells us the user is price sensitive and "sell my car online" shows intent for an online transaction. What display advertising allows marketers to do is to target user context and what they are reading or watching at the time. For instance, if a user is reading a page about skiing gear, that information can used to target them with an ad for skiing holidays, rather beach breaks.
Each piece of data collected provides invaluable guidance for ad targeting but can also be limited by the channels it’s sourced from.
Remarketing second party data across channels
However, with a bit of work, digital marketers can expose data which is valuable data across all channels. The technique is to extract data that comes through implicitly via online traffic from partners and use that in a first party data environment.
When planning data strategies markers need to consider what works best, and equally worst, for each channel and how those differences can be managed to make informed, strategic decisions for future engagements.