In a digital landscape inundated with ads, there is rising pressure for advertisers to reach their target audiences. In this article, Kevin Tan, CEO of Eyeota, debunks some common misconceptions about programmatic and audience data.
“Cookies mean no privacy”
Cookies are in large part aggregated and anonymised. This means that personal information, for instance, your home address, email address and phone number, is not collected. Although they get some bad press, cookies are vital to the ecosystem because they allow users to have personalisation of content. It also enables brands to learn more about their users’ favourite content by tracking cookies so that the same ad does not repeatedly appear.
“I just need to buy inventory programmatically, I do not need data”
The average time taken from the start of real-time bidding to the time an ad is served when the page loads is 150 milliseconds. This allows media buyers to buy ads rapidly through automated programmatic buying processes, replacing the once tedious procedure. The fallacy is that you achieve uplift quickly once you use programmatic, but even this machine learning plateaus quickly. Audience data is needed to help marketers discover: where to buy ads, what messages to deliver and what time to send it.
“First-party data is all I need”
Audience data is classified into first, second and third-party data. They all offer marketers unique insights, yet often, marketers believe that first-party data from their own location, transactional or CRM data is all they need. But, first-party alone does not allow marketers to understand their target audiences beyond the realm of their websites, and because of this, marketers often apply lookalike modelling to create scale. Yet what some marketers do not realise is that third-party data can achieve this scale – without losing the quality of their data in the process.
“My budget for digital advertising will cover targeting”
The central puzzle piece to an advertising campaign is a proper data strategy. Without this, marketers cannot focus in on certain aspects, for example, your exact target user. Although putting aside a budget for digital ads is a step in the right direction, marketers also need to figure out what percentage of their advertising spend will be set aside for actual targeting. Having the budget for digital ads does not equate to targeting.
Marketers: “The more the merrier, when it comes to data”
Marketers today are blessed with the bundles of customer data which can be used to understand their target audiences. However, resources to gather, process and store this data are required also, therefore an approach of gathering as much data as possible may not be the most practical option. Marketers should instead ask which data will be the most relevant for their campaigns – and leave the rest. Research by Forrester supports this claim; they found that only 12 per cent of data collected by brands are actually used. Meaning that it is the quality – not quantity – of data is important to marketers.
Publishers: “Every piece of data should be gathered”
If you are a publisher who attempts to capture every segment of data about your visitors, it will not take you long to realise the hundreds of hours of manpower to set up and analyse an almost infinite amount of data sets. Budgets and interest will also deplete long before any profits are generated. A more rational approach is to use a marketplace to sample the data you are collecting. This will provide a platform to review your data in the context of your competitors and the regional marketplace. You can then start defining the data segments you need to capture for the open market, private sales and your internal knowledge base.
By Kevin Tan