Measuring effectiveness: where first-party data meets advertising agility

What’s more, research from Gartner shows that top-performing brands are using their own data to better understand their customers and target them with more relevant advertising, leading to improvements in site traffic, revenue and customer satisfaction. The challenge is not that advertisers don’t understand the importance of first-party data and building trust. It is rather taking that strategy and turning it into a realistic plan through to an active deployment.

“Given how valuable first-party privacy initiatives are in staying competitive in the market, survival of the fittest is not about who is the strongest or the smartest but who is best able to adapt to change,” says Alex Wilkins, head of analytics at Incubeta. “We’re not necessarily reinventing the wheel, we’re trying to achieve the same things as last year and the years before that - it’s the methods through which we achieve them that are changing.”

The key is to start gradually testing these first-party data elements to begin the process and build up on it. That way, advertisers start to “build a pipeline to the processes required to understand the techniques and tools that you need in order to lay that foundation,” says Wilkins.

Smarter decisions

Without access to the data advertisers may have previously used, it’s about making the most of the data advertisers do have. Marketing decisions have to be made incredibly quickly - so having all the information ready for the algorithms is key. The cleaner, more accurate this data is, the more effectively AI-powered tools can work their magic with high accuracy to fill gaps, find new growth opportunities and optimize to the results that matter.

“The benefit for advertisers is that if you are using consented first-party data, you don’t need so much data,” says Aymen Tabbakha, data and measurement partner account manager, Google. “It’s about finding the data points you need to deliver a profitable campaign and get the best ROI or uplift in conversion numbers. To create an advantage with first-party data, brands should set a data strategy to guide what data is collected and how it’s activated.”

By prioritizing the collection of consented first-party customer data in volume and on a regular basis, machine-driven processes can then handle the complexity - advanced machine learning, for example, can model that data to make better decisions off the back of it. “When we understand these patterns better, we can make smarter decisions,” says Wilkins. “We’re no longer in a fully observed world, so it’s critical to think of everything as observed plus modeled.”

Modeling tools and products, like Enhanced Conversions & Consent Mode, can take what data you have and extrapolate it to ensure the modeled data is as accurate as possible. “There is a true value we are trying to get towards and first-party data products give us a way to arrive at that true value - or at least get much closer to it,” says Wilkins. “It may not be observed but it is more accurate and better than the data that we have without it. With observed plus modeled, you get closer to the truth.”

Simplifying measurement

Advertisers are thinking about data differently, but when it comes to first-party data measurement, the same campaign measures apply - just using different data points.

“Measuring the success of first-party data shouldn’t be seen as an isolated item, but measured using the quality definitions of measurement that you would use for any campaign,” says Wilkins.

Google’s Tabbakha adds: “When measuring first-party data, you need to be able to relate the success of a deployment to the success of a campaign - that’s how you know if the quality and deployment of the first-party data is strong enough or not. If you’ve seen a successful marketing campaign, then you’ve seen the successful ingestion of first-party data as well.”

Simplifying measurement prevents miscommunication and creates actionability from the data. And it’s important to agree on a simple metric that everyone in the business can understand - versus metrics that don’t mean anything to anyone outside of a very small team.

“The simpler that is, the easier it is for people to understand the impact of what is happening, which makes it easier to have conversations about doing the next thing and the thing after that to build back towards it,” says Wilkins. “That helps to start building momentum about how you are using first-party data as an organization. That metric will depend on what matters to your business, but everything you test should be aimed at ultimately improving that metric.”

A durable framework

With success measures determined at the start of the campaign, any activity should then be monitored and understood in a way that relates to that, and measured again and again to be able to get closer to the value. This can be a simple process:

1. Define the KPIs to measure the impact: you don’t need to reinvent the wheel or create a brand new set of KPIs specifically for measuring the impact of first-party data on campaign performance. These KPIs should reflect the nature of the campaign itself.

2. Monitor it using these KPIs: is the test proving to deliver a positive impact in ROI or cost per conversion or whatever the metrics you’re working to optimize towards based on what’s important to your business? These are ultimately the metrics your privacy-led, first-party data led initiatives should impact.

3. Measure the impact again and again: adopt a test and learn mindset to see the benefits of the initiative, explore pre/post analysis and experiment with A/B methodologies to evaluate the success of leveraging first-party data performance.

Measurement remains consistent in its core objective, which is enhancing campaign performance by leveraging the most pertinent data available. The success of the deployment hinges on whether it has achieved its intended goals.

“Companies who are doing things well are taking privacy seriously, putting time, effort and money into making sure they’re on the right side of things and are significantly more agile,” says Wilkins. “Those who have built the infrastructure with data in mind, have teams who are cognizant of it, and understand what they’re trying to do, are able to access the data very easily and bring in the right tools to drive growth and value.”