Best Practices for Planning a Data Strategy

The most universal challenge facing advertisers today is how to achieve better performance and ROI for their online campaigns. For the last decade, behavioural targeting has been at the forefront of advertisers’ minds in order for them to deliver more relevant information to internet users based on their previous behaviour.

Within the last five years, other signs of data coming from customers started to emerge. For instance, data can be a review that somebody left on a website about your business. It can be social signals, it can be a mobile device ID, and it can even be a person’s physical location, captured via apps and beacons. If the visitors to a website property are considered first-party data, so are the real people who enter a store.

The problem for marketers, however, is that the information provided to them for behavioural targeting is in the past, it is history. Advertisers need to predict consumers’ next intent to buy, and past behaviour is only one part of what makes up future intent.

As consumers alternate between search engines and websites, on mobiles and on desktops, they are leaving behind far more than just behavioural data – they’re leaving a trail of digital signals that are subtly telling marketers exactly what they are looking for and where they are going next, not just where they have been. Savvy marketers must recognize these signals in real-time and turn them into actionable insights, incorporating the right data into campaigns in order to cherry-pick the perfect prospects for better conversions. First-party data is like someone virtually tapping an advertiser on the shoulder and telling that advertiser what they want. While these signals are great, there are not enough of them to power the number of impressions needed. That’s where second-party data and third-party data becomes necessary for most advertisers to get the volume of target audience needed.

Creating a data network is a solution that allows clients to gain access to relevant audiences found in multiple DMPs.  The sheer amount of data that exists from every user’s behaviour on the Internet is endless.  Brands struggle to understand who their own audiences are, and then they are faced with the added complexity of locating unknown users who they believe would be interested in what they have to offer.  Brands are often able access third party data sources, but run the risk of paying for, and using, unverified data sources mainly collected through media buys across the internet that are aggregated and sold on a CPM basis.  Brands also have the option of directly contacting many publishers, vendors, and technology partners on their own to create a combination of first and second party audiences that will be more successful than third party but will take more time and resources to execute.  The creation of a data network would make data strategy execution much simpler for brands – a single source that is able to aggregate verified data with the added option of overlapping the brand’s own first party data for usage, in an incredibly simple way.

1. Before you start using data to target audiences in campaigns, clearly define goals!

Although first-party data is touted as the best for ad targeting, there really is no “best” when it comes to data, only what’s best for the advertisers. It really depends on their campaign goals. For example, if customer experience is a KPI, consider investing in a first-party data strategy. However, if traffic lift is of top importance, consider using third-party data.

It’s critical that advertisers understand the distinctions between the different types of data and the benefits and limitations of each type in order to use those differences to craft strategies for maximum benefit and ROI. If you’re not familiar with the different types of data, check out this infographic that distinguishes between each of the three types of data, specifically what each type is, the benefits, limitations, and examples.

A combination of all three might be the better strategy. However, keep in mind that obtaining the right type of data can be time consuming, and potentially costly – all factors to bear in mind when weighing up the options and devising a strategy.

Given the inherent strengths of first-party data, premium publishers who have access to large quantities of such data are in a great position to use it to build revenue steams by combining their mountain of premium data with an advertiser using their own high-quality, first-party data which will drive superior performance and ROI.

2. Conduct a data audit

Review all the sources from which you capture consumers data – forms on websites, website analytics, social media data etc. and ensure that you are collecting the right kinds of data that will be useful to your marketing efforts. For example, are you making use of customized pixels (to not only anonymously track online behaviour but also to gather deeper information such as purchases made) adding to a more robust customer data profile?

Evaluate your planned data sources (first, second, third-party) within a campaign-centric framework (e.g. upper vs. lower funnel tactics). You might find that the biggest challenge is driving engagement with your web properties – finding something of enough perceived value that will compel consumers to share more and more of their personal data. Customers will come into contact with brands at different stages throughout their buying cycle, and each touch point presents an opportunity to provide relevant messaging that will help move the customer to the next stage of the cycle, and potentially capture more data that can be used to secure a purchase.

3. Create a plan for how best this data can be used

Your data plan is going to include objectives and tactics to achieve your overall goals that were defined at the very beginning, as well as the processes for the management and storage of data – not just the collection and use of it. Your data will help you determine the following, which should be addressed in your plan:

    • Who are you trying to reach? What is the ‘persona’ of your audience?
    • Where is the best place to find them?
    • What creative message will resonate best with them?

Let’s take the example of a grocery store. It has the goal to increase revenue by 10% this year. This is a very broad goal, so a more precise objective would be to turn one-time customers into repeat buyers. The tactics used to do this might be through an introductory offer to join the rewards club, and then by emailing offers based on knowledge of the customer to encourage repeat visits.

Audience ‘personas’ are essentially a wish-list of who you’d ideally like to be reaching – for example, the characteristics of the audience most likely to convert, or the audience that is most profitable.

Audience personas are more than gender, age and location. Marketers want to know the audience’s behaviours and preferences, their wants and needs, their habits and perspectives – this will avoid wasting ad impressions on people who are not interested in what you are offering. When the audience is targeted with more relevant messaging, they are more likely to respond.

4. Embrace a flexible test-and-learn approach to working with data from multiple sources

Advertisers should measure and compare performance of different depths, different types, and different sources of data. It is crucial to understand the online behaviour of consumers’ pre and post conversion, in addition to their needs and buying process – essentially looking at their interests, intentions and interactions.

5. Evaluate data performance against campaign KPIs

Tracking, segmenting and revamping messaging to match the audience is an on-going process.


Achieving better performance and ROI from online campaigns is a universal challenge facing advertisers today. However, successful advertisers are the ones who can recognize the data signals being left behind by consumers, and use these signals to craft campaigns that will provide more value to consumers, and lead to more conversions.

While first-party data is the preferred type of data to use because of its superior quality and relevance, its scarcity means that advertisers have to find other sources in order to power the number of impressions needed in their campaigns, which is where second and third-party data sources play a key role.

Have any questions?

Ask our team of data experts at