Trust in Data, Data in Trust: Perspectives from the CMA Media Council

Every meeting, our council discusses one pressing issue to understand the implications, appreciate new perspectives and consider opportunities. This time we chatted about data. Data is one of those things that has been weaponized, hyperbolized and metabolized into every aspect of marketing but in particular, media strategy, planning, buying and optimization. We talked types of data, applications in media and the value of data lakes to a brand, agency or partner.  


First-party data is the information that companies can collect and own from their properties – from both online and offline sources – such as the company’s website, mobile app, CRM or surveys. In the context of display advertising, first-party data is most often cookie-based data. Some examples of using first-party data include:

  • Email marketing;
  • Facebook custom audiences;
  • Service support, follow-ups and loyalty offers (using CRM first-party data); and
  • A TV broadcaster using set-top box data to understand their own audience.

Second-party data is info from trusted partners. You may not have a direct relationship with the partner’s customers, as you do with first-party data, but you’re still able to access it through a relationship with another organization. This data could be valuable because competitors may not have access to it and it can help form a larger, more holistic view of your customer.

Third-party data is data that you have from outside sources that are not the original collectors of that data. These companies are often referred to as data aggregators and they bring in data from various platforms, surveys, companies and websites where it was originally generated. These aggregators have a deal with partners, publishers and other data owners for their first-party data. They collect it into one large data set, often improve upon it through cleaning, modelling, and integration and then sell it. Third-party data is great for demographic, behavioural, and contextual targeting and plays a critical role in solutions like audience targeting and audience extension. The following is a good example of using third-party data in digital advertising. If you are a car dealership and you are looking for a niche audience such as automobile shoppers with intent to purchase within next 3 months, some of them may be found within your own, but not all.

Hence, this is where third-party data suppliers come in. They can gather audience data from both online and offline sources and use various methodologies to find people clearly researching their next car purchase. This data can then be onboarded to extend the target audience the car dealer had from their own site, thereby overcoming a challenge of using potentially undersized audience pools from first-party data. A couple other examples of using third party data include: understanding your customer and consumers (i.e. demographics, psychographics etc.) and a TV broadcaster using Numeris data to understand their own audience and competitive audiences.


It goes without saying that trust in data is more important now than ever before. Unfortunately, many brand marketers have data overload. Data is presented by vendors and agency partners as being the very best and most trustworthy. Platforms are supplying data around brand lift and research companies are using data to inform marketers who their customer is and how to reach them.  The content agency is showing the marketer data to understand what gets brand engagement and the PR agency is showing data that there are more impressions in media.  

The only data point that matters is sales. The C-suite is looking at sales data and saying sales are down. The challenge is pulling together the data to understand how, why and what decisions to make next to improve sales. The justification of any marketing metric needs to ladder back to sales.

There is more than enough data, but what data and whose data do you trust in order to make effective decisions? In order to shed a little light on this question, we discussed the ways publishers, agencies and advertisers can help to ensure high-quality data to generate better business results. These include:

Collaboration and new ways of working – The best way to address data quality and ensure people are singing from the same song sheet is to bring the experts together and work together to build confidence. They should develop a framework for measurement and decision making. The worst way to approach data is to have teams working in isolation from each other and making decisions that can contradict each other. If the data is in real time and performance marketers are optimizing quickly there should be an approach to how they collaborate with the client for approval or parameters for decision making.

Data sources, merging and solid benchmarks – Everyone needs to understand “what good looks like”. There should be a discussion at the outset of a campaign or project that clearly defines a vision, mission, objectives, key performance indicators and the metrics associated with those KPIs. This is when the team decides on the data sources and how they merge for decision making either based on a cross-channel measurement plan, or a customer journey model. Vanity metrics should be pushed aside if they don’t tangibly lead to a business outcome. This should also be the time when benchmarks are put in place to understand what is happening today that will influence the direction of an outcome for tomorrow.

Exploration from all angles – The agency teams should understand collectively what is good and what influenced the result. There are many times when a conversion issue is attributed to a media agency when in fact they don’t control the user experience or site traffic. On the opposite side, the digital team may have a conversion issue because the quality of the audience to the site is poor. Collectively, the team should be willing to explore and ask difficult questions to get to the right decision and chart the new path forward for the client. The magic of a metric or a data point comes from knowing the context and not exploration in isolation. This comes from looking at the market, economics, behavioural and other research that shows the reasons why. It can also come from shifting the visualization of data and understanding the competitive landscape. This exploration behaviour should always be in place no matter the client or project.

It is well known that the industry is struggling with the excess of data that can create decision paralysis or overreactions. We recommend a level-set for executives to revisit the business plan and work backwards as to what helps demonstrate the value of marketing and how they are showing that story through data with full context of the reasons why and the value of the plan.

Authored by the CMA Media Council

Tell Us What You Think
  1. If you haven't left a comment here before, you may need to be approved by CMA before your comment will appear. Until then, it won't appear on the entry.
    Thanks for waiting. View CMA's Blogging Policy.