A Media Brand’s Audience And Data Strategy: A Bridge Between Current Assets, Future Goals | AdExchanger
Media brands’ survival depends on a sound audience and data strategy. But before the strategy can be devised, a deep phase of audience and data auditing is required.
With results from audits’ fact-finding processes in hand, media brands can move on to define their commercial and marketing goals and create audience portfolios through the following steps:
- Data audit
- Audience audit
- Set goals
- Create a taxonomy representing an ideal portfolio supporting the newly set company’s goals
- Make that ideal portfolio a reality by creating or sourcing new data to fill existing gaps.
The audience audit
The different phases are in this sequence because a data strategy is not a company’s end goal but a means to an end. It is the foundation of monetization and an instrumental element in the achievement of the company’s commercial and marketing goals.
While the data audit focuses on identifying user touch points and individual data points available across the company, the audience audit is aimed at understanding if and how those data points are processed and translated into a single customer view.
The audience audit analyzes current audience profiling, segmentation and taxonomies. It also examines existing tools, platforms and technology partnerships and collects information on the company’s current monetization channels and clients. It should be performed agnostically and extensively since it is the stepping stone of the future strategy.
The audience audit also includes a review of all the third-party tags on the company’s digital properties and the reason why they are in place; it is incredible how much data leakage is caused by old or redundant tags.
A single customer view supporting multiple taxonomies
Once the data and audience audits have been completed, the next phase can be approached with total clarity: the defining of the company’s commercial and marketing objectives.
Sitting down with the CEO, CMO, CRO, COO or other decision-makers is crucial. It is an opportunity for listening and understanding of the board’s needs and plans, but also for influencing choices with a better understanding of how data can shape the company’s future goals. For example, both a paywall and an advertising strategy need to be supported by data in their definition and execution phases.
For the next steps, we need clear targets to shape the audience and data strategy to meet those goals.
A taxonomy is the conceptual representation of the data strategy as a bridge between current assets and future goals.
The starting point should always be the translation of the data and audience audits into a taxonomy and a portfolio of available data segments. One could think of it as a picture of the audience reflected through the characteristics of the content and the data collected, such as articles on different topics or information received through subscription or participation in events, competitions or app use.
The taxonomies must cover the needs of the editorial, commercial and marketing teams separately yet together. As an example, for subscriber-retention purposes the marketing team might need the audience segmented and targetable by subscription package, something the advertising team might not care about. At the same time, knowing that a user is in-market for a smart fridge would be relevant for advertising and of little use for marketing.
While we support the individual departments within the company, building a single customer view is paramount. Publishers’ different business units should avoid having different, partial versions of the same person on different platforms that are managed by different teams. I’m not referring to the challenge of cross-device identity management but to different departments using different tools, platforms, processes and data sources across the company. This is a pillar of any serious data strategy and a classic source of many issues if not corrected.
The media brand’s current portfolio of audience segments will be factual, representing available data sources. The recently defined company goals should be followed by the creation of a second version of the audience segments portfolio, a sort of wish list representing what characteristics and user info are needed to achieve the company’s objectives.
For example, the marketing team might have a goal of lowering churn rates and need to segment the audience by likelihood to churn.
The advertising team should also review any advertisers’ and agencies’ briefs that the publisher failed to successfully respond to due to lack of specific data, a gap that the strategy should overcome.
We now have a portfolio of currently available data and segments and an aspirational version (still in the form of taxonomy, its “shell”) covering all the segments needed for different departments to achieve their newly set objectives.
How do we complement the current portfolio and fill the gaps in the data, with the objective to make the aspirational portfolio become reality and support the company’s goals?
I believe that, except for certain types of data and needs, there are several avenues that should be explored before resorting to second- or third-party data, the latter often of unknown quality and offering no differentiation against competitors.
One powerful tool to increase and improve the first-party data being collected is natural-language processing (NLP). It allows publishers to extract additional data points based on semantics (the content) rather than classifying the article by where it sits within a site’s navigational structure.
A piece of content about how to reduce electricity bills by better insulating doors and windows would once have been categorized as “money” or “home and garden” interest, depending on the section it belongs to. With NLP, the same article might generate the following additional data points: “home and garden,” “utilities/gas and electricity,” “personal finance” or “green and environmental issues.”
Success is in the details
This results in a much wider and diverse portfolio of segments reflecting an audience’s characteristics and interests.
I see the ideal taxonomy as both static and dynamic: static to represent the nature of the business and enable fulfillment of objectives, and dynamic because NLP allows discovery of new topics and trends within the content and creation of new segments and engagement opportunities.
There is a popular misconception that the platform used — which could be a data management platform, a customer data platform or even a demand-side or supply-side platform — will determine the success of a media brand’s audience and data strategy. This assumption couldn’t be more wrong, and it is a barrier and root cause of media brands’ lack of control of their audience data.
Media brands must maniacally create flawless foundations upon which their strategies are built. A platform is just an enabler that’s as powerful as the strategy that leverages it.