“Data-Driven Thinking” is written by members of the media community and contains fresh ideas on the digital revolution in media.
Today’s column is written by Nico Neumann, assistant professor and fellow, Centre for Business Analytics at Melbourne Business School.
Advertising agencies are often masters of creating and developing brands. To survive and keep striving, they have learned to constantly adapt and cater to new customer needs.
In recent years, however, the large advertising holding groups have struggled to grow their stock market values, with several suffering from share price losses over the last 24 months. And many media agencies are still recovering from the aftermath of the transparency discussion triggered by the K2 report in 2016.
The large holding groups have made different bets on what may help them grow again. IPG and Publicis invested large amounts ($2.3 billion and $4.4 billion) to make two data companies – Acxiom and Epsilon, respectively – their own. Likewise, GroupM has built its own mPlatform to offer proprietary data and technology solutions to its clients. The idea seems reasonable: Data and technology will be more relevant in the future, which is why they made corresponding investments.
While it can be convenient to use relevant technologies and data assets that are owned by your agency, there are some important issues that may suggest staying away from this strategy.
1. History repeating: conflict of interest
Media agencies are supposed to act in their clients’ best interests. But if they recommend their own data and technology solutions, clients may question related media spend decisions. Being a data broker, technology provider and agency/ adviser at the same time clearly presents a conflict of interest.
We should not forget that one of the main reasons for the all-time low in client-agency trust is rooted in the conflicts of interest that many agencies previously created to generate more revenue. Hence, one needs to wonder whether repeating the same controversial strategies – first investing in and then recommending the “next big thing clients are after” – will help regain brands’ trust.
2. Customer preferences and reach limitations require data agility
Let’s imagine that every major agency would primarily sell or promote one data source only – for example, the one they own. Betting on a single data source, even when rich in first-party data, is dangerous for several reasons. First, no single data source provides a complete picture of customers’ behavior. Even large ad networks may miss crucial pieces of information when used in isolation, as shown by a peer-reviewed study in 2016.
Second, someone else’s first-party data is still second- or third-party data to your brand. The problem with trying to leverage other people’s rich data assets is that identifier match rates between different data sources are often small. To obtain large enough reach requires access to multiple data sources. The increasing demand for privacy regulations and customer protections will make data matching even more challenging in the future.
Third, customer preferences change continuously. This will also affect the long-term value of different data businesses, each of which may have unique relationships with apps, websites or other companies. Yesterday a data broker deal with Pokemon Go or Whisper sounded good, today a data partnership with TikTok seems worthwhile, and tomorrow working with GrabQPons may be a more promising avenue.
In a nutshell, a fast-changing world requires data agility. And therefore it seems more reasonable to rent rather than buy data whose value can disappear quickly.
3. Don’t disrupt the decision-intelligence onion
Connecting all data pipes with a company’s legacy tech stack is cumbersome. While implementing new solutions typically takes six-12 months, building a completely new marketing technology from scratch in a large organization can take years. And employing customized analytics solutions, new data layers or dashboards on top of the core tech creates additional work.
We can illustrate the different layers to create decision-making guidance in an onion (see image below). The deeper the layer in the “decision-intelligence onion,” the greater the disruption will be if you must fix, remove or replace a piece. Technology and data will have the greatest impact as they affect all layers on top.
This is not news to organizations. Because of the negative consequences companies avoid unnecessary data mitigation and technology system changes. For example, enterprise resource planning systems often have life cycles of 10-15 years.
In contrast to enterprise software, clients on average change their agency of record about every three years or only commit to project work. Since technology lifecycles outlast agency tenures, having committed to an agency’s data asset or technology could mean that you need to set up or fix your mar tech stack again when switching your agency. It seems therefore more reasonable to select independent data and technology that you can keep using when selecting a new agency.
The Decision-Intelligence Onion
Recommendations for clients and agencies
There are good reasons to not have your agency own technology and data assets. To avoid conflicts of interest and minimize disrupting the decision-intelligence onion, clients should seek independent technologies that can be used across agencies and easily integrated with many market platforms. Having flexible and agnostic data platforms is critical for data agility and implementing new data sources quickly.
What can we take away as strategic lessons or possible positioning for agencies?
The decision-intelligence onion has further implications. In line with the goal to protect their onion, brands are well advised to have as many of the technology- and data-related processes in house as possible.
However, in reality there will be limitations to what clients can accomplish themselves. They will always need external help with maintaining, fixing and leveraging their decision-intelligence onion. This is the customer pain point where ad agencies can shine, in particular for the latter: Given their experience in building brand experiences, ad agencies should focus on how to turn creative ideas into actions that leverage many different data and technologies.
Instead of trying to compete with tech unicorns and hoping for short-term revenues from conflicting investment decisions, agencies should go back to their roots and become masters of service excellence and creativity. They are best suited to understand how various technology, software and data pieces fit into the overall brand or campaign stories.