Writing in a WARC Best Practice article, GroupM’s Brand Safety and Digital Risk Director for EMEA, Bethan Crockett, acknowledges that the only way to remove all risk is to buy directly, “where the surrounding content is known in advance and will not change for the entire duration that an ad is shown.”
Rarely, however, is the answer that simple. To buy directly is to “limit access to available inventory, audiences and tools”, Crockett argues. A “risk-based approach must be used.”
There is a diminishing scale of insight when moving from single to aggregated inventory sources and from direct to indirect buying models, Crockett notes. Yet, the risks of unsafe environments can be minimised by establishing a more transparent supply strategy.
“Every brand holds core values which inform a baseline approach to risk. However, most will offer multiple products or services, targeting different audiences, which will further inform risk tolerance level”, Crockett writes.
Across the three pillars of high, medium, and low risk tolerance, Crockett explores the philosophies behind each. For a high tolerance, “performance and results are [the] most important concern.” Meanwhile, a medium tolerance will emphasise a balance.
Unsurprisingly, a low tolerance philosophy will “only run with inventory sources that offer full controls.” Two fairly simple mechanisms can ensure baseline protections: blacklists and whitelists. Though they too have drawbacks.
“Blacklists provide reactive protection with no downside therefore should be used as default but must be continuously maintained for effectiveness. Whitelists provide proactive protection but must be scaled in alignment with marketing goals,” Crockett writes.
The author highlights international telco Vodafone’s development of a “Global Brand Safety Strategy,” the principles of which came to govern “engagement with all digital media and intermediary partners.”
As a result, across its entire ad buying operation, Vodafone began to use whitelists tailored to each market, a set of non-negotiable categories and keywords blocked on all activity, commitment from media partners, and the scrupulous use of verification technology.
Data sourced from WARC