Programmatically traded advertising has undoubtedly come under fire during 2017 particularly in the area of brand safety.
Naturally the issues garnered many headlines, but it’s worth remembering that brand safety concerns aren’t unique to programmatic. Unfortunate juxtapositions are nothing new – in the past it was down to eagle eyed newspaper editors or TV planners insuring brand messages were delivered in the right environment at the right time.
I recall an unfortunate TV spot error in the UK in the mid 2000’s when an advert for United Airlines ran during a Discovery documentary about United Airlines Flight 93, the flight on 9/11 that crashed in Pennsylvania.
What is new is the size and scale of today’s campaigns and available inventory (the scale of inventory being the reason for automating trading in the first place). This means that as well as sharp eyed humans, technology has to step into the guardianship role.
Whilst theoretically less fallible than humans, once an error does occur it can be harder to win back trust. Not least as the reasons humans screw up are easier to understand than the complex chain that causes the occasional automated error.
I’m not downplaying concerns. Undoubtedly new technologies bring challenges as well as significant opportunities.
However, despite what some recent programmatic-apocalypse headlines might suggest, programmatic trading isn’t going away. Media buying became automated as a necessity when the inventory-sphere exploded post digital. The scale is now simply too big and we aren’t putting the programmatic rabbit back in the hat.
In 2016, Nielsen research outlined that two-thirds (63%) of Australia’s digital advertising inventory across all devices is now bought through programmatic or ad network services. In the US the figure is even higher at 80 percent and is tipped to rise to 85 percent by 2019. And other mediums like TV and radio are beginning to trade that way; at Nine’s upfront earlier this month, it announced it was automating the buying and selling of its television inventory through its 9Galaxy platform.
Gonzales Fuentes, Global CEO Media & Digital Practice at Kantar Insights, summed up programmatic well in his keynote at IAB Australia’s MeasureUp conference last month when he said “Programmatic is the future, but very inconsistently delivered in the present.”
Yes there are some inconsistencies out there – but the current scrutiny should accelerate efforts to address.
From where I’m standing, there are already several reasons to feel optimistic about brand-safe programmatic trading in Australia.
Accelerated evolution
Programmatic is still in its relative infancy – and yet there have already been significant strides made to promote and foster brand safety.
There are numerous technologies that already exist that are built into many programmatic platforms. At the most basic level, vendors have blacklists that dictate where their ads must not run, the flip side of which is whitelists that limit impressions to specific brand-safe inventory. In addition, third-party providers like Integral Ad Science, DoubleVerify and White Ops work to ensure brand safe environments during bid requests.
Historically these tools haven’t been used by advertisers and vendors. In part this has been due to fears over frequency capping from buyers and poor communication from vendors. However, less reputable players in programmatic have actively avoided the scrutiny these tools bring.
That is now changing. Underused or misunderstood quality-control measures are becoming a thing of the past as buyers are now asking much more questions of vendors. Further, clients are demanding that third party verification providers are bought on board.
Ultimately, the scrutiny has resulted in much more collaboration and communication between agencies, brands and platforms to ensure filters and preferences are in place from the very beginning. And this type of collaboration will have long-term impact.
Machine Learning
I know from my own company that there is a lot of time, money and resources being devoted to developing highly sophisticated state-of the-art technology to further improve brand-safety.
Artificial intelligence (AI) and machine learning is playing a significant role in enabling developing technologies. The benefit of AI is that not only can it weed out today’s offensive or undesirable content at scale, but it can also look to prevent the threats of tomorrow (at the same time as more precisely targeting individual consumers with relevant creative).
When combined with the traditional methods outlined above – blacklists and whitelists – it should be able to safeguard brand perception and lead to increased confidence between buyers and sellers.
Premium is performing
It isn’t just technology like AI that is evolving to improve brand-safety – the programmatic marketplaces themselves are evolving. Globally, the share of deals made across open exchanges is declining, whilst we are seeing the rise of programmatic direct (automated trades between publisher and advertiser) and DMPs (private marketplaces where select inventory is offered to a select group of advertisers).
By their nature these tend to be premium brand-safe inventory that advertisers trust and are set to take an increasing share of all programmatic deals in the future. It’s a good example of how the programmatic space has – and will continue to evolve to adapt to client needs.
I certainly hope that this time next year we will be having a very different conversation around programmatic. However, this will require everyone in the programmatic chain to step up to the plate.
As long as there are companies willing to purchase questionable inventory to get cheap advertising and we don’t take steps to weed out the bad players who try and game the system with things like domain spoofing, the problem will persist.
We need the entire programmatic ecosystem, from brand to platform to media owner, to stop enabling scale at the expense of quality.
By Alex Khan, APAC MD of platforms at Oath