AI fever is everywhere. And, our industry wasn’t an exception. AI in the Adtech industry has gone far from basic data analysis and mining. We’re seeing increased adoption of AI and machine learning to do literally everything in the media buying process.
On a surficial level, Artificial Intelligence may seem like ‘a sci-fi movie’ where robots and machines beat humans to win the jobs (We’re being optimistic). But, people working in AI tech know the importance and researchers carefully note the progression AI bringing us every year.
On the other hand, with all the jargon and multitudes of middlemen, we couldn’t streamline the Adtech ecosystem. As the industry grows, the severity of calculation, bidding, etc, becomes more difficult.
Hence, we’ve invited AI into our industry. Since then, we’ve seen substantial growth in accuracy and scaling. Today, we’re gonna take a look at how AI has been used by some innovative companies to better the Adtech landscape.
Table of Contents
Maxus – Lucy
The media agency Maxus uses an AI called ‘Lucy’ to process and organize data (structured) for increasing the effectiveness and efficiency of the campaigns.
“AI is not going to rupture media agencies. Technology like this will be how we evolve,” said Gaines, Chief Planning Officer of Maxus.
For instance, Maxus utilized Lucy to segment the audience and understand the influence of media (TV, Social, etc.) have on them. This, in turn, helped an Asian Cosmetic Brand and a Hotel Chain to optimize and reach potential prospects effectively.
Rubicon – nToggle:
Rubicon bought nToggle, a startup that developed a platform to streamline bid requests last year. nToggle uses Machine Learning (ML) to weed out impressions that DSP would be better off without bidding.
The process depends on ‘traffic shaping’. In the ‘Header Bidding’ era, the volume of impressions a DSP is supposed to handle is too high. This increases the operating and infrastructure costs and makes the buying process less optimized.
Rubicon’s nToggle allows DSP to bid on the right impressions and facilitates its smooth operation. Quality impressions in a growing volume are what nToggle promises to buyers using Rubicon.
MediaMath – IBM’s Watson:
MediaMath has partnered with IBM Watson at the end of last year. When eMarketer interviewed Chris Victory, Vice President of MediaMath to learn about the goal of AI partnership with IBM, Chris stated that the focus is on three areas.
Decisioning – MediaMath is planning to infuse AI into decisions. As AI can consolidate and analyze millions of touchpoints and industry data, MediaMath looks to drive more optimized economic decisions through this partnership.
Ad Creative – Ad creatives aren’t flat anymore and interstitial creatives are on the standard rise. Chris said it’s the next generation ads where you can ask questions directly to the ad and it can answer back with a Call-to-Action*.
*Recently, Emirates Airlines used chatbots in banner ads and reportedly saw an 87 percent lift in engagement, per source.
Insights – Another area where AI can be leveraged is reporting and analytics. With more advanced researching and data mining capabilities, AI algorithms can be used to derive impactful insights for any campaigns.
Also read: We’re in a race with AI!
Blackwood Seven – se7en
Volkswagen (You’re right, the Car company) relied on its media agency ‘MediaCom’ until Blackwood Seven upped the numbers.
Blackwood Seven is a Danish media company, which uses AI to predict and Optimize digital ad spends. After using the algorithm, Volkswagen’s Lutz Kothe, Head of Marketing for Passenger Cars said, the campaign results were increased by up to 14% in a few months. She also reported a 20% difference between algorithms’ and agency’s results in some instances.
Every Car Model demanded a different strategy and the algorithm predicts the best medium and frames the strategy by analyzing and forecasting tons of data including Fuel Price, Competitor’s Price, Region, Car Registrations in the area, orders, and market data from Nielsen. As an AI algorithm, the Blackwood Seven’s strategies became effective with the increase in input data.
How Publishers can use AI to thrive in this Competitive Market?
AI wasn’t meant to help just the buy-side of the chain. Publishers can embed AI/ML to optimize their strategies and automate manual processes.
Reformatting:
In the publishing industry, there is no one-size-fits-all content. Everything has to be fine-tuned for the respective platforms and audience states (emotions). For instance, there will be a huge gap between Facebook and Reddit users and so, you can’t push a common message expecting conversions. Reformatting has been a problem for publishers until the Associated Press showed a way.
Associated Press (AP), a legendary publisher who produces 2000 digital stories everyday uses AI for reformatting purposes.
Auto-tagging:
Tagging is an important part of publishing an article. The tag decides the suggested content, Google indexing, targeted ads, and more. As you can’t replace this task, use machine learning to automate the task.
Here’s how The New York Times uses its own ‘Editor’ to auto-tag the contents.
Moderating:
‘Content Moderation’ has been a burden for publishers as they have to manually remove the spams and qualify human opinions. Before February 2017, NYT’s staff moderators had to examine around 11,000 comments posted to 10% of their open articles daily.
But now they’ve partnered with Jigsaw (Alphabet’s subsidiary) to weed out spams. Jigsaw’s perspective, a self-learning algorithm can compare the comments with thousands of ‘toxic’ comments to determine the veracity. It allocates a score and then, eliminates comments below the particular score.
Chatbots:
The arrival of surplus technologies has reduced our attention spans. Publishers need to keep the readers engaged as long as possible to understand and deliver the right data.
To make it the engagement more conversational, publishers have started to use chatbots which answers readers FAQs, suggests content, and connects the support team.
Predictive Analysis:
Artificial Intelligence is known for crunching millions of data points and structured data/reports. It helps marketers make informed decisions and develop successful strategies. On the other hand, AI can be used to drive monetization for publishers.
How? The New York Times used Machine Learning and data science to analyze, find, and market the most engaging content which ultimately driven their subscriptions up. Read here how the publisher did it in detail.
