It’s hard to believe we are a little over 12 months away from the year 2020, the start of a new decade, a milestone year as we mark 20 years into the 21st century. With programmatic fully into adolescence, we expect to see the run-up to 2020 as a period to “Do Good, Better” (which happens to be one of our MediaMath values) as we learn our lessons from the GDPR, Cambridge Analytica, poor measurement and continual battles with fraud. From AI to identity, a few MediaMathers share their predictions for 2019 below.
“From ‘omnichannel’ to ‘big data’ to ‘artificial intelligence,’ the buzzwords of corporate strategists, CEOs and marketing executives are once again making their cyclical ~5-year shift. As ‘AI’ and its ingredients—machine learning, deep learning, facial recognition, natural language processing— become the words de rigueur of the next few years, we are going to continue to see Moore’s Law-like impacts of technological change.
What this means is that those companies who only talked the talk and did not walk the walk in terms of big data (and omnichannel) strategies are going to be left further behind at a quicker pace, as AI relies on big data. And exacerbating this phenomenon will be the dearth of talent in analytics, engineers, big data and AI, which will lead to further and faster divergence between flourishing and collapsing companies.” — Laura Carrier, VP, Data & Measurement Strategy
“There will be a surge in the use of deep learning techniques by ad-tech companies in all aspects of marketing: customer ad response prediction, audience lookalike modeling, dynamic creative optimization, probabilistic identity tracking, fraud detection and optimal bidding. Deep learning models have been shown to be vulnerable to imperceptible perturbations in data that dupe models into making wrong predictions or classifications. With the growing reliance on large datasets, AI systems will need to guard against such attacks on data, and the savviest advertisers will increasingly look into adversarial ML techniques to train models to be robust against such attacks.” — Prasad Chalasani, Chief Scientist
Whereas two to three years ago TV buyers were saying ‘talk to my digital buyers about connected TV,’ now both sides are saying we should be the ones to control this industry. There was about $10 billion dollars spent last year that’s forecasted to grow, and the industry hasn’t even reached full potential. We need to stop saying that this is the year of connected TV because we are already there. But we’re at a point where buyers are saying I want to be buying into connected TV and buying into addressable TV not as a value-add to their linear buy but as a distinct channel.” — Mike Fisher, VP/Head of Advanced TV & Video
“Mobile is not a channel anymore, but the channel of the other channels. Mobile is becoming the centerpiece of overall advertising spend, building the bridge between online and offline through location data ( DOOH, TV, audio, desktop, mobile). We expect a big consolidation of location data providers—they’ll partner with DSPs, TV and DOOH solutions to win market share. We also expect location data improvement of quality vs. quantity as the stakes for accuracy become high with attribution and we dig into the correlation between visits and sales. From a vertical perspective, location attribution will become especially interesting for CPGs, which don’t always have access to sales data, and for retailers as an additional down-funnel KPI for them to drive users to in-store purchase.” — Floriana Nicastro, Director, Mobile Product Solution
“We’ll see continued/increased scrutiny of major first-party data players’ privacy, security, anti-fraud, brand safety and election-related practices, with expansion of that scrutiny to third-party companies, and improved industry standards to proactively address these concerns, along with emerging technologies such as connected TV and IoT.” — Alice Lincoln, Vice President of Data Policy & Governance
“I expect browsers to continue attempting to represent consumer interest via privacy-restricted browsers that block third-party advertising, This tips the scales further in the direction of walled gardens, which can operate on the first party on self-contained advertising stacks, which makes it hard for advertisers to reach their target markets (without using three different siloed ad tech stacks), harder for the free and open Internet to continue operating on ad revenue and specifically hard for independent publishers to continue providing content (of ANY quality) back to consumers without charging subscription fees. This means 80 percent of browser use will default to blocking tracking and thus interest-based advertising.
I see the advertising industry finding ways around the blocking, technically, and I see privacy advocates litigating bad actors into submission. Ultimately, the way we continue to do business is by engaging industry and consumers with consent, to continue delivering free and open Internet, funded by advertising and by commerce. This is a global thing—it’ll be there next year, following Europe by just 12 months.” — John Slocum, VP, DMP
“Consent management companies, customer data platforms and data management platforms will continue to become critical components for managing the interaction of consumer preferences, personal information and pseudo-anonymous online identities. With DoubleClick’s decision this year to stop populating the encrypted UserID field in DCM and DBM for consumers globally, marketers wanting to maximize campaign effectiveness and achieve maximum relevance for consumers must now explore alternative solutions in earnest.” — Ellie Windle, VP, Global Business Development