The New Era of Big Data and Digital Advertising
Advertisers have demanded a better ROI on their marketing budgets, and as a result, the industry has become a frontrunner in embracing technology to connect with the right audience.
Join the DZone community and get the full member experience.Join For Free
Traditional advertising has always relied on user data from a plethora of products and services. However, gone are the days of broadcasting ads to a wide audience on TV, Radio or Print Media, and hoping to convert a small % as customers. Advertisers have demanded a better ROI on their marketing budgets, and as a result, the industry has become a frontrunner in embracing technology to connect with the right audience.
Initial days of digital advertising justified the need for metrics presented through ad impressions, clicks and conversions to measure the performance of people who were using a computer browser. Not only has that changed in the last few years with rise of mobile and social media, but it has also made these very metrics irrelevant in today’s age.
According to a 2014 comScore report, mobile usage as a whole accounted for 60% of time spent, while desktop-based digital media consumption made up the remaining 40%. Google confirmed that since 2015 more searches take place on mobile devices than on computers in 10 countries including the US and Japan.
On one hand, the ability to access information anytime/anywhere has provided great opportunities to the ad tech industry; it has also made the situation more precarious on the other. With the explosion of the internet, has also risen the prevalence of malware and fraudulent bot traffic that can be used to generate any kind of ad impressions/clicks deemed necessary for success. As a result, $7.2 billion is expected to be lost globally due to non-human traffic in 2016 according to Association of National Advertisers.
Naturally, there are calls to overhaul digital advertising offerings to come up with new performance measurement techniques that suit current user behaviors. Companies have started looking into the vast amount of data being collected continuously for new opportunities. Needless to say, Big Data is reshaping the industry. To understand the impact, we need to take a closer look at the current state and needs of the online ecosystem.
- Multi-device users – Users are always online using multiple devices several times during a day. Having the ability to target the same users across those devices is valuable for advertisers. However, it is not possible for third party ad tech companies to cookie users on mobile apps, as they can on desktop browsers. This has lead to something called mobile device fingerprinting which essentially gathers information from multiple signals coming out of a phone/tablet and then runs it through an algorithm to create a unique fingerprint for every device. Such techniques require data accumulation from millions of devices, real-time decision-making based on the fingerprint and need for the data to be preserved for future refining of the inherent algorithms.
- Addressing real-time needs – Smartphones have made us impulsive. The industry is also evolving to target such audiences during these ‘Moments’ when we are more receptive. There is more importance being given to what a user is doing at a particular time of the day vs. who the person is demographically. Some good examples would be use of a phone to influence a purchase decision while in a store or searching your smartphone for a solution to an unexpected problem. These users expect relevant answers in that very moment and enterprises rely on their Big Data infrastructure to act on such situations happening several times during a day.
- Dynamic Ads – Big Data is helping to deliver a highly personalized online experience. The ability to show a unique ad to a particular user based on their preferences is on the rise. Advertisers are creating several versions of the same ad to target different audiences online. Ultimately, technology is moving towards delivering dynamic ads via real time analysis of user data. The Open RTB (protocol used to transact ads programmatically) specification on Native Ads has an inbuilt ability to configure different ad components on the fly during an ad transaction, which will lead to precise targeting and better performing ads eventually. Similar concepts are being used in Digital Out of Home (DOOH) advertising that trigger dynamic ads on billboards and other public places based on the audience.
- User data distribution – Today’s user is constantly online on various Apps, Web, and Social Media across multiple devices. The amount of user data being collected by different entities in the process is unfathomable. To address a user’s online needs across the ecosystem, it will require all of these parties to share data with each other and that too in anonymized way. Many companies have already starting sharing their invaluable insights with others at a premium price. However, most consumers are struggling to manage and use that data effectively. Although, we are still far from collecting all data sources and combining for universal access, there is an immediate need of NoSQL database expertise in the area.
- Fraudulent/Non Viewable Traffic – Along with non-human traffic there is an existence of web page areas delivering ads that are never seen by a human eye. These impressions are a total loss for advertisers who want the ability to detect bots and measure viewability every time before any ad is shown. The Interactive Advertising Bureau has been tackling these issue head-on, which has given birth to new companies that can provide viewability and non-human traffic details after the fact. The industry will be required to start transacting on these dimensions fairly soon. However, the supporting technologies still need to standardize ways to absorb/analyze thousands of ad requests every second without failure before making a correct decision.
As we can see, Big Data is primed to have a major impact across the entire Digital Advertising universe. Still, leveraging the power of Big Data effectively requires the expertise in next-generation technologies such as scale-out databases (key-value datastores, NoSQL databases, and many others) as well as a robust data management infrastructure that can be easily managed, and can protect data in real-time.
Published at DZone with permission of Jeannie Liou, DZone MVB. See the original article here.
Opinions expressed by DZone contributors are their own.