DATAx presents: The transformative power of AI in marketing

As marketers get to grips with the vast potential afforded them via AI, we caught up with Prasanna Kumar, global ecommerce/data lead and president of the APAC regional client team at WPP, to discuss the ways AI is transforming his industry


AI is not only acting as a disruptive force on marketing departments the world over, but it is also completely transforming the way marketers approach their day-to-day roles, deal with clients and generate business. Its potential is huge, albeit tempered by a general lack of understanding or willingness to embrace the technology.

To get to the bottom of why the industry appears to have been slow on the uptake and to understand the potential of the technological revolution being shepherded by AI, we spoke with Prasanna Kumar, global ecommerce/data lead and president of the APAC regional client team at WPP. Kumar, who will also be among a stellar collection of speakers at DATAx Singapore this March, also discusses key trends he expects to come to the fore over the next 24 months, the importance of relevant content and consumer data accuracy, and what it is that makes Singapore such a unique city in which to conduct business.

DATAx: Why do you think AI adoption for media buying remains low and what hurdles does WPP face in respect to AI adoption?

Prasanna Kumar: When it comes to AI in the media buying world, it's certainly moving slowly. I wouldn't say it's surprising though, due to three reasons: A lack of understanding, cost structures and data accuracy.

Clients are still learning about AI through pilot programs or research. Marketers continue to consider AI as more of an innovation tactic, rather than a core part of their media-buying practices, with the main use so far being improved targeting for programmatic ads. As a ballpark figure, fewer than 20% of clients are either ready to or are already using AI for their media buying.

The industry doesn't yet have a stable cost structure for the technology. While some solutions are free, others ask for 10% of media spend and most vendors do not define their offerings correctly – it's being shown as a silver bullet, which is not correct and rather confusing.

The biggest hurdle, however, is the accuracy of consumer data. For many marketers, their data is not in good-enough shape to be fed to AI. Lots of clients are focusing first on improving their identification capabilities. Before marketers can even think about using AI, they have to make sure that the technology they use to match cookies or other IDs back to individual consumers is accurate – otherwise, AI is pointless.

DATAx: At DATAx Singapore, you'll be discussing how data analysis can drive digital personalization. Can you provide examples of how WPP has implemented these solutions and what more can be done to ensure society benefits from digital personalization?

PK: My favorite example is in skincare where we created an online skin testing platform to interact and provide customized solutions for beauty enthusiasts. We were able to identify and reach out to customers via machine-learning algorithms, as well as through their past actions in search, commerce and social media.

There are three areas in which we need to improvise to ensure society benefits from digital personalization, which are interactivity, scale via machines talking to machines and content relevancy.

In respect to interactivity, the smart use of augmented reality (AR), VR and IoT technology, rather than utilizing just static banners when comes to personalization, works brilliantly. And it should not be tactical. It is proven to work, should marketers know the code to create the required experiences. At WPP, we continue to educate our clients and strive to enhance interactivity via technology to grow a brand's business efficiently.

As an example of achieving scale via machines talking to machines, [m]Platform/Xaxis's turbine engine can do what was once unthinkable and is able to handle numerous queries per second, almost three billion messages per day, and more than 5TB of data stored with hundreds of active data partners.

Marketers also need to consider that, while personalization assures the right target and context, unless content is appealing, it will not be as effective in order to complete the acquisition. We cannot ignore the need of creativity in a world where machines and humans should work together, rather than against each other.

DATAx: What are the biggest differences you've noticed about your industry's attitude toward AI today, compared to when you first entered the field?

PK: Wow, we are talking about couple decades ago vs. today. AI was seen only in robotic movies back then. However, in the past half-decade, the inclination to try, test and learn from marketers has increased significantly.

Today, it's seemingly possible and it's just matter of time before AI will become part of 'core' advertising, rather than just part of a tactical innovation 'checklist'.

DATAx: What key trends, in respect to data science, do you see your industry embracing over the next two years?

PK: There are six important developments that are sure to impact the world of marketing in 2019–20:

– The world is quickly turning to virtual and augmented realities.

– Voice is becoming the preferred search-marketing tool.

­– AI-powered solutions are everywhere – from small businesses to large enterprises.

– Video views also lead internet traffic on commerce platforms.

– The onslaught of native advertising is pushing marketing to be more creative.

– Programmatic advertising moving into the mainstream.

DATAx: What can you tell us about the opportunities and risks digital channels offer organizations in respect to recognizing customers and personalizing experiences?

PK: Personalization comes with its own risks and failing to understand these could make it a bitter experience which even the world's largest digital companies have recently gone through.

In terms of aggressive personalization, there's a fine line between engaging with customers and annoying them. If one small interaction with a brand triggers a long series of personalized content, the customer will be hesitant to interact with your brand at all.

Customer data is critical to personalization, but marketers' efforts often backfire when they use information that a customer didn't willingly provide. In regard to unauthorized personalization, it's no wonder that a large number of consumers feel that personalization is less useful. Customers want to feel understood by brands, not spied on.

It's important to remember that some personal information is extremely sensitive. When it comes to insensitive personalization, just because you can leverage someone's information doesn't mean you should – especially if your marketing initiatives might irritate or offend them.

Personalization relies heavily on customer data – and more specifically, accurate customer data. If your database is full of outdated or incorrect data, your personalization efforts will backfire completely because of its inaccuracy. If you want to execute a successful personalized marketing campaign, data hygiene must be a priority.

DATAx: Finally, in your opinion, what makes Singapore stand out as a location to establish and build up a tech startup?

PK: There has always been interest in Singapore's tech and startup scene, especially when the city-state became the only Southeast Asian country to rank in Startup Genome's Top 20 Startup Ecosystems in World report in 2015.

It has steadily growing entrepreneurial activity: Over the last five years, Singapore's startup landscape has grown tremendously which can be attributed to a conducive business environment, ample government support and a strong entrepreneurial community.

According to a study by venture capital (VC) firm Monk's Hill Ventures and tech startup event Slush Singapore, in 2017, startup investments in Singapore reached $7.3bn, making up 45% of all deals in the region. Singapore has been steadily growing to this number, with venture funding alone growing from $136m to $1.37bn in just a span of five years (2012–17).

Singapore has seen the birth of five unicorns – Grab, Lazada, Razer, Sea and Trax currently valued at a collective $12bn. Being headquartered in Singapore gave them access to the talent, funding and connections that they needed as startups.

The city-state also offers tangible government support through Enterprise Singapore, part of the Singapore Government, which has been actively working with different partners to develop Singapore's startup landscape. This support could mean mentorship, funding and investments, and incubations, which are all critical to the growth of startups.

Singapore is ranked third globally on the ability of startup leaders to connect and form relationships with entrepreneurs in other countries. The strong startup community in Singapore allows founders to build relationships and partnerships, not only locally, but also internationally.

With an ever increasing number of deals and VC investment amounts in deep tech, the Singapore government has been focusing on attracting and building more startups in deep tech sectors like medtech, cleantech, fintech and future mobility, among others.

The Lion City is also one of the best countries for women entrepreneurs. In a study recently conducted, Singapore ranked as the fifth-best market globally for having strong supporting conditions and opportunities for women to thrive as entrepreneurs. Women made up 27.5% of total business owners in Singapore, which is among of the higher proportions in the total markets studied.

And while the numbers may seem paltry, the conditions on the ground are not. There is a growing, vibrant community of women entrepreneurs in Singapore that aims to inspire women to create the next big thing.

Whether you're a homegrown Singapore local, or an international entrepreneur searching for a friendly place to plant your startup roots, Singapore is one of the best places to consider.

Making sense of the confusing world of iot platforms small

Read next:

Making sense of the confusing world of IoT platforms