Telcos in APAC Are Betting on AI to Gain a Competitive Edge

Compelling use cases are encouraging telecommunications companies in Asia Pacific (APAC) to invest in, adopt and implement artificial intelligence (AI) solutions at a faster pace than other sectors.

The telecom industry is investing heavily in AI. Investment in this sector is expected to touch US$2.5 billion by 2022, at a compounded annual growth rate (CAGR) of 46.8 percent between 2017 and 2022, and APAC is expected to show the highest CAGR during this period.

A recent study, Artificial Intelligence is Critical to Accelerate Digital Transformation in Asia Pacific, published by Forrester in partnership with Appier, points to the high adoption of AI in the IT and telecom industry in the region. Sixty-four percent of the respondents from the telecom sector say that they are already implementing AI solutions within their businesses. This adoption rate is well ahead of other sectors in the region.

The Impact of AI on Functions Across Enterprises

AI today is impacting every function within an enterprise, from research and product design to sales and customer service.

In one study, 32 percent of decision-makers surveyed believed that by 2020, AI’s greatest competitive impact will be on sales, marketing or customer service. The 2017 Boston Consulting Group (BCG) – MIT Sloan Management Review study “Reshaping Business With Artificial Intelligence” states that customer-facing activities, such as marketing automation, customer service, and support services, will be most impacted by AI over the next five years.

AI has a lot to offer the telecom industry in particular, across different aspects of the business. Telecom operators have access to mind-bogglingly vast amounts of customer and network data. Using AI, they can make sense of this data at a granular level and access rich insights to improve customer service, boost operational efficiency and introduce new revenue streams.

As the Forrester study finds, the telecom sector in APAC aims to use AI in innovative ways in order to stay competitive and flexible, so they can better anticipate and respond to market changes. Additionally, telcos aim to speed up the generation of customer insights and improve existing products and services.

Using AI to Enhance Customer Service and Improve Marketing

Undoubtedly, the fastest growing and most visible use case for AI in this industry is customer service. Telecom companies around the world are using automated chatbots to streamline customer operations and applying machine learning algorithms to make customer service more seamless and cost-efficient.

For example, Ask Spectrum – Spectrum’s AI-powered virtual assistant – can help customers with their questions around their account or Spectrum’s services, and can even help troubleshoot issues. In 2017, Vodafone launched its chatbot TOBi, which simulates human conversation, and live chats with customers to respond to their questions. Closer home, Indonesian telco Telkomsel now only employs chatbots to answer most customer queries.

AI can also be immensely valuable in sales and marketing functions, by creating a personalized user experience and boosting customer engagement and retention. Telcos can use AI tools to analyze users’ preferences, interest areas and past purchase patterns, profile subscribers, and map these onto typical conversion rates to make personalized and relevant recommendations. AI-powered insights can help them recommend the appropriate packages to users, and push these to them at the right time – for example, marketing an add-on package when customer data shows that the subscriber is close to using up their data quota for the month.

Applying AI in Operations

However, the customer angle is only one half of the story. AI can also find application in network optimization and predictive maintenance. AI-powered algorithms monitor network data to detect problems as they occur in real time, and help in quick and efficient resolution. AI can enable a network to self-heal, self-optimize and self-learn, and make decisions autonomously. This ability to detect and solve potential issues before the customer is even aware of them, leads to proactive and enhanced customer service. Additionally, telcos can offer better service delivery through better network performance and improved reliability.

Take the example of SK Telecom, considered a pioneer in the use of AI through the implementation of Tango, its AI-powered network operation system. Used across all of SK Telecom’s telecommunication networks, Tango uses machine learning to analyze network traffic information and optimize network operation.

There are various such examples of how telecom companies all around the world are successfully using AI and machine learning within their operations, but AI implementation in this sector is not without its challenges.

Challenges to the Rapid Adoption of AI

Different studies, including the Forrester report, point to the prevalent barriers to AI adoption across enterprises, including those in the telecom sector. These include:

  • Difficulty in attracting and developing skilled resources that can implement cutting-edge AI solutions and transform business
  • Difficulty in building a skilled and agile cross-functional team to work with AI solutions
  • Security concerns around the adoption of AI solutions
  • Internal resistance to change within the organization, especially given the popular notion that AI replaces people
  • Lack of leadership support for AI initiatives, and competing priorities for investment

With AI offering unprecedented opportunities to obtain effective insights into customers and operations, telecom companies in the region are focusing on finding a way around these obstacles in order to accelerate their digital transformation and stay ahead of the competition.

 

* For more insights into AI adoption across different industries, read our industry-focused infographic and the full Forrester study.

Gain Quality Leads That Won’t Break Your Budget With Artificial Intelligence

Cost per lead (CPL) has long been considered the gold standard in marketing – it’s hardly surprising, as 63 percent of businesses cite generating traffic and leads as their biggest marketing challenge.

However, while CPL can be a reliable metric, it does have one rather large omission: it doesn’t consider the quality of the leads. A low CPL is desirable, but it shouldn’t come at the cost of high-quality leads – this will result in a low conversion rate, and fewer sales for the brand. No one wants that.

The good news is that a low CPL doesn’t have to mean low-quality leads. By harnessing the power of artificial intelligence (AI), marketers can find high-quality leads with a high conversion rate, and at the same time bring down the CPL.

Pick Out the High Performers: Identifying and Targeting Lookalike Audiences

An AI-enabled platform can identify which devices customers use, and build a behavior profile based on their cross-screen activities. Once you know who your highest performing audience groups are (i.e. those most likely to convert browsing into purchases) by demographics and interests, the platform can identify other users who share those attributes. This is called finding lookalike audiences.

Brands can therefore identify and buy access to the best audience for their campaigns, expanding their reach beyond the traditional domain.

Maybe your best audience is the one who viewed 10 products over the course of three days, as opposed to those that viewed five products in a single day, for example. The traditional marketing approach would be to target each of these groups in turn and hope for the best. However, AI removes the guesswork by allowing you to focus on those highest performing customers for maximum return.

Remarketing to reduce delays in sales

Artificial intelligence also helps with remarketing (reaching potential customers who have expressed interest in a product but not yet converted). By reaching the user on every screen they use, you will shorten the time between the initial visit and the actual purchase.

For example, if a user browses a product on his laptop and later on his smartphone, an advanced system can remarket to him via both platforms, through email for the laptop and by an app notification for the smartphone, say. This makes it much easier for the consumer to purchase, as it only involves a couple of clicks, rather than them switching devices and/or from an email program to a web browser.

A Person, Not a Device: The Importance of a Single Customer View

An AI-enabled platform lets you tailor recommendations and advertising creative to each user’s unique cross-screen behavior and browsing history (aka the Single Customer View). Because it recognizes these activities as those of an individual, and not just a device, it builds a picture of them as a person, with certain interests, behaviors and habits. Not only does this increase the chances of a higher conversion rate, it also makes it easier to find lookalike audiences, because you know more specifically what you are looking for.

For instance, you are interested in the cross-screen web browsing history of a male consumer named A. Using traditional marketing techniques, you would only see that content being consumed on different screens is about sports, technology, finance and travel. Multiple touchpoints make it hard to tell how many consumers are using these devices, and these topics are also too vague to really build a picture of what he is interested in.

By analyzing the cross-screen behaviors and the keywords within the online content A consumes, AI can link all the devices owned by him, and determine that he is searching for basketball with virtual reality, bitcoin, budget bed and breakfasts while travelling, and so on. Immediately you have a much more vivid picture of who A is and where his interests lie, which enables you to tailor your marketing materials to him as an individual. Crucially, you can start a dialogue with A – and with those who share his interests – which will build a relationship between him and your brand.

Put a Limit on It: Frequency Capping and the Power of Saying No

Of course, with all this power at your fingertips, it’s tempting to bombard the consumer with marketing messages, but that risks overloading them, which will only serve to alienate them from your campaign. It will also cost you more.

Instead, smart AI algorithms employ frequency capping to ensure you don’t overwhelm your customers. It also means you spend your budget efficiently and limit wasted impressions. As it works across every screen your user owns, it knows they are the same person instead of assuming a new individual for each device. That way, you won’t send them mixed marketing messages.

Capping is available per day or per action during the life cycle of a campaign. AI can find the best rule depending on your preferences. You can also limit it to a total number of impressions or clicks.

Taking global beauty and skincare brand Estée Lauder as an example, it employed the above techniques to huge success. It used Appier’s CrossX Lookalike feature to identify new, high-value, young audiences from the profiles in the CrossX database with data collected from over 3,000 campaigns run by Appier. Estée Lauder increased its number of leads by 167 percent, while reducing its CPL by 63 percent. It also shortened the time to conversion among valuable users who were interested in the brand by using CrossX’s remarketing and frequency capping tools.

AI is an immensely powerful tool for marketers looking to increase the quality of their leads, while decreasing their cost per lead. It shows that when it comes to marketing leads, low cost can mean high quality.

How to Convert First-Time Visitors Using AI-Driven Personalization

It’s a norm for marketers these days to drive meaningful engagement with returning customers through personalized marketing campaigns and content. However, with marketing automation tools powered by artificial intelligence (AI), you can now personalize your website content for first-time visitors to turn more traffic into conversions.

In today’s noisy digital space, brands place high premium on personalizing their content and services to their audiences. Tailoring the content on an app or a website to match visitors’ interest can be a powerful way to convert them into leads or even buyers.

According to the 2018 Personalization Pulse Check from Accenture Interactive, 91 percent of consumers are more likely to purchase from brands that provide relevant offers and recommendations.

How Far Can Traditional Methods Go?

Marketers have so far approached personalization by researching the existing customers and refining target personas to create more tailored content over time. As you get more data on the type of visitors who return to your site or app, you get a better understanding of the creative that should be produced for these visitors.

While this helps in delivering personalized experiences for returning customers, a major portion of visitors convert on their first visits. According to a recent research, 84 percent of conversions happened during visitors’ first-time visit. This means you might be missing out by not personalizing experiences on you site to new visitors.

There are some visible attributes of new visitors you can use to personalize your site’s user experience. Traditionally, data points like IP address, location, device used and source of traffic can provide marketers some insight into what new visitors expect to see.

Geolocation filters, for example, can help clothing brands personalize their online stores to show season-themed clothing in local languages based on first-time visitors’ IP addresses. Similar first-party data points can also help surface promos, time-sensitive discounts and freebies on popular purchases from your store exclusively for first-time visitors.

How AI-Driven Personalization Can Do Better

However, these traditional methods still have limitations as they are based on information collected from user behavior within your website or app. They are also incapable of distinguishing behaviors from the same user across devices, thus losing out on opportunities for device-specific personalization.

Today, with the help of AI, you are able to understand consumers interests and behavior outside of your own online channels, which means even before visitors make their first visit to your site. So, you will have a better chance to personalize experiences accordingly.

  • Cross-device personalization

Despite the explosion of marketing automation tools in the last few years, marketers still struggle to track consumers across touchpoints like devices and channels. For example, a new visitor may browse a fashion retailer’s site on her tablet on Sunday afternoon, bookmark her favourite dress on her PC the same night and decide to check out only on her mobile later. These three interactions would traditionally be attributed to three separate users.

With a proactive marketing automation tool powered by AI, you are able to map the user journey and gain a single customer view across all possible screens, which enables you to deliver relevant messaging at the right time, such as a push notification remind her to check out on her way to work on Monday morning.

  • More accurate profiling based on interests from third-party data

In addition to this first-party data, marketers can also benefit from leveraging third-party data about a potential future visitor’s behavior and preferences on external sites. While some platforms provide syndicated online consumer data, they are usually collected from the behavior of logged-in users or through surveys conducted by a panel of selected users.

What is more powerful is an AI-based tool that can analyze billions of data points to identify patterns from across devices, to model how users behave and move between different screens. For instance, Appier’s AIQUA manages to do just this by analyzing more than 2 billion data points from its thousands of campaigns to identify user interests and keywords. You can layer such data on top of first-party data to segment audiences more precisely and personalize content for each segment on their first visit.

More importantly, such AI-powered tools go one step further and analyze the keywords within the offsite content consumed to reveal more specific interests like “basketball” and “FIFA”, instead of “sports”, or “bitcoin” and “virtual reality”, rather than “technology”.

Considering a first-time visitor to your travel site might have varying intents based on the articles he read on external sites – mere exploration, planning a safari in Kenya or a wine tour in Tuscany. Based on his interests, you can categorize him into “outdoor adventurist” or “wine lover”, and tailor the web content or offers accordingly when he first lands on your site.

While you can’t always expect your online or app visitors to convert on their first visits, it doesn’t mean you can’t increase the chances of this happening by using AI-power marketing automation tools to deliver a hyper-personalized experience.

Accelerate Digital Transformation With Artificial Intelligence in APAC

As customer expectations continue to intensify in an increasingly competitive environment, companies can stay ahead of the curve only by adopting the latest and best that digitalization has to offer. For those frenziedly pursuing a digital transformation agenda, artificial intelligence (AI) has the potential to positively impact the way they do business.

A new commissioned study conducted by Forrester Consulting on behalf of Appier shows that AI has a critical role to play in accelerating digital transformation in Asia Pacific (APAC). Companies are well aware of the powerful insights that AI can generate. More importantly, they are testing how they can use AI to transform their business models and enhance experience across the entire customer life cycle.

The June 2018 study surveyed 260 business and IT leaders across eight markets (Japan, South Korea, Singapore, Taiwan, China, India, Australia and Indonesia) who are directly responsible for technology purchasing decisions.

Covering the key industries of telecom, insurance, banking, IT and retail, the study threw up some interesting insights on how companies perceive AI’s role within their businesses.

Some of the key benefits that companies are expecting by using AI tools are:

  • Easily discover relevant prospects, and boost chances of conversion
  • Improve customer interaction using deep insights into customer behavior
  • Maximize customer value, offer an enhanced user experience and improve customer loyalty
  • Optimize the marketing mix to improve ROI
  • Personalize digital experiences

Streamline Operations and Offer an Enhanced Customer Experience With AI

In a customer-obsessed world, 71 percent of respondents were surprisingly clear that AI would primarily help them improve efficiency of operations, while 59 percent spoke of improved scalability.

Respondents also expect that AI will improve the digital customer experience by helping companies deliver smarter and more personalized service. More than half of them expect that AI will help better predict customer behavior better and 45 percent are convinced of the deep consumer insights they can gain.

Investment in AI is Still Nascent

Yet, the study found that when it comes to investing in AI technologies, companies in the region are still in the nascent stage. Most respondents are investing in data management technologies and platforms in a bid to first build a strong base for future digital growth and their AI journeys.

For instance, 61 percent of respondents are investing in enhancing their data security and privacy capabilities while 57 percent are investing in data cataloguing solutions for analytics. Only around half of the surveyed companies are investing in AI solutions that improve customer view across digital channels and even less are using insights to build data-driven AI applications.

The reason for this seems to be the challenges APAC firms face around gathering and integrating big data, as well as building the right predictive analytics platform for their needs. While these challenges demand that firms invest in data management technologies, an insufficient focus on core-AI technologies such as computer vision and natural language processing could cause AI adoption to slow down.

Prioritizing Business Objectives to Build Relevant AI Journeys

Investment may be slow in coming but companies are already building their business plans based on what they hope to achieve using AI. The IT and telecom industry, for example, are prioritizing AI solutions that will allow them to predict market changes better while the FSI segment is looking to improve the accuracy of customer behavior prediction. Retail firms want to use insights gathered through AI to develop new products and services.

The Forrester study points to the immense potential that AI holds for APAC companies, which can use this technology to address core business objectives of streamlining operations and enhancing the customer experience.

Companies in the region are also expected to leverage AI to deliver business value at every stage of the customer lifecycle, from prospecting to upselling and retaining loyal customers.

For a more in-depth analysis of how companies in the region are using AI, download the complete study here.

Keep Your Audience Engaged With AI-Powered Push Notifications

Transform your app marketing with artificial Intelligence (AI), making your messaging and notification strategy personalized, relevant and customer-centric.

Did you know that around 52 percent of app users find push notifications annoying? An unpleasant experience around app messaging can easily result in a user opting out of notifications they find interruptive. Worse – they could even uninstall your app, especially if they find the notifications completely irrelevant.

The solution does not lie in completely doing away with notifications. When used wisely, these are an important element of your brand’s messaging strategy. Notifications help keep your app top-of-mind; they encourage users to open and re-engage with the app. Ultimately, a well thought through notification strategy encourages app usage, growing the lifetime value of the customer.

Effective Notifications Are Personalized, Relevant and Timely

In a world where 35 percent of notifications are generic broadcasts to all users, useful notifications are those that offer real value to customers, and are personalized, relevant and timely.

Netflix is often cited as an example of a brand that has nailed successful messaging through notifications. It only sends out notifications announcing the launch of a new season of a show that you follow, or recommending a newly available movie that is similar to others you have watched. Evidently, Netflix’ recommendations are personalized to your preferences.

To make notifications relevant, marketers must use the swathes of data that they have around user behavior, interest areas and past purchases to offer information that is actually valuable to the user. AI can enable this kind of insight.  

Using AI to Understand Each User at a Granular Level

AI is making it easier than ever for marketers to personalize user experiences, and encourage engagement and retention. It enables proactive recommendations or notifications on products, services or features that are aligned with user interests, pushing up the likelihood that they will engage with the app and complete the conversion KPI.

AI helps personalize messaging through:

Segmentation

Effective personalization starts with accurate user segmentation. AI tools can help you minutely segment your audience by enabling you to:

  • Analyze data around in-app user behavior, past purchase history, action on push notifications, etc.
  • Learn about individual preferences and interest areas
  • Detect patterns, and
  • Predict future behavior

Relevance

Use AI to also ensure the relevance of the notifications. AI tools can analyze vast amounts of user data around preferences and interest areas, and use this to recommend content that they would most likely to engage with. Seventy-five percent of what Netflix users consume is a result of recommendations that are triggered in this way, and 35 percent of Amazon’s revenues come from recommended purchases.

Solutions like Appier’s AIQUA, an AI-powered marketing automation tool, allow brands to hyper-personalize their messaging by analyzing the user’s in-app behavior and journey, and mapping this onto their interests outside the brand’s platforms. Insights from onsite-offsite user interest mapping enable a single customer view, facilitating a better user experience through relevant and personalized messaging.

Timeliness and frequency

How many notifications should a brand send out? Studies have shown that 37 percent of respondents would disable push notifications if an app sent between two to five notifications a week, while 22.3 percent of them would stop using the app.

This means that no matter how personalized or relevant your notifications are, if they are too many in number, chances are that your users will perceive them as disruptive and screen them out. Hence alongside relevance, marketers must also consider frequency and timeliness of notifications.

Here again, AI has a role to play – by analyzing data patterns around when users engage with your app, for example. This will help you send out notifications at the optimal time, when users are seen to be most responsive and more likely to take the looked-for action. For instance, marketers who use AIQUA not only use the platform to personalize notifications, but also to identify the right moment and right way of reaching each user.

Thanks to the power of AI, marketers today can ensure that messages and campaigns are based not on guesswork or intuition, but hard data. The kind of proactive personalization described here is just one way in which you can use AI to enhance your messaging strategy. When it comes to leveraging AI in marketing, the possibilities are limitless.

Proactive Ad Fraud Prevention With Artificial Intelligence

As marketers grapple with the problem of ad fraud and its mounting losses, artificial intelligence (AI) is proving to be an effective weapon that can reverse the tide.

Marketers in Asia Pacific continue to throw money at advertising, as ad spending is expected to increase 10.7 percent to US$210.43 billion in 2018, according to eMarketer. However, the ever-growing problem of ad fraud is skewing their reporting and standing in their way of showing better returns.

Even mobile marketers who expected more safety with app installs, faced 30 percent more fraud during the first quarter of 2018 compared to the same period last year, according to AppsFlyer’s “The State of Mobile Fraud: Q1 2018” study. Mobile app marketers were exposed to US$700-US$800 million in ad fraud losses worldwide. What makes ad fraud such a challenging problem today?

More Sophisticated Ad Fraud Methods Today

In the early days of ad fraud, the methods adopted by fraudsters were relatively simple. They used bots focused on driving large volumes of traffic to websites, bought cheap traffic through auto redirects or employed people to install apps in click farms. Once a click was made or an app was installed, their job was done. However, this was soon caught by advertisers and their focus began shifting to examining post-install quality, engagement, last-click attribution and return on investment (ROI).

Today, ad fraud has evolved into a completely different beast. For example, in a technique called ‘click injection’, they try to steal credit for app installs by triggering a click right before these apps get installed. Or, they stack up ads on top of each other to generate impressions for multiple ads.

Fraudsters also realize that to remain undetected, they not only need to drive traffic to websites or generate large amount of app installs, but also remain engaged subsequently and mimic human behavior. They even employ human workers in click-farms to imitate real human interactions, develop apps that hijack devices to generate additional clicks and create simulators that generate fake installs from bot networks.

Brands Turn to Technology for Help

Big brands and major publishers have begun to act against ad fraud. The Guardian recently collaborated with Google and MightyHive, a programmatic solution provider, to investigate programmatic fraud. Adobe Advertising Cloud has partnered with cybersecurity firm White Ops to tackle the problem in the streaming TV media.

Another technology that is gaining traction among marketers is blockchain, thanks to its ability to enforce decentralized monitoring and independent verification. However, there are still many challenges to be addressed, like handling the massive transaction volumes involved in real-time bidding and getting universal acceptance from everyone involved.

Joe Su, Chief Technology Officer and Co-founder of Appier, explains, “I think it will be a while before everyone in the supply chain – from media buyers to ad exchanges and publishers – cooperates to opt into a universal standard to make it feasible.”

Sophisticated Detection Powered by Machine Learning

Traditional approaches in fraud detection rely on simple rules created by humans through the measurement of three signals (metrics like conversion rate and click time) or dimensions at most. For example, IP blacklists to block suspicious traffic and filter out installs with low click-to-install-time (CTIT) have worked in the past.

However, the problem with these fixed rules is that they are pre-defined, as advertisers approach detection knowing what they are looking for. For example, one might decide to exclude app installs with CTIT below 10 seconds, due to the higher likelihood of bot-operated installs in these cases. But if such rules are fixed ahead of time, it’s only a matter of time before fraudsters figure out ways to circumvent them.

A more effective solution is to leverage technology that can keep pace with, and more importantly, stay ahead of today’s ad fraud techniques. That means marketers need to go beyond simple, fixed rule-based criteria, towards AI-powered solutions that are capable of learning new fraud patterns and refining the rules on their own.

As fraudsters employ new techniques that are capable of mimicking human behavior, these machine learning algorithms can help marketers look for fraudulent behavior not immediately evident to the human eye. This is especially critical in the case of app installs, where detecting fraudulent clicks and impressions before install becomes paramount.

“Looking for signals like suspiciously low time between clicks and installs or CTIT can indicate fraud, but by the time the install has occurred, it would be too late and an attribution for the install has already been counted,” said Su.

One foundational AI-approach, called the “tree-based model” works by analyzing a massive number of signals to achieve maximum coverage and accuracy in detecting outlier behavior. Consider the case of “the chameleon”, where fraudsters mimic legitimate publishers and generate installs at a later date, when the natural user retention is expected to drop.

Another scenario is that of an “inventory burst” where inventory count from a suspicious publisher spikes at a time when generally the in-app registration falls. As machine learning algorithms learn from gathered data over time, both these sophisticated ad patterns can be detected and fed back into the filters for improved detection in the future.

By detecting more cases of ad fraud, marketers can weed out poor quality traffic and measure their returns on advertising spend (ROAS) more accurately. In a study of 5.2 billion data points from mobile app campaigns in the region, Appier detected twice the number of suspicious installs using an AI-based approach and realized 4 percent more ROAS, compared to a traditional approach.

It won’t be long before fraudsters develop even more novel techniques to try and escape detection. Advertisers who are vigilant and proactive in preventing this with AI-based fraud detection will be in a better place than their competitors to reap the benefits of their investment.

Personalization at Scale With Artificial Intelligence

Every consumer is different. They have their own interests, preferences and concerns. Sending the same message to every one of your customers and prospects is unlikely to win their hearts. Instead, it will only see your efforts quickly ignored and leave a sour taste in their minds.  

For marketing to be effective in any industry, you need to find a way to speak to your audience on a personal level, and using personalized techniques backed by artificial intelligence (AI) is the way to go.

Marketers’ Missing Opportunities for Personalization

It is now a common practice to use marketing automation tools to reach a wider audience, but the most obvious mistake that marketers tend to make is simply ignoring the option of personalization. By monitoring how users interact with your site, you can get very clear signals of what they are looking for, which device they use and at what time of the day. Failing to engage them with personalized messages means you are missing out on a hot lead.

Other marketers begin personalization first interacting with the audience, but stop short of tailoring messages to the individual throughout the customer journey across devices. For instance, a visitor looked at a t-shirt on your e-commerce site using a smartphone, but he soon left before making any purchase. Later he used a tablet to search for a sweater due to the change of weather. Without knowing this change of behavior, you would continue to send him push notifications about the t-shirt on the phone, rather than the information that meets his real need.

Personalize Campaigns for a Wider Audience Based on Interest

Now, with AI-powered marketing automation tools, marketers can tailor messages on the individual level based on their interests and behavior patterns.

For example, Emma came to your travel site and saw two package deals: one to Tokyo and the other to Bangkok. She clicked on the Tokyo offer and found out the dates were unsuitable. So, she stopped reading and then went on to the Bangkok page where she spent much longer time reading about this destination. Often, this behavior is seen as Emma having equal interests in both locales based purely on the clicks. However, by taking a holistic view, AI is able to identify that she is more likely to go for the Bangkok package. Hence, you can send personalized messages relevant to her real interest.

While learning about your existing customers to delight them is a great step for many companies, you need to make it count – and that means being able to scale campaigns effectively.

Being able to scale campaigns by segmenting the audience based on user interest and behavior is imperative. Not only this way you can personalize content for more individuals who share the same preference, but also target only the people who have shown real interest in a product – perhaps they have checked items multiple times or left something in their cart – rather than just because they have downloaded your app.

An example of poor personalization would be a travel site that sends details of a sale on hotels in Tokyo to all its users, regardless of them all showing an interest in Japan or not. Instead, a more logical approach would be to send details of the various cheap deals to the users who have searched for specific destinations.

An extra step would be to offer deals on car rental or things to do in their destination city. Preferences like this can often get lost in the ether, with prospective customers searching on their phone, tablet and computer and on numerous sites. The average consumer in Asia owns three devices, after all. If this valuable data cannot be tracked, marketers would miss out on a way to improve their strategy.

With the right AI on your side, you can quickly find the signals that people are showing and triggers that will convert them into customers. Many businesses struggle as they grow because they can no longer spend as much time on each customer, but with AI software doing the hard work you can be much more efficient in this pursuit.

Marketing automation can help a business boost its marketing campaign to a certain degree, but without the data and nuances brought about by AI-powered tools, you are often shooting in the dark. Marketing is about getting the right message to the right people, and artificial intelligence is the most efficient way to find out what people want.

How Far Are We From Explainable Artificial Intelligence?

Artificial intelligence (AI) is heralding a revolution in how we interact with technology. Its capabilities have changed how we work, travel, play and live. But this is just the beginning.

The next step is explainable AI (XAI), a form of AI whose actions are more easily understood by humans. So how does it work? Why do we need it? How will it forever change the way industries – especially in marketing – function?

The Mystery of the Black Box: The Problem With Current AI

No one would deny that artificial intelligence produces amazing results. Computers that can not only process vast amounts of data in seconds, but also learn, decide and act on their own have turned many industries on their heads – according to PricewaterhouseCoopers, the market worth of AI is around US$15 trillion. However, in its current form, AI does have one major weakness: explanation.

Namely, it can’t explain its decisions and actions to humans. This is sometimes referred to as the “black box” in machine learning – for example, the calculations and decisions are carried out behind the scenes with no rationale given as to why the AI arrived at that decision.

Why is this a problem? It doesn’t engender trust in the AI, which in turn raises doubt about its actions. Explainable AI is expected to solve that.

How XAI Works

XAI is much more transparent. The human actors interacting with the AI are informed not only of what decisions it reached and actions it will take, but how it came to those conclusions based on the available data. It aims to do this while maintaining a high level of learning performance.

Current AI takes data into its machine learning process and produces a learned function, leaving the user with a number of questions such as: Why did it do that? Why didn’t it do something else? When will it succeed? And when fail? How can I trust it? And how do I correct an error?

By contrast, XAI uses a new machine learning process to produce an explainable model with an explainable interface. This should answer all the questions above.

This carries its own risks. Any decision made by an AI is only as good as the data used to make it. While XAI increases trust in the decision made, that trust could be misplaced if the data is unreliable.

Another problem is how well the AI explains its decisions. If it is not comprehensible to the user – who could be a lay person with no technical background – the explanation will be worthless. Solving this will involve scientists working with UI experts, along with complex work on the psychology of explanation.

Risk, Trust and Regulation: Why We Need XAI

In so-called “big ticket” decisions like military, finance, safety critical systems in autonomous vehicles and diagnostic decisions in healthcare, the risk factor is high. Hence it is crucial that the AI explains its decisions in order to boost trust and confidence in its ability. However, there are a host of benefits for businesses in other industries.

XAI can address pressures like regulation, as it will enable full transparency in case of an audit. It will encourage best practice and ethics by explaining why each decision is the right one morally, socially and financially. It will also reinforce confidence in the business, which will reassure shareholders.

It will also put businesses in a stronger position to foster innovation, as the more advanced the AI, the more capable it is in terms of innovative uses and new abilities. Interacting with AIs will soon be standard business practice in many industries, including marketing. Hence it is vital that users can do so comfortably and with confidence.

Experts think this will empower marketers, effectively turning AI into a co-worker rather than a tool.

“In order to trust AI, people need to know what the AI is doing,” says Hsuan-Tien Lin, Chief Data Scientist, Appier. “Much like how AlphaGo is showing us new insights on how to play the board game Go, explainable AI could show marketers new insights on how to conduct marketing. For instance, AI can reach the right audience at the right time now, but if future XAI can explain this decision to humans, it would help marketers understand their audience more deeply and plan for better marketing strategies.”

It could also usher in a new way of working, with marketers accepting or rejecting XAI’s explainable suggestions with reasons in order to help the AI learn. “Today, it is likely that many great suggestions are rejected because they are not explained, and so humans overlook their power,” says Min Sun, Chief AI Scientist, Appier. However, these days could soon be over…

The Defense Advanced Research Projects Agency is currently running an XAI program until 2021. The program is expected to enable “third-wave AI systems”, where machines can build underlying explanatory models to describe real-world phenomena based on their understanding of the context and operating environment. Other experts also predict XAI will become a reality within three to five years.

XAI is no doubt the next step for AI, improving trust, confidence and transparency. Businesses would be wise not to overlook its potential.

Enabling Dynamic Personalization Through an AI-powered Single Customer View

For marketers today, the holy grail is no longer just customer acquisition; rather, they are increasingly focusing on growing the lifetime value of the customer by improving customer engagement and retention. Artificial Intelligence (AI)-powered solutions can help achieve this through hyper-personalization and a seamless user experience across devices.

In an earlier blog, we discussed how personalized marketing could drive better conversion.  For online businesses, personalization forms the backbone of user experience and revenue growth. However, in a survey conducted by eMarketer, 91 percent of decision makers acknowledged that their companies needed to improve their personalization capabilities.

Many enterprises are unable to truly personalize their marketing campaigns simply because they lack the single customer view (SCV) that is essential for personalization success.

The Limitations of Marketing Automation Tools in Developing an SCV

Put very simply, SCV consolidates customer data from your different marketing and customer service channels in one place; thus, offering a complete and comprehensive view of your customer. This, in turn, can help you create better targeted and personalized messages that boost interest, engagement and conversion.

But implementing this is easier said than done. Across the region, companies face a number of challenges when it comes to effective personalization through SCV:

1. As the number of touchpoints grow, with customers browsing, evaluating and purchasing products across devices, companies must communicate with them across devices and channels as well. Often companies lack the technology needed to map the same customer across different devices, and this leads to an inconsistent customer experience.

For instance, a marketer may reach out to the same person with the same message three times on three separate devices, simply because their marketing automation tool did not recognise the customer as one person. In a different scenario, if you send a push notification through a marketing automation tool to remind a shopper of something that she has already purchased, this information would be irrelevant or even annoying.

2. Current marketing automation and data gathering tools (email marketing, google analytics, social listening, etc.) allow companies to access vast amounts of customer data, but with these acting in silos, the data is dispersed across different databases. The result – a fragmented customer view.  

3. Marketing teams are able to gather data about their customers’ behavior and journeys on the website and company app, but lack details about what the same customer do outside the company’s online platforms, leading to an incomplete picture of the customer.

Using AI to Build an SCV for Personalization Success

This is where artificial intelligence solutions can help. An AI-powered proactive marketing automation solution like Appier’s Aiqua, for instance, automatically links your audience across devices. To achieve this, the system requires a massive amount of user behavior data, which comes with Aiqua and a brand might not have in a short time. AI will then find the patterns through the data and link back to a specific type of users and the devices they own. This gives you a consolidated view of each customer’s activity and lets you engage seamlessly with them across devices.

Additionally, it maps their journey inside your platforms with their behavior and interests outside, offering you a comprehensive view of your customer’s complete online behavior.

With this data, you can hyper-personalize your marketing campaigns and engage with users across devices by sending out your messaging at the right moment on the channel or device that is right for each user.

With AI enabling marketers to understand and identify audiences based on interests outside of the company app and website, such solutions become integral to customer discovery as well – allowing you to decipher user preferences before they even engage with your site or app. Using the data consolidated from various channels, marketers can then personalize content for customers who have not even engaged with them as yet, making messaging relevant and specific.

For example, a marketer at a travel company can identify which users are likely to be interested in traveling to France even before they land on the website. They can reach out to these prospects with personalized messaging meant to arouse interest and also proactively set rules to personalize the content of the website on the prospect’s first visit.

AI-powered SCV hence allows you to hyper-personalize marketing and offer customers products that they are actually interested in, thus shortening the purchase cycle and driving conversion. Use SCV as a foundational platform for your cross-channel marketing efforts and leverage the insights derived from the data to target the right customer at the right time on the right channel.

Making 1:1 Personalization Possible With Artificial Intelligence

The way a company markets itself and its products or services can have a huge impact on its profits, brand strength and longevity. While a poorly planned and executed marketing campaign is unlikely to help a company greatly, using modern techniques such as personalization powered by artificial intelligence (AI) can push your campaigns to the next level.

Marketing Personalization Is A Necessity

While many brands have already started to personalize their messages using marketing automation tools, online marketing allows for much more specific targeting, almost on an individual basis. Potential customers leave breadcrumbs of their interests across the internet, through social media, shopping sites and search engines. This data can help you craft a better marketing campaign, one with tangible benefits.

Personalization at scale can lift revenue by 5 to 15 percent for companies in the retail, travel, entertainment, telecom and financial services sectors, according to McKinsey & Company.

There is no surprise to see that personalization is becoming a major consideration as McKinsey highlighted that more than 90 percent of retailers believe this is a top priority. However, only 15 percent of these companies are actually doing a good job at it.

Challenges With Marketing Automation Tools Today

One of the most important aspects of marketing is making sure you get your message to the right people. The more specific you can be with your target audience, the more likely you are to see a positive return on investment in your campaigns. There is little point telling a vegetarian about your new steakhouse, or advertising a luxury resort to backpackers.

A major limitation of many marketing automation tools today is the lack of available data. Whether it is because you are looking at a small data pool or that you cannot track a user over multiple devices, having holes in your data can make your personalization less accurate and therefore less effective.

For example, a user may access your site on their computer, phone and tablet. Many personalization tools don’t have the capacity to see this as one user and will instead report three different people looking at your site. This gives a disjointed view of the user and means you miss out on valuable information about the buyer’s journey.

Marketing automation tools do not always understand human actions or emotions either. They could fail to recognize that after someone has made a purchase, he or she will no longer want to buy the same item. This can often come down to a lack of data about a user’s behavior.

Another challenge businesses have is that they find it difficult to scale their marketing efforts and manage engagement across all their channels. One cause for this is the number of different tools they need to cover all the bases, and upgrading each of these can be a costly or impractical exercise. With different tools in place, there is often a lack of communication between what marketers are seeing, making it hard for them to act quickly and efficiently.

How Can AI Improve Personalized Marketing?

There are many ways that AI techniques can help brands improve their marketing campaigns. This is particularly true in terms of zooming in on one particular type of customers and scaling up for large projects.

Without any sort of AI, the information companies know about a visitor when they visit a website can be limited. However, the visitor may already have shown information about the type of person they are and what sort of product they are looking for. For instance, a consumer often purchases fashion goods from an ecommerce site A, but she also buys baby products from site B. Having this insight available for site A means the visitor can be exposed to relevant products and information to keep them engaged further.

AI can also help you engage potential customers once they leave your site. For instance, someone who has abandoned a cart on your site can receive a reminder about the product that they have left waiting, perhaps with personalized suggestions for other items if that one wasn’t quite right.

Another scenario could be knowing whom to send certain newsletters to. You may have a sale or promotion on products that might be of interest to some of your contacts. Sending one email campaign out to all your subscribers isn’t necessarily the best approach. So, knowing who has shown an interest and who would be likely to buy that product can help make that email marketing campaign more effective.

AI-powered personalized marketing is proving to be one of the most effective ways to maximize a marketing budget and it’s something that brands are increasingly making use of today.