How Machine Learning Unlocks Marketing Performance on the OtherLevels Platform

By Jacek Serafinski

At OtherLevels, our ultimate goal is to help our clients broaden their ‘customer reach’ by engaging customers across multiple communication channels through second-generation digital messaging, making sure the reach and outcome of every marketing campaign launched through meets their expectations.

In most cases, there are many different drivers behind the success of a given campaign. Finding and understanding those drivers is key to meeting expectations and improving our clients’ outcomes – and the more data we collect, the more clear and accurate our analysis of the relationship between our clients and their customers can be.

Through data-driven intelligence, we can see how fluid and dynamic these brand-consumer relationships really are across various devices, channels, message types, and more. The fact that we expose users to second-generation digital messaging – reaching far beyond the standard email campaigning to leverage push notifications through browser, in-app messages or rich inbox communication – also impacts the dynamic interactions we see in our customers’ data.

Armed with that data, OtherLevels serves clients through the power of machine learning. In recent years, machine learning has proven its usefulness in business areas where the outcomes of the decisions of users in the past can be leveraged to support decisions in the future: Usage of automated machine learning models and algorithms allows for the discovery of patterns and interconnections between certain parameters of the decision and the outcome.

To maximise client outcomes, OtherLevels uses machine learning based on users’ interaction with multiple channels – taking into account various parameters such as time of message sent, channel per customer, average response time and many more – to drive predictive analytics: Pre-calculated prediction models that allow us to support marketers’ decisions about the specifics of campaigns in near real-time during the selection of specific parameters.

Essentially, the predictive models allow us to learn behavioural patterns in users’ interaction with marketing campaigns while the usage of machine learning techniques allows the models to maintain high accuracy over time as behaviours change.

These predictions are very subjective to individual characteristics of the audience of our customers – which, of course, are fluid and can vary across our different clients. But as we collect and amass more data, the quality of that data – and its power in predictive analytics –amasses as well, helping us support the definition of our clients’ campaigns with efficiency in order to ultimately minimise marketers’ workload.

As early adopters of second-generation digital messaging, our clients are among the first to realize the usefulness, value, and impact of new marketing communications channels before they become mainstream. At OtherLevels, we’re thrilled to be helping our clients unlock stronger marketing outcomes through the power of machine learning-backed messaging.

Filed Under: Tags: , , , With: Engagement, Marketing Techniques

Using Machine Learning to Drive Stronger Marketing Outcomes

By Brendan O'Kane

Much like artificial intelligence, “machine learning” is a tech-industry hot topic seeing increased application in B2B and B2C marketing. Though a complex concept, machine learning is largely about using technology to analyze data for patterns, then automatically create and deploy rules (based on those patterns) to solve problems or improve processes.

For marketers looking to power more impactful outcomes for their companies’ bottom lines, machine learning is key asset to look for from a digital messaging vendor. With the right partner, marketers can use machine learning functionality to take their campaign efforts a step beyond personalization to the “holy trinity” of individualization: The right message, on the right time, on the right channel.

Pattern Recognition Creates Personalization Power

Machine learning provides value beginning with pattern recognition. When dealing with massive volumes of users (both known and anonymous), it’s impossible for marketers to compile and process such huge amounts of data coming from multiple sources – ranging from apps and mobile websites to purchase behavior and offer redemption.

As such, it’s historically been a challenge for marketers to predict what campaign strategies will be most effective for each individual customer. Yet as personalization has grown increasingly important to the success of email marketing and other digital messaging strategies, machine learning has risen up to help them analyze historical campaign data to deliver highly targeted marketing offers.

In addition, machine learning helps marketers in other key areas, including:

Customer segmentation – By finding patterns in large pools of data, machine learning can help marketers segment their messaging campaigns down to the smallest, most homogenous groups of customers displaying similar behaviors and preferences.

Channel selection – Rather than blast all customers via email or another single channel, machine learning can automatically enable marketers to target users on the channel the individual end customer is most likely to engage with (based on past behavior).

Customer churn prediction – Machine learning helps prevent churn by recognizing the drop-off habits of customers who churned in the past, thus deploying rules to lessen the likelihood of brands losing those users who are at a high risk of churning.

Message timing – Timeliness is an underappreciated key to marketing effectiveness.

Customer lifetime value forecasting – Machine learning can help marketers predict and capitalize on the customer lifetime value (LTV) of existing customers to cultivate higher conversions from brands’ most valuable customers over time.

With a digital messaging strategy powered by machine learning, companies across many sectors including igaming, wagering, travel, and hospitality can see much stronger retention and conversion rates that drive meaningful outcomes on their bottom-line performance.

Contact OtherLevels today to discuss how we can drive stronger digital marketing outcomes for your brand.

Filed Under: Tags: , , , With: Marketing Techniques, Personalization

Mobile Commerce: Capitalizing on the ‘Multi-Device Path to Purchase’

By Brendan O'Kane


With consumers now spending an average of 174 minutes per day on their mobile phones, one BI Intelligence report projects that by 2020, mobile commerce will make up 45 percent of total e-commerce – totaling $284 billion in sales.

Yet that big number is a huge three times more than what’s projected for 2016: BI Intelligence predicts mobile commerce will hit just 20.6 percent of overall e-commerce, or $79 billion, this year.

With so much mobile activity in play, why isn’t “m-commerce” moving faster toward 2020-level totals? Because many mobile users take a multi-device path to purchase: 1 in 3 shoppers does research on a mobile before buying on desktop or offline (1 in 2 among shoppers ages 18-34). Why is this happening?

A common reason: Dissatisfaction with the mobile shopping – or mobile buying – experience. According to the results of a Facebook study, the top reasons people cited for buying on their desktop or laptop versus their smartphone or tablet included:

It’s easier to see all the available products on desktop/laptop (54 percent);

It’s difficult on a smartphone/tablet to compare products/retailers and get all of the information I need (26 percent); and

Entering personal data (address, credit card number, etc.) is not very user friendly on mobile devices (26 percent).

Addressing these issues is a challenge for retailers. After all, it’s unlikely consumers will actually triple the amount of time they spend on their phones between now and then.

Ultimately, while e-commerce is increasingly a mobile-first ecosystem, it’s still very far from a mobile-only ecosystem. To maximize outcomes in today’s multi-device landscape, marketers should embrace smart digital messaging strategies that help them reach customers right where they are on the “multi-device path” in a given moment. With an intelligent messaging toolset, brands can automatically target consumers on the exact channel each individual end user is most likely to engage with, given their existing interaction habits and preferences.

With targeted offers and high-impact, data-driven messaging tactics, brands will see their audience making more purchases across a range of devices and channels. To learn more, contact us.


Filed Under: Tags: , , With: digital marketing, Engagement, Marketing Techniques, Mobile Marketing, Personalization

Building Customer Loyalty: Is Your Brand “Present” Enough for Your Audience?

By Derek Strong

Building Customer Loyalty - OtherLevels

According to recent loyalty research, the key drivers of consumers’ engagement with brands are shifting from rational values (such as pricing and features) to emotional – meaning the quality of your product may be less important to your customers than how your brand makes them feel.

It’s a marked shift, but it aligns with larger trends: Given the massive number of marketers clamoring for their attention, consumers are in a position of control (leading them to become more selective about the brands they engage with, and more connected to the ones they love).

This creates both opportunities and challenges for marketers – nearly two-thirds of whom are still focusing their brand messages on the functional elements of their products, despite changing dynamics. Now that creating an emotional connection with your audience is key to winning their loyalty, it’s time to invest in building one-to-one relationships with customers – in part, by making your brand a more engaged presence in their digital lives.

‘Engagement’ and ’email’ aren’t synonymous

Brand marketers understand that they need to consistently interact with their customers in order to win their loyalty. For the last decade, email marketing has been their primary avenue for communicating and engaging with an existing base. It’s easy to see why: Since the dawn of online retail, email has been an embedded element of the e-commerce experience.

With 86.7 percent of companies increasing email marketing budgets in 2016, the continued dominance of email marketing is making consumers’ inboxes more crowded than ever – creating the need for marketers to “break through” to them with more personalized, context-driven offers.

Personalization is an area where brands think they’re doing well: According to a recent study by Forrester, 66 percent of marketers rate their personalization efforts as “very good” or “excellent.” But when it comes to issues facing their personalized marketing strategy, 45 percent say their top concern is “consumers delete most email offers and promotions without reading them.”

That outlook essentially places marketers’ problems on customers’ shoulders: We’re doing a great job giving them what they wantThey’re just ignoring it. 

Going beyond the inbox

That thinking reveals a larger disconnect. Across the board, consumers engage with targeted offers from their favorite brands when the offers are delivered they way the consumers, as individuals, prefer. The problem is that by relying so heavily on email, marketers are failing to be present – that is, failing to meet customers where they live, across various apps, platforms, and devices.

Now that consumers shop regularly on apps and mobile web browsers, email’s prominence in marketing is steadily declining. 61 percent of consumers now start the purchase process on a smartphone and continue it on another device, making it important to engage users on their preferred device and channel if they want to drive return purchases and build loyalty.

Sophisticated automation solutions use customer profiling to automatically decide, in real-time, which channel is most likely to drive engagement at the user level. That functionality can help marketers reach 100 percent of their audience with each campaign and tie measurable, attributable revenue results to digital messaging efforts – often at lower costs than through email-first automation systems.

Ultimately, earning an emotional connection between the customer and the brand is vital to business outcomes. To be present for their audience (rationally, and emotionally) marketers need to understand their customers’ preferences at an individual level and meet them, every time, with an omnichannel-optimized toolset.

Filed Under: Tags: , , , , , , With: blog, digital marketing, Engagement, Retention

5 Ways a ‘Rich Inbox’ Benefits Brands (and App Users)

By Brendan O'Kane

From emails to push notifications, digital messaging is a proven strategy for driving clicks and conversions among any brand’s audience. But when it comes to growing engagement, few channels are as valuable as those inside a brand’s app or platform – and that’s where “rich inbox” functionality comes in.

Rich inboxes are ’email-like’ inboxes within a mobile app or company website – but the benefits they pose go far beyond traditional messaging avenues like email and SMS. Here are five reasons every retailer, gaming company, loyalty program, or hotel brand should put rich inbox on their radar.

Reach 100% of your audience. There is no optin! Hence every audience member can see the rich inbox content on desktop, mobile web and inside an app.

An easy avenue for action. Google research shows that 90% of smartphone owners move between devices to accomplish their goals. In this multi-device marketing landscape, a key to driving outcomes from digital messaging is reaching users when and where they will be most prepared to take action on offers or promotions.

With rich inbox, marketers can reach their audience at the exact moment users are already interacting with the brand – increasing the likelihood of those users being motivated to click and convert.

A new experience, every time. Consumers often get bored with apps when the newness factor wears off. Rich inboxes can help mitigate churn and boost engagement by keeping users interested.

Unlike other messaging channels, the data behind rich inbox systems is pulled from a server – rather than pushed out – enabling brands to ensure there’s fresh content in the app each time the user arrives. App publishers can control the content dynamically to make sure it’s timely, relevant, and actionable (and compel users to come back time and again to see what’s new).

Customization is endless. What makes an in-platform inbox “rich” is its look and feel: Companies can personalize their message portals with imagery, logos, and colors to reinforce their branding. They can make messages viewable in eye-catching visual formats according to user preferences – for example, by showing messages as tiles instead of text – to attract the end user’s attention.

Alternatively, brands can “commercialize” their rich inboxes to create additional marketing opportunities. Clickable banner ad-style headers can give drive clicks to partners or referral forms; or messaging partnerships with other brands can allow you to deploy high-value offers from advertisers from inside your own app.

Outstanding analytics and behavioural insights. What is better than seeing which content the user reads, and re-reads ? For example you may be a sport site, yet the user has not yet expressed any preferences. If they repeatedly click on tennis content, then tracking that event means they can be served more tennis content. Rich inboxes offer endless ways to identify users unexpressed preferences.

No matter your marketing and digital messaging goals, a branded rich inbox can help you achieve them. To get started building a custom rich inbox of your own, contact OtherLevels.

Filed Under: Tags: , With: Engagement, Marketing Techniques, Mobile Marketing, Retail