AI can profoundly transform how marketing campaigns are executed. With new machine-learning tools, you can create marketing campaigns that listens and adapts automatically to audiences to improve performance. Tools can hyper-personalize the messaging using dynamic content, adapt email send times, and more.
Motiva.ai is a company offering solutions with these capabilities. I discussed their solution with David Gutelius, CEO and co-founder of Motiva AI. He has worked in the Silicon Valley for the last two decades, and played a part of building some of the most pioneering artificial intelligence products over that time.
For example, he worked with the U.S. national security community post 9/11 on a wide variety of R&D topics, including DARPA’s PAL Program – the largest machine learning program in the US government’s history. That work produced the core IP for SRI International spinout companies, including Siri, Social Kinetics, Tempo AI, TrapIt, and Desti (and still counting).
Interest in AI and human augmentation led to starting three machine learning companies, Social Kinetics (personalized behavioral healthcare – acquired by RedBrick Health) and Proximal Labs (proactive, personalized search and recommendations – acquired by Jive Software). Motiva AI is the most recent.
Now, over to the interview!
What is Your Company’s Background?
Digital marketers often treat customers more like walking transactions than people. They’re chasing the single click — which hopefully (maybe) leads to that single transaction.
Marketing technology (“Martech”), meanwhile, has blurred the line to where advertising and marketing have become hard to tell apart. It’s Madmen meets Ad Tech — where “targets” are constantly bombarded and their personal data surveilled and mined, all to deliver a small bump in response rates. It’s almost as if they think we are dumb enough to confuse an optimized digital offer with a relationship or with an understanding of our needs. We aren’t that dumb.
And we’re all tired of it.
Motiva started from a different place. If Martech is typically all about push messaging, Motiva is about listening, learning, and serving.
What does that mean? Let’s start with the why.
The marketing and sales process is really about building relationships. Transactions or sales reflect good relationships and trust. Our ultimate goal at Motiva is a deeper relationship between firms or organizations and the people they serve. Marketing used to be about this — not clicks.
Key to building a good relationship is active listening. Listening is different than spying. Active listening is like sitting down to coffee with a new friend and engaging in a give and take conversation. Martech today (following Ad Tech) is like sitting in a tree outside someone’s house with binoculars, rifling through their trash, and pelting them with pebbles when they come out. That’s not the world we want to enable — or live in.
Our motto: Be a friend, not a creep.
Relationships pave the way for more than a single transaction. They can lead to multiple transactions, word-of-mouth, good karma, and all kinds of unexpected positive outcomes.
On the other hand, optimizing only for transactions leads to doing whatever you have to in order to get the click or sale done. We think that’s short-sighted and leads to questionable choices that are not in best interests of customers or firms.
We offer assistive, intelligent technologies for marketing and communications that are geared to building persuasive, high-trust relationships between organizations and the people they serve. We mean both B2B and B2C. We also mean governments engaging citizens, nonprofits engaging supporters, and schools engaging students and alumni alike. In short, anywhere we can help build a relationship that provokes action based on listening and learning, we’ll pursue.
It’s a hard, high-quality problem. The technology we need largely doesn’t exist today. It’s not slight improvement on something, it’s not a remix. It’s a different approach, solving for a different set of outcomes.
What Problem Does Motiva Solve?
Motiva AI creates marketing that listens and adapts automatically to audiences, saving time for marketing teams and enhancing the response performance of every campaign, and any scale.
What Does Motiva do, and why is it Important to Your Customers?
Motiva AI brings fully automated campaign optimization to marketing, and delivers better performing campaigns through machine learning. It’s essentially like having a data scientist in a box.
Automatically find and exploit the best messages with multivariate experimentation and campaign optimization.
Motiva AI can test anything about an electronic message: subject lines, secondary subject lines, copy / body text, design elements, graphics, etc – or all of these at once.. Motiva AI experiments automatically shift investment from lower performing towards higher performing messages. No manual intervention is required.
Find the right send time.
Motiva AI will tell you when your best send time slots are for any given campaign.
Understand an audience.
More often than not, Motiva AI exposes high-performing responses among subpopulations in a given segment. It may be the audience you have is actually three or four different, but distinct sub-audiences. This can be a great way to use data to influence the campaign and/or creative design process. This can also expose new subpopulations or customer characteristics that can help challenge assumptions about customer audiences and even become a tool for adjusting SEO, PPC, and other marketing strategies.
Lighten the workload.
Stop trying to design the right valid a/b tests, exporting data, interpreting results, rerunning the campaign, etc. Motiva AI will do all that for you, and more. Marketing teams can save thousands of hours over the course of a year just by letting Motiva AI handle the tedious tasks.
Motiva AI leverages a completely proprietary approach to machine learning to deliver these capabilities that can operate at both tremendous scale and high accuracy.
How Does Motiva use AI?
Artificial intelligence (AI) is touted by many as “the” solution to radically transform marketing automation. Yet specifics are often absent about how we move toward a more productive future.
To understand where you can apply AI, consider how you define a typical email marketing campaign. In its simplest form, you define a population, email message, and the date and time of delivery. In a multi-step campaign, you specify sequences of messages often with time delays and conditional logic governing the delivery of each message. In other words, you take your best guess at what messaging experience will best resonate with the population.
The classic first step in removing the guesswork is conducting an A/B/N test to identify the most compelling message out of two or more possible options. The approach seems simple in principle: randomly select a fraction of the population to serve as the test segment, and then randomly allocate the test segment contacts to cohorts of equal proportions. Each cohort receives one of the messages under consideration. The message option that produces the most significant response in terms of a specified performance criterion, such as unique open or click-through rate, is deemed the winner.
Immediately questions arise about how to apply the test. How large should the test segment be? How do you calculate a measure of confidence in the result? When should you reject the result? What should you do after a rejection? Instead of burdening platform users with these details, machine learning can step in to deliver far better outcomes without the headaches.
The Motiva AI platform does this today by applying a machine learning procedure for incremental testing to remove two issues with A/B/N testing. Prior to conducting an A/B/N test, it’s impossible to know how many test contacts we’ll need to discern the best option with high confidence. That depends on the magnitude of the difference in performance between the best option and its closest competitor.
As depicted below, we address this by spreading campaign execution over time and avoiding the need to define a test segment. In essence, everyone is in the pool. Random subsets of the population get treated at a regular frequency over a specified number of days. As evidence builds, the platform decides whether further testing is required or sufficient evidence exists to commit to the current best option.
A representative Motiva campaign searching for the best message option in terms of potential lead rate, where a potential lead is defined as a click-through without a corresponding unsubscribe.
A more significant weakness we address is the lack of A/B/N testing efficiency. Due to the equal allocation of contacts to message options, a majority of the test segment receives an inferior option when more than two options are sent. Furthermore, as the number of options grows, the fraction of suboptimal message allocations increases. Not good!
Our platform delivers greater efficiency and a better user experience by varying the allocation of contacts to message options based on the available evidence. As responses arrive during the campaign, the learning algorithm updates the daily allocations to send only the most viable options weighted relative to their expected performance. This results in higher overall response rates in Motiva AI-run campaigns without the need for human intervention.
While our work has benefited from the latest research in multi-armed bandit learning, we’ve had to tackle several nuances associated with this scenario. Two in particular are batch experimentation and delayed responses. Most classic multi-armed bandit algorithms assume immediate feedback after a single experiment. In our context, cohorts of audience contacts are treated simultaneously with unknown response delays. The Motiva AI platform integrates all available evidence in a principled manner and explores the most compelling options throughout the campaign with efficiency and ease.
Now what if we could use all available customer data?
At a high level, leveraging customer data in a deeper way creates the opportunity to uncover meaningful populations that exhibit shared content preferences. This opens up new ways to understand customers, both individually and in larger segments, as well as tailor personalized messaging experiences.
In essence, the answer involves learning what relationships exist, if any, between customer attribute and behavior data and their responses to the available message options in an ongoing campaign.
In machine learning terms, the change in the learning objective represents a shift from multi-armed bandit to contextual bandit learning. Here we’re bringing all available context into the learning task to support true customer-level, adaptive personalization of the experience.
By learning models that predict the likelihood of message engagement based on customer attribute and behavior data, we are automatically learning definitions of the underlying populations that are most likely interested in the associated message content. Coupled with an evolving model of message content similarity, the game changes. The possibility of continuous learning across campaigns without human intervention is within reach.
In Your Mind, What is the Future of AI in Business and Marketing?
Most of the current crop of marketing automation platforms were created in the early 2000s to help marketers do this at greater scale. The basic approach of mass push is the same, however, even if the tools have changed: combine customer database, messaging server, and rules engine. Add reporting. Add external integrations. Now sprinkle some AI on top: Salesforce Einstein, Adobe Sensei are recent examples.The underlying assumptions, however, haven’t changed.
But here’s the thing:There’s an alternative coming, and it’s built with Artificial Intelligence at its core — not as an afterthought or add-on. It’s different.
The next wave starts with intelligence and an ability to adapt and learn as the foundation. Add a combination of human marketing experts, incremental learning, and data. Start with the notion that there isn’t one campaign, one funnel, one persona, one channel or one lead score that will win. Make the unit of analysis the person instead of broad segments. This new approach has the potential to upend how we communicate with customers.
Here are five basic ways that things change, and sooner than you may think:
One campaign or a million.The main reason why we run only a few marketing campaigns at a time is because that’s all we can manage as humans. Machines have no such limitation. The latest advances in Bandit-based and Reinforcement Learning (which are two interesting classes of machine learning algorithms out there getting lots of attention) suggest that given minimal input and contextual knowledge, machines can essentially bootstrap themselves into creating complex personal campaigns on demand. Not personalized campaignsin the sense of {token} replacement; I mean the system figures out what the right steps are (and in what order, cadence, channel, and form) to motivate a customer to take a desire action. In other words, instead of a single nurture campaign for a million people, you launch a million campaigns, each of which is learning from all the others, all the time.
The segment of one. Know your audience, craft the message. That’s the game. Humans have an incredibly finite ability to understand people at scale. This is okay for lots of use cases; we’d go nuts otherwise. But people are weird and diverse and they change over time. Imagine instead a machine-human team working together to understand that the right segment for given marketing goal isn’t a broad, generic interest group. Instead it’s dozens or hundreds of specific subgroups that could be addressed in a more targeted and effective way — or a single person.
Content that adapts. Get ready for Deep Learning (another class of machine learning algorithms, based on neural nets) to push the art of the possible forward in sophisticated language and image experimentation. Combined with advances in Reinforcement Learning and an ability to adapt based on incremental human feedback at scale, we’ll see a new kind of ability to create compelling digital content and experiences that improves over time. It won’t take away jobs on the creative team; it will take those jobs in whole new directions.
The New Dream Team: machines + humans. With machine-driven intelligence at the center, marketing (and sales) orgs will either adapt and thrive or they’ll keep doing what they’ve always done and fall behind. Advantage will go to teams ready to adapt workforces, roles, culture, and incentives, collaborating with and learning from intelligent tools. The differences between the old and new will be stark and measurable. The winners will leverage these new capabilities to attract the most important precious resource of all: customer trust. Trust will flow to the companies that earn it authentically, and flow away from those who continue to mindlessly pitch their wares.
A very different model of marketing is coming and will directly, deeply challenge the current major incumbents and how marketing gets done alike. Intelligent at its core, the new marketing engine won’t just incrementally improve things. It’ll change the game for the better, for good.
Do You Think Marketers Will be Replaced by AI Robots?
The nature of marketing work will/is changing. Some tasks that are done by humans today will be fully automated and that work will go to machines.
The rule of thumb I use is the following: If you can think of any particular task or set of tasks that are defined by rules and simple heuristics, that work will be automated – and it will probably happen faster than most marketers imagine.
This is actually great news for most marketers. AI will free up time to spend on doing more creative, complex tasks that machines will not be able to do as well. Who wants to spend time doing a job that’s so repetitive?
But the bigger story here is about augmentation rather than automation. The right human-machine hybrid team combines the creativity and brilliance of humans with the scale and learning of machines.
Those higher performing hybrid teams will quickly outpace teams that are slow to adopt AI into their marketing practices. We’re already seeing massive shifts in consumer retail, where the more forward-leaning marketing teams are just outclassing rivals and as a result driving revenue at the expense of their competition. We’ll see this trend pick up dramatically in 2018-19.
In this interview, David Gutelius provided fascinating insights into how artificial intelligence can be used for campaign optimization and personalization. It is an interesting thought to run one million campaigns targeted to one customer only, rather than one campaign targeting one million customers. I am sure AI and machine learning will transform marketing campaigns onwards!
I am an author, speaker and consultant in marketing automation and artificial intelligence.
Do you need help with marketing automation or AI-based solutions? Contact me and let’s discuss how I can help you!
