MarTech Interview with Robin Bordoli, President at RollWorks

Tell us about your interaction with the new-age technologies like AI, Machine Learning and Robotics.

Before joining RollWorks, I was the CEO of Figure Eight, the leading training data platform for Machine Learning teams. Figure Eight sold its platform into the automotive, financial services, media and entertainment, retail, and technology industries supporting both Natural Language Processing (NLP) and Computer Vision use cases. Thus, I have a broad and deep understanding of the current state of AI and Machine Learning in the real world. Beyond Figure Eight, my experience with computer vision goes back almost 30 years when I was working on Primitive Object Identification for robots in manufacturing plants when I was studying engineering at Cambridge University.

How did you start in this space? What galvanized you to start at RollWorks?

I have a 25-year history with AI and Machine Learning. In the early 90s, I was exposed to Object Identification problems while at Cambridge. In the mid-90s, I was at Micromuse working on inference in real-time fault correlation. In the mid-2000s, I worked on search and advertising relevance problems at Yahoo, followed by Content Recommendation problems at Jive and Lead Scoring problems at Marketo in the early 2010s. Most recently, in Figure Eight I worked on both Natural Language Processing (NLP) and Computer Vision use cases across the aforementioned industries. When presented the opportunity to lead a technology platform like RollWorks, I was drawn to the opportunity because of their Machine Learning pedigree and the ability to apply this capability to the problem of Account-Based Marketing (ABM) at scale.

What is RollWorks?

RollWorks is an account-based platform powered by Data and Machine Learning. It’s used by ambitious Marketing and Sales teams to align on the accounts that matter to them, engage the buying committees within those accounts, and confidently grow revenue.

What technologies do you provide to Marketing and Sales companies?

The RollWorks Account-Based Platform is a SaaS offering for Marketing and Sales teams. Within the platform, there are three solutions: Identification, Engagement, and Measurement. The Identification Solution enables Marketing teams to align with Sales on an ideal customer profile (ICP), create and prioritize a target account list, and select contact data for the buying committee within the target accounts. The Engagement Solution then enables Marketing and Sales teams to engage the known and unknown buyers within the target accounts via Digital Advertising and Email Automation sequences. The Measurement solution allows Marketing teams to measure the impact of these ABM programs on pipeline creation and pipeline acceleration.

Underpinning these solutions in the platform are two foundational layers: Data Graph and Machine Learning. The Data Graph layer is based on an open data strategy that combines multiple proprietary and third-party data sources. User profiles and their attributes are scored based on the level of agreement or disagreement from all sources—leaving us with a view of the best data. The Machine Learning layer applies Machine Learning models at scale across the entire spectrum of the three solutions from dynamic messaging and bidding and budgeting to account scoring and suggestions. Our infrastructure allows us to operate at over one million Machine Learning predictions a second.

Tell us more about your vision into growing revenue opportunities within the ABM industry?

ABM isn’t a fad. It’s a fundamental change in how B2B companies market. Marketing has always been focused on leads or individual people. Sales has always been focused on accounts or companies. Given this, there has been this fundamental disconnect between Marketing and Sales. ABM is about fixing that disconnect and aligning how Marketing and Sales teams go to market. Why would a Marketing team ever spend a dollar on a lead that is from an account that will never buy from them? It’s a waste of money. At RollWorks, we believe that over the next decade every B2B company will transition from lead-based marketing to account-based marketing.

This transition has not been possible until recently since it was too manual for Marketing teams. However, the quality and quantity of B2B data and the scale of Machine Learning capabilities have reached a point where automation is a possibility. Our addressable market over the next decade is approximately 500,000 B2B companies in the U.S.

How is the B2B data market evolving with better digital frameworks and advanced analytics technologies?

The B2B data market continues to evolve with new providers like Clearbit and Bombora emerging alongside the incumbents such as Dun & Bradstreet and Thomson Data. In this ever-changing landscape, the right strategy is to manage an open data strategy with three core attributes: combining first-party and third-party data, ingesting multiple third-party data providers and using the level of agreement and disagreement to score the confidence of the data attributes.

What kind of technology do you leverage to boost ABM adoption?

The Data Graph and Machine Learning layers in our account-based platform are mission-critical to making ABM possible at scale. The Data Graph layer is based on an open data strategy that combines multiple proprietary and third-party data sources. User profiles and their attributes are scored based on the level of agreement or disagreement from all sources – leaving us with a view of the best data.

The Machine Learning layer applies Machine Learning models at scale across the entire spectrum of the three solutions from dynamic messaging to bidding and budgeting to account scoring and suggestions. Our infrastructure allows us to operate at over 1 million Machine Learning predictions a second.

Which regions have been the fastest to adopt MarTech platforms? Which markets are you currently focusing on?

We’re focused just in the US at the moment.

What is the biggest challenge to Digital Transformation in the market you cater to? How does RollWorks contribute to a successful Digital Transformation?

RollWorks is focused on companies that want to transition from lead-based marketing to account-based marketing. So by definition, they have already started their Digital Transformation and doing Digital Marketing. At RollWorks, we then use our ABM Readiness Assessment Framework to help Marketing and Sales teams understand where they are on the journey to applying ABM at scale and how to uplevel.

Where do you see AI-enabled customer journeys and other smart technologies heading beyond 2020?

There are two big fundamental challenges for the AI-enabled customer. The first challenge is identifying unknown buyers so that their journeys are visible to users. This is a data problem which is why we have been investing so deeply in our data graph and our open data strategy.

The second challenge is being able to engage buyers throughout the entire customer journey. This is why Digital Advertising is central to any serious ABM platform. Digital Advertising is the only channel that can consistently reach known, unknown, engaged and unengaged buyers through the lifecycle.

What is your opinion on “Weaponization of AI and Machine Learning”? How do you promote your ideas?

There is a formula I use to explain what AI is to executives that are interested in applying AI to their business. The formula is AI = TD + ML + HITL.

To understand this formula, let’s break it down and imagine a company is trying to create an AI solution that can categorize customer support tickets by severity level based on the unstructured text showing an exchange between a customer and a customer support rep discussing a particular topic or problem.

TD is Training Data. Training data is a set of inputs with the correct outputs or examples with the correct labels that can be used as an example to train the machine. In this example, the input is the unstructured text inside a support ticket. The output, or answer, is the label “severity level” which has been applied by humans according to definitions of severity levels specific to the company in question. An automotive manufacturer will want to define these severity levels differently from a retail banker or a wearable technology company.

ML is Machine Learning.  The Machine Learning capability is the ability to convert training data into a predictive model that can be applied to new inputs—in this case, that would be new support tickets with unstructured text. You want the Machine Learning model to apply its predictive power to create new outputs—in this case, that would be the “severity level” label. One of the advantages of machines compared to humans is their ability to understand their own confidence level. Humans are notoriously overconfident at evaluating their own judgments. So you can accept or reject the prediction based on the machine’s own assessment of its confidence level. For example, if a support ticket has words and phrases which haven’t been seen in the training data, or seen very infrequently, then the machine will objectively assess its own confidence level as being low for that particular prediction.

HITL is Human-in-the-Loop.  This is the critical third component of commercially viable AI. If the Machine Learning model is not confident in its prediction it can route it to humans to review and answer. In this blended model, you take advantage of the speed and scale of Machine Learning to address the less difficult tasks, while the humans handle the harder tasks. AI is not about machines replacing humans; it’s about machines augmenting humans.

My point of view on “Weaponization of AI” is that we need to remember to keep “humans in the loop.” Machines are computation engines. Humans are context engines. Machines and humans are better than machines alone or humans alone. That is our challenge.

What start-ups and labs are you keenly following?

I’m an advisor to two startups that are addressing big, unsolved problems. Datatron is focused on the problem of Machine Learning model governance. If a consumer accuses a financial services company of discrimination in their Machine Learning model, how does the company respond? This is the new domain of machine learning model governance.

NameCoach is another startup I am advising. They are focused on the problem of respectful identity acknowledgment. By some estimates, there are over four million unique first names in the US. Saying someone’s name correctly in situations such as graduation or during the sales process is the problem NameCoach is focused on solving.

What’s your smartest work-related shortcut or productivity hack?

When I was CEO of Figure Eight, I was running a small scrappy startup, and it wouldn’t have been appropriate to hire a full-time executive assistant. I used claralabs.com as my digital EA to schedule external meetings for me. I would copy clara.james@figure-eight.com on my emails and “she” would take over scheduling a meeting. “She” was so efficient that sometimes visitors at our office would ask to meet Clara because they assumed she was a real person. If you’re a scrappy startup or a small business you should definitely check out claralabs.com to streamline scheduling.

Tag the one person in the industry whose answers to these questions you would love to read.

Lukas Biewald, Founder of Figure Eight and Weights and Biases

Thank you, Robin! That was fun and hope to see you back on MarTech Series soon.