Interview with Lyle Stevens, Co-founder and CEO, Mavrck

Tell us about your journey into martech and how you started Mavrck?

Prior to Mavrck, I actually worked in the aerospace and defense industry as a cybersecurity product manager. One of the initiatives I worked on was building and deploying an enterprise social network to help employees securely collaborate with each other, so they wouldn’t use LinkedIn or Facebook and increase their exposure to social engineering attacks. I later helped to develop a data-driven methodology for identifying employees we deemed “subject matter experts” based on the topics they were posting about and earning the most engagement around. These subject matter experts could be used to disseminate information faster within the company. It wasn’t until I was visiting family for Thanksgiving, where two of my three younger siblings were arguing over who was more popular on Facebook, that I got the original inspiration for what is now Mavrck. Being the data nerd that I am, I asked my younger brothers to link their Facebook accounts to a very basic app that would iterate through their social posts and tally the likes and comments they received, giving them a “score” and settling the argument. Upon my return to Boston, I chatted with my now co-founder Sean, and we were curious if marketers would want to understand which customers were the most “popular” or “influential” on social networks and leverage them to spread their marketing messages. Turns out, most marketers do.

How is Mavrck different from other influencer marketing platforms?

Mavrck is the only all-in-one platform for enterprise brands to identify and activate all personas of influence. If you look at the influencer marketing platform industry, most platforms can be segmented into two camps: influencer databases and influencer marketplaces. Influencer databases have an inventory of various types of influencers for marketers to choose from, but the influencer database company doesn’t have any relationship with the actual influencer, so the marketer is on its own to build that relationship. On the other hand, influencer marketplaces do have relationships with the influencers, but usually with only one or two persona types, who jump from campaign to campaign based on the highest bidder, creating no real relationship with the brand. At Mavrck, we’ve taken a hybrid approach that allows a marketer to embed our technology within an existing database of customers that they already have (e.g., ecommerce website, email list, mobile app user, loyalty program), identify the customers with various tiers of influence within that “database” to build a genuine relationship, and activate them to drive the most value on behalf of the brand (e.g., social posts, blog posts, youtube videos, referral links, ratings, reviews, or surveys).

As a data geek, how do you foresee the power of data science in shaping up automation industry for marketing and sales?

It is an exciting time to be working in marketing. We are witnessing two massive paradigms shifts in society collide with each other. 87% of the three billion people on the internet have a social media account and spend an average 1.7 hours per day on a social network. This mass adoption of social media across the globe has resulted in a fundamental shift in how we consume and trust information from institutional to distributed, giving birth to the age of the consumer. Additionally, 90% of the data on the internet was created in the last two years, equal to 2.5 quintillion bytes per day (that’s 18 zeros worth of data per day), paving the way for our fourth industrial revolution around machine learning and artificial intelligence. The collision of these paradigm shifts is resulting in the explosion of new industries like influencer marketing and chatbots that leverage deep learning technologies to automate and scale interactions with people. Examples of these deep learning technologies include automatic speech recognition (ASR), natural language understanding (NLU), natural language classifiers, and image recognition.

How does Mavrck change the way brands communicate with their customers and displace the traditional display ad? Please elaborate.

It comes down to ideas people trust. At Mavrck, our manifesto is to empower brands to rise above the noise through content people trust and we believe people trust content from other people, not brands. In fact, studies have shown that more than 80% of consumers trust content from people they know versus a brand’s ad. Our ultimate vision is to help brands reduce their dependence on display ads to spread their messages because they have the ability to get their consumers to spread those messages on their behalf at a similar scale. One key to making this a reality is to require brands to relinquish some control over their messages, allowing their consumers to define and say them in their own authentic voices. The marketers who accept that they no longer define their brands as much as their consumers do and leverage that dynamic to drive business value will be better positioned to win in the age of the consumer.

What is the current state of Influencer technologies in 2018? How much of that state is influenced by the maturity of data science and customer experience platforms?

The current state of influencer technologies is great! Here’s what the data says: More than 80% of marketers are using influencer marketing and say it is an effective tactic. A majority of marketers are now managing always-on influencer marketing programs, versus one-off campaigns. In 2017, search traffic on Google around the topic “influencer marketing” surpassed the search traffic around the topic “social media marketing.” Over the last three years, the volume of #Sponsored content on social networks has doubled each year, but it still equates to less than 1% of all posts created on social networks, meaning we have a lot of upside still. Data science has certainly been a key ingredient for helping to make influencer marketing more automated and scalable. For the longest time, the attributes you could programmatically derive about influencers were based on text only. But the majority of the signals on social networks these days is visual, so using visual recognition to programmatically detect the attributes of influencers has been game-changing for the industry.

How is the social media analytics industry today different from when you first started? How do current audience analytics tools enable influencers to outgrow their audience reach?

We’ve entered the third phase of the social media industries evolution, or what I sometimes call Social 3.0. Phase one (2007 to 2012) was characterized by brands investing heavily in their owned social channels, amassing Facebook page fans and Twitter followers who in turn engaged with brand posted on a regular basis. However, as social networks IPO’d and the pressure of investor expectations mounted, the need for additional revenue accelerated the transition into Phase Two (2012 to 2016). This phase was characterized by brands shifting their social strategy to a paid strategy, that was fed by social listening and monitoring. Owning a social channel started to become permission to play in order to promote paid content to target audiences. While “mommy bloggers” became all the rage in 2009, we didn’t enter Phase 3 until 2016, when the number of “influencer” platforms doubled in a year. Phase 3 marked the shift of the social strategy to be more consumer-centric, with consumers spreading the word of brands at scale. With more than 21 million #Sponsored posts being created in 2017, we are now at a point consumers have developed their own personal brand and need tools to manage that brand just like a marketer would. At Mavrck, we’ve begun testing tools to help make tasks easier and more automated for the individual influencer.

How do you see the influencer marketing strategies evolving around omnichannel customer experience standards?

At Mavrck, we believe a cornerstone of your influencer strategy should be identifying and activating existing consumers with influence in the moment of interaction with your brand. Whether that’s your website, your mobile app or your customer support line, you need to know where in the influence spectrum someone falls, and leverage that information accordingly. Our Influence Plugin technology aims to do just that within these existing customer journey touchpoints. Additionally, marketers are now looking to influencer marketing to impact the entire customer journey. Traditionally, influencers were used to drive top funnel awareness, mainly because that was the easiest to measure.  With advancements in influencer marketing measurement and attribution, marketers can impact the awareness, consideration, purchase, loyalty, and advocacy stages of the customer journey.

Which startups in martech and adtech industries are you keenly following?

I’m following startups who are applying artificial intelligence or blockchain to marketing use cases. For example, Netra is using machine learning to derive visual intelligence from photos, including logos, age, gender scenes and context. Dock.io is using blockchain to give consumers more control over their data, while also enabling developers to get access to data from many sources (e.g. Facebook, Google, LinkedIn) without needing to have users authenticate each source separately.

What marketing and sales automation technologies do you use?

At Mavrck, we use platforms like HubSpot, Google Analytics, and Drift to help automate our marketing, sales and customer support processes.  We also leverage Zapier to build automation between platforms that may support integrations out-of-the-box.

Tell us about the new standards of Influencer martech engagements. Who does it best and how?

The new standard of influencer marketing involves running “always on” programs across the full spectrum of influence that work together to maximize business results. For example, we’ve powered an integrated strategy for Astral Brands that has enabled the company to run its referral, advocate, and influencer programs via Mavrck, where each program feeds the next. The result is consumer activation and value creation at scale. One of Astral’s brands, COSMEDIX, won a Shorty Award in Beauty for its influencer strategy and partnership with Mavrck.

How do you prepare for an AI-centric ecosystem as a technology leader?

I try to sponge as much info as possible on the subject, that satisfies my focus on being a lifelong learner. My go-to resources include Medium, podcasts, and subreddits on the subject. There’s also this Medium post that got me into some really interesting podcasts: The 10 Best AI, Data Science and Machine Learning Podcasts. I also like to engage in discussions with peers and colleagues on the potential impacts of artificial intelligence to make sure I understand various perspectives. Recently, I was debating with colleagues about the Google Duplex demo, and while I was really excited about all the possibilities, one colleague mentioned the fear they had around Google Duplex being able to simulate my voice and give direction to the company without it really being me.

One word that best describes how you work.

Data. I love to analyze and understand it. Every decision involves data and if the data isn’t good enough, I make a decision with commitment to revisit that decision when the data is better. At Mavrck, every company, team, or individual objective has key results that are measurable with data, making it very clear what success looks like.

What apps/software/tools can’t you live without?

Breather. When you’re on the road 50% of the time, especially when doing a day-trip to NYC, having on-demand office space to get work done, take a customer call, or just unplug for a few minutes is game changing. No matter the city, the spaces are consistent and dependable with their amenities.

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

Gmail Filters. I try to check email only three times a day for an hour each: once in the morning, once midday, and once at night. Otherwise, I could spend all day reading and writing emails. In order to make each session as efficient as possible, I’ve created 58 gmail filters that categorize my email and allow me to prioritize certain types of email messages during my three daily sessions. I’ve even trained my team to put “Action Required” in the subject line if they need me to take action on something within 24-48 hours. Anything more urgent than that requires a phone call.

What are you currently reading? (What do you read, and how do you consume information?)

I typically read and listen to non-fiction audio books for maximum retention. Currently I am “reading” A Higher Loyalty by James Comey and Shoe Dog by Phil Knight. I also like to use Blinkist first to get a summary of the book, and if I like the summary, I’ll then buy the full book.

What’s the best advice you’ve ever received?

One of my first mentors gave me great advice around prioritization. He said that you can plot all work on a matrix of urgent versus important. Anything that’s not urgent or important, you can eliminate. Anything that’s urgent but not important, you should delegate. Anything that’s urgent and important you obviously do, but a key to success is to work on at least one important but not yet urgent task everyday so that important things never become urgent. I later learned this was called the Eisenhower principle based on something he once said in a speech: “I have two kinds of problems: the urgent and the important. The urgent are not important, and the important are never urgent.”

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

Matt Moog, CEO at PowerReviews

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