Tell us how technology became your career. What inspired you to join Treasure Data*?

I like to joke that I’m either “the oldest millennial” or “the youngest Gen X” because I was born in 1980, but got my first computer when I was 6 and grew up as a “digital native” back when we were just called “nerds.” From there I’ve always been an “early adopter” of technology and paid my way through college partially by building websites for small businesses, etc. I had my own social media startup in my 20s in Japan for a while, which was successful if you don’t count money! That experience inspired me to learn about Sales and Marketing best practices. So, I got a job in SaaS sales at a public company and worked my way up through Sales and Marketing Management.

As for Treasure Data, I was actually about to join another company a few years back when a contact of mine at Treasure Data encouraged me to come for an interview. I remember liking the “feel” of the office the moment I walked in and while talking with the Japanese Founders I said, “Look, I don’t know much about the data business, but I know how to build Sales and Marketing teams and I speak Japanese.” They hired me that day and it’s one of the best things that’s ever happened to me!

What makes Treasure Data unique in a crowded Customer Data Platform (CDP) ecosystem?

We’ve always been a “data-first” company. Our mission statement is essentially “to help companies activate data-insights by bridging the gap between the exponential growth of data and the shortage of skilled data workers.”  Treasure Data’s roots are in hard-core data management. We have authored some of the most popular tools used by Data Engineers and Data Scientists today and we are very good at solving the “messy” parts of data. Not just when data is siloed in different applications, but when entire technology stacks are spread across different subsidiaries, teams and systems.

Many other CDPs excel at more “focused” use cases where data is coming in predictable/common types such as Advertising, Email, CRM, Web and Mobile and that can be a great fit for SMB and mid-market companies. Companies come to us when they have to deal with all of the above, plus data from legacy systems, home-grown platforms, product data and IoT data — especially in retail. We have the technology and expertise to help these organizations. From my perspective it all boils down to this: if companies truly want to personalize the customer experience, they can’t just be working with the same Marketing datasets everyone else is, they need to think bigger and Treasure Data helps them do that.

How has the acquisition by Arm impacted Treasure Data?

It’s been overwhelmingly positive! You might not immediately think that a semiconductor design company and a data/CDP business make sense together, but they’re actually highly complementary business models, especially as Arm leans into IoT. Treasure Data already had retail, gaming and automotive customers blending IoT datasets into Marketing use cases to connect “online” and “offline” customer journeys. We think that trend will continue and likely accelerate. With Arm’s expertise in IoT, mobile and “smart” devices and Treasure Data’s expertise in enterprise-scale data management, this is really a perfect combination.

What is Treasure Data’s role in the Arm Pelion IoT platform?

Treasure Data is the data layer, so it’s pretty much business as usual as far as we’re concerned, But, there are two important things to highlight. One, Treasure Data will continue to focus on CDP-related use cases for the foreseeable future as that’s where the growth in the market is right now. Two, as the use cases for Arm Pelion expand, we’ll plug in to solve the data collection, preparation, analyzation and activation needs for clients. Since we’re already working with IoT datasets in retail, gaming and automotive, it’s really nothing new for us. The Arm Pelion platform means Treasure Data now has more reach, more resources and even bigger goals. It’s quite exciting!

How will IoT impact marketers?

Well, the prediction is that there will be 1 trillion connected devices by 2035 and 10 trillion data streams. The amount of Interaction, Personalization and Customer Experience that will be possible with that level of connectivity is mind-boggling and rather hard to predict. But, If we reduce the scope a bit, we’re already seeing three common use cases.

1. Retail: Combining real-world, in-store customer interactions with WiFi, beacons, and smart-shelves with digital datasets in advertising, mobile personalization, etc.
2. Entertainment: Blending in device data to personalize the experience in Augmented Reality (AR) and Virtual Reality (VR).
3. Automotive: Analyzing safe driving behavior to offer new, personalized programs, discounts, and incentives with various ecosystem partners.

How does AI fit into your CDP offering?

Treasure Data has a hypothesis that the minimum number of segments a company needs to have in order to truly personalize the customer experience is around 100,000. However, Machine Learning (ML) and Artificial Intelligence (AI) will be needed to uncover all those segments. As a human professional marketer, I could probably come with 100 segments to personalize – but that’s it.

When you connect all your datasets – even non-marketing datasets — and you run a Machine Learning algorithm against them, ML can be trained to find all kinds of signals in the noise that a human would miss. To help with this, we include a number of predictive algorithms out of the box to help marketers start uncovering hidden segments but we also include our open-source Machine Learning library (Hivemall) so that digital marketers and data-scientists can import their own algorithms and run those on the unified datasets we helped them build.

How do you prepare for an AI-centric world as a Marketing leader?

The two biggest considerations are to get your data together but be careful to stay human. First, tactically — companies have to have all their data together in order to compete. The results from ML/AI will only be as good as the underlying data-set. Consequently, if you only feed in partial data, you’ll probably get poor results. “Getting your data sorted” should be the priority so you can start experimenting with ML/AI now. Working with the best, cleanest data available will help maximize results.

Strategically, marketers need to remember to be human, to always think of their customers as people. The human interaction is critical. Your customers and partners will always tell you what they want and how they want to be marketed to — just ask them! Once you have that goal in mind, then apply the AI/ML to your unified data set to see how you can best achieve that result. But, remember, it’s always people first, AI second.

What should companies do with more government regulations on the use of data likely to come in 2019 and beyond?

Embrace the changes. And focus on quality over quantity; experience over the scale. The current wave of data protection is good because it will effectively kill one of the worst trends in Marketing we’ve seen in the last 10 years: growth hacking. Now, instead of scraping/hacking together a bunch of contact data that was collected from a number of legal and “not-yet-illegal” methods and essentially using your brain to trick people into clicking on things, marketers need to think about getting back to basics and providing value in exchange for data. As someone who got into marketing because I enjoy the human/emotional element of it. I’m super happy about this change.

What Marketing and Sales Automation tools and technologies do you currently use?

Salesforce, Marketo, Engagio, Clearbit, Hushly, BrightTalk, LinkedIn Navigator, Outreach, 8×8, Google Ads (social ads, etc.), Chartio and, of course, Arm Treasure Data. One thing we do with all our technology partners is connect them into our own system and then try to solve complex problems using our own solution. This helps us maintain expert-level status on the most cutting-edge Marketing problems, and also keeps us honest as we know first-hand when we’re good at solving problems on our own and when it’s better to pull in a partner.

Could you tell us about an outstanding Digital Campaign in your career?

I’m most proud of is our shift into Omnichannel Enterprise Marketing and our move away from pure digital campaigns. With digital tactics I was always optimizing for conversion rates at the top of the funnel or mid-funnel and, while that’s perhaps Marketing best-practice, it wasn’t necessarily business best practice since it didn’t translate to Sales. In late 2017, we decided to move to an Enterprise Marketing strategy with ABM tactics (away from growth hacking) to align Marketing down the full Sales funnel to optimize for WINS and not just vanity Marketing metrics.

To do this, we took a step back and set executive expectation that things might slow down in the short term, but that we were going to aim higher and close bigger. Tactically, we used a combination of good, old-fashioned market research and data analysis. We looked at groupings of customers to spot trends in use-cases, we talked to their Marketing leaders to understand pain-points, we collaborated with our Sales team and re-aligned them to have a vertical focus (instead of territory). We also got input from our partners and engaged with industry analysts. We put together a target account list based on all these insights. Then we used our Treasure Data technology to prioritize certain accounts based on lookalike/predictive models, segmented and prioritized them.

Next, we put together integrated, Omnichannel campaigns to target specific companies. We engaged and iterated at every step, as often as possible. We used the Treasure Data platform to augment or build the functionality we wanted. This involved a combination of personalized web experiences, emails, advertising, and field events where we had tailored messaging for specific verticals and target companies.

The process was slow and took a good 6 months until we started seeing results. To help show progress to the executive team in the short term, we did two things:

• Used an account scoring model with Engagio to show how we were gaining traction in target accounts over time
• Used our own technology to build complex customer journey maps where we could track how campaigns touched different personas at different target accounts over time to prove how our programs were having an effect on different Sales stages.

Then the larger deals started to come in. At first, it was just a couple, then it became a steady flow of inquiries from the larger companies in our target verticals. Now, we routinely see deals that are 10x the value of the deals we saw just 12 months ago and everyone agrees the strategy was a success.

Any predictions for 2019?

Here are three predictions – they mostly revolve around the theme of data privacy:

• Growth hacking will finally die as data security makes it too risky to scrape together profiles and engage them without consent.
• The rise of the “data troll.” These are people who will try to exploit the new data policies and look for any potential breach to essentially ransom money from a company as an alternative to notifying the data regulatory authorities and incurring potentially crippling fines.
• The return of Experience Marketing. Marketers will have to get creative to create interesting or experiences that customers are willing to engage with in exchange for personal data.

How do you inspire your people to work with technology?

I try to do it by running the “Treasure Data runs Treasure Data” program where all their campaigns — digital or field — should run through our own technology somehow, so they can become an expert on how CDPs affect their specific job functions and Marketing channels.

One word that best describes how you work.

Directly.

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

Nothing really. For every “must-have app” I have a backup for when my first choice doesn’t work. I think that’s the product of startup-life: Give up attachments to what you think you like, and focus on getting the job done with whatever tools are available.

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

Get in before everyone else (I’m at my desk at 7:30 am) so I can get ahead and set the pace for the team for the rest of the day. For creative work, however, I like to spend an hour before bed with a glass of whisky, a pencil and a sketch pad. I think a lot of my best creative ideas come after 10 pm.

What are you currently reading?

For work, I read a lot of things. I probably get 30 different publications sent to me daily and I wake up and just scan headlines for anything that seems important and then do a full read of 2-4 articles that I think are the most relevant to what we’re working on. For fun: I love sci-fi – always have, always will.

Though this is hardly an original choice, William Gibson is probably my favorite author of all time — the way he can expand on technology’s impact on the near-future is frighteningly accurate and seriously entertaining.

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

Everything in moderation, including moderation.

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

(Editor’s note: This interview was first published at MarTech Series in January 2019. This is a republished content.)