Welcome to the 2019 Predictive Analytics World (PAW) Conference Series. Recorded live in Las Vegas, this series is bringing interviews straight to you from exhibitors and speakers at this year’s event. In this interview, host Jamin Brazil interviews Brian Shindurling, VP of Marketing at Big Squid.
Find Brain Online:
Hi, I’m Jamin, and you are listening to the Happy Market Research Podcast. We are live at Predictive Analytics World, Marketing Analytics World, and many other worlds. My guest right now is Brian, VP of Marketing at Big Squid. You guys have the coolest booth, I think, on the show floor by the way.
Oh, thank you.
The hot pink’s great. I have a kick-ass sticker that I’ve added to my bag. Thanks very much for the tchotchke stuff.
Tell me a little bit about the company.
Cool, so, we’re Big Squid. We’re based in Salt Lake City, Utah. We’ve been around for ten years actually. The company’s kind of evolved out the marketing and analytics world. We’ve done a lot of business intelligence consulting in our years past. And, really, kind of the genesis of where we are today was we were consistently hearing a lot of the same challenges with our customers where leveraging business intelligence for an analytics environment, you’re building really interesting and cool dashboards all the time, but you’re always looking at data in a historical context, which led our customers to the next questions, which is “What’s going to happen? “Why is it happening?” and “What can I do about it?”
I mean that’s the Holy Grail.
Absolutely. That’s right. So, a couple of years ago, we launched our product we call Kracken. It’s an automated machine-learning platform.
Thank you, thank you. And the approach that we’ve taken is to integrate with the analytics infrastructure that BI analysts are using every day. So, most of your enterprise data warehouses that are on the market today, most of the major business intelligence platforms… We’re able to round-trip data in and out of those environments where we’re basically just enhancing it with predictive metrics to give analysts a little bit better idea of what’s going to happen.
That’s really cool. What kind of data are you dealing with?
It depends. We deal with all kinds of different data. We’re kind of a horizontal play. We work with companies across basically any vertical that you can think of. The data that we play with is always structured, again coming out of kind of that business intelligence environment.
So it’s been cleaned.
It’s been pretty well curated, pretty well cleaned. This is data that has been used or is being used for reporting on a day-to-day basis. So we’re lucky in that sense; there’s already been some thought behind the business questions that we’re trying to support with analytics. Yeah, structured data and set up in a way that it’s being used in reporting environments.
Who is an ideal customer?
Good question. So, our ideal customer is the BI-analyst or data engineer, those that are leveraging these platforms like a Snowflake and/or a Tableau, Looker, Click (Places where they’re leveraging data on a day-to-day basis to derive insights and then reporting on and telling stories to their executive stakeholders about what’s happening in the business and what do we think is going to happen, how should we be thinking about making things better. But they haven’t really been classically trained on data science and machine learning in the past. So what we’ve done is we’ve created a platform that enables them to very easily navigate towards that concept that Gartner calls a “citizen data scientist.” So, leveraging an automated platform that really brings the R and the Python, the math, and the stats that most data science practitioners have in their back pocket to the analyst who is more closely aligned usually with the executive personnel, the stakeholders who are making decisions on a day-to-day basis.
There’s been a lot of movement not just in this space but across the board like in primary data and others where it’s like this democratization of access to the actual insight, which prior was almost impossible because it required, at a minimum, some advanced math and stats, which is like Master’s level. And now, it’s like you’ve got these solutions that are allowing the common folk, as you said Gartner cited correctly, the citizens… So, when you kind of frame things out a little bit more, is the buyer the analyst inside of the organization? Or is it happening at the CTO level? Who’s the one that’s sequestering the budget?
Another great question. It’s an interesting space. This is absolutely an emerging market; it’s red hot right now. What’s fascinating is most often companies haven’t budgeted specifically for…
It’s like a new dollar, right?
It’s a new dollar, absolutely. That’s a good way to describe it. So for us, we take an approach where we try to create champions, our actual users, people who are excited about being able to expand their skill set to have a bigger impact on their organization, to make a name for themselves. But in almost every occasion, we have to navigate our way up to executive level and, most frequently, C-level sponsorship in order to reallocate budget into the environment where they can make an acquisition of a product like ours.
Yeah, totally. Is the initial sales strategy a little bit like a B2C. I know you’re not connecting in it at that level. But does it feel a little bit more starting in the trenches of the organization?
It can be, yeah. I like to think it the way we approach the market as kind of top-down and bottom-up.
Yeah. Both directions?
Both directions, yeah. And, if we’re lucky, we meet in the middle. We have an informed executive, and we have a champion who knows exactly how they want to leverage it, and they’re pushing for the insights that they can build.
So, with all the success that you’ve had (this next question is going to be kind of hard to answer) but I’d like you to pick one example of a project where it was just your favorite. You connected with it; it seemed like it went really, really well; clearer kind of view of how people are using you.
Yeah, OK, cool. I’d say for me my favorite customer story is probably that of Skullcandy. For those who don’t know, Skullcandy is a very innovative player in the audio space, kind of a life style brand. They build headphones of all shapes, colors, and sizes.
I have some.
OK, good. I hope you enjoy them. The story with Skullcandy is they were trying to leverage some kind of predictive analytics to inform the way they’re developing their products. So, what they were able to do, actually leveraging some NLP earlier on in their data pipeline, they were able to call out key words in customer reviews from places like Amazon and things like that. So they were getting an understanding of, from a commentary perspective, where are things failing. If we have a one- or two-star review, what are the key words that are being used over and over again. We took the outcome of the NLP model and started to feed it into a supervised machine-learning model with all of the attributes around specific products, new products going to market. And the cool part about their story is they’re able to now predict what new products that they’re launching are going to have failures and where. So, before they go too far into market with a new product, they have the opportunity to retool, improve the quality of their product, increase the customer experience, make a better product, improve the user’s experience, make everything feel better, and lower their warranty claims costs overall. So they’re having to deal with less claims coming inbound; they’re having to deal with less returns and shipping costs and sending free product, and all those types of things.
That’s really cool. That is really cool. I love that. So, you said… It sounded like what they were leveraging was existing reviews like Amazon-type, Yelp-type stuff, right?
So, you’re able to incorporate things like social media as well as primary data sources, any type of data.
Sure, yeah, yeah. If the data is being captured as long as it can be piped into a data warehouse type environment, we can ETL things together. It becomes a really rich data set that’s fit for machine learning.
That’s really cool. So, what do you think about the show?
The show’s been outstanding. It’s been really, really fun, actually. We’ve had a lot of fun dealing with a lot of data science practitioners and kind of getting their insights on where and how they might leverage something that’s a little bit more automated, where their pain points are, and how these emerging technologies can speed up their workflows and, ultimately, enable them to do more. We’re also had some really fascinating conversations with analysts who are reading and learning and trying to understand this data science space. I think that’s why a lot of people have come to this event; overall, it’s just to learn more about what is this predictive analytics world, so to speak. So it’s been really fun. I feel like our message has been resonating really well, and it’s been a very productive show for us, absolutely.
I think the attendees here are… I mean you have this really nice cross-section between business professionals (these are like buyers) and then also executioners (the people that are actually doing the analyst side of things). So, you’ve got a very nice representation across those two. And then, you got a lot of big brands, right? Big budgets are represented as well. And then, the show floor feels very friendly.
Yeah, I agree.
Yeah, it’s kind of nice.
Yeah, it’s been really fun. We’ve had upwards of 15, 20 people hanging out in our booth at any given time. Wish we had some Polaroid action going on. Come to the booth and get a Polaroid. And people have just been kind of light-hearted and having a good time. It’s been fun.
That’s great. Brian, if somebody wants to get in contact with you or sales at Big Squid, how would they do that?
Probably the best bet is to visit our website BigSquid.com. Obviously, fill out a contact form there. Really easy to find. You can also check out a free trial on our website as well. Sign up and get a 14-day trial and get kind of a feel for what automated machine learning looks like and feels like and what types of insights you can derive.
And, of course, we’ll include that information in the show notes. Brian, thanks for being on the Happy Market Research Podcast.
Thank you very much for having me.
Everybody else, appreciate your time. This show is coming to an end. If you found value in this episode like I did, I hope you will screen capture, share it on social media, LinkedIn, Twitter. I’d greatly appreciate it. Have a wonderful rest of your day.