673: Predictive Analytics: What You Need to Know Before Getting Started w/ Vaughn Aust

ABOUT THIS EPISODE

In this episode we talk to Vaughn Aust, EVP of Digital & Product at MarketStar.

Click here to connect with this guest on LinkedIn.

Are you struggling to come up with original content weekend and week out? Start a podcast. Interview your ideal clients, let them talk about what they care about most, and never run out of content ideas again. Learn more at sweet fish Mediacom. You're listening to the BTB growth show, a podcast dedicated to helping betb executives achieve explosive growth. Whether you're looking for techniques and strategies or tools and resources, you've come to the right place. I'm James Carberry and I'm Jonathan Green. Let's get into the show. Welcome back to the BDB growth show. We are here today with Vaughan Aust he is the evp of digital and product at market star. Van Hey, awesome to day, James, glad to be on. I'm really excited to chat with you today. Vaughn. We're going to be talking about predictive analytics and and really how you can utilize predictive to accelerate sales. But before we dive into that,...

Vaughana, love for you to just tell our listeners a little bit about market star and what you guys are up to. Yeah, some market star. We actually just registered trademark sales of the service, and so we are an outsource sales company. We've been doing that for thirty years for a lot of the fortune hundred brands, but recently we've really found a lot of growth in upandcoming SASS companys that are looking to do growth hacking and grow in a new markets, new areas and and many times move down stream from enterprise S and B as they evolved their products set. So sales as the services what we do. I love it. Savan in and looking at a predictive analytics. Obviously you guys doing sales as of service for for your customers. Predictive as something that's huge part of, I would imagine, the results that you're able to drive for your customers is we were talking offline. You said that there are a few things that people need to look at before they ever get started...

...and starting to experiment with predictive can you talk through kind of what are things that people need to know before they go down the path of really trying to take advantage of predictive? Yeah, I mean predictives kind of like really, really good reporting or analytics. Your your data has to be in a pretty, pretty good spot, and by pretty good spot it's got to be connected. So a lot of clients that we work with, you know, they have their their marketing system and their marketing data. They have their sales system and their sales data, they have their earp and and they have their, you know, pos data there and a lot of companies have got those connected. But if you don't, then I would definitely start with that, because predictive and machine learnings is not a silver bullet to solve data that doesn't interconnect. And so I usually look at what clients are at and and it's interesting. A lot of our clients we we do those connections for them. We...

...pull those data together in a data warehouses and make it available for for predictive of machine learning. But I would start there and and then the second thing is there's lots and lots and lots of data, and with machine learning specifically, it's okay to have lots of data. And so automating your data collection as much as possible through your marketing system, your sales system and your point of sale will help you get a richer data set that will give you better results. And so I'd start with the data. I love. It makes perfect sense. And then let's say that you know, people have their data in order there they're ready to get going. Let's let's talk about what all is possible with predictive analytics and as they're going out and developing of the they've got all the tech and place, have got their data online. They're ready to pull the trigger. What can they expect or what should they be looking to be able to do once they start really leveraging predictive yeah, I think...

...the number one thing to start with is a business problem that will provide value to their business. And it sounds simple, but a lot of times, you know me, included three four years ago, I to go to a conference, I hear some awesome things on how predictive and machine learnings being used and things like healthcare and policing and like Oh man, if we can apply that to sales and particular, that that we've got something there. And so sometimes people just jump right into it. I I take a step back and say what is a key business problem or key question that I have that, if I answer it's going to provide real value, and then I would experiment and I would pass and I would refine it before you blow it out into you know, changing the way that you do business, because it really has a potential to do that. But if you jump into that as the end in mind, this is going to change everything, I think you might be disappointed. And I've definitely been kicked in the teeth a number of times and...

...got through that cycle with several clients where okay, all right, let's go back and get something that we know we can derive business value from and move the team forward. Got It. So could you share some examples on of some business problems that can be solved by leveraging predictive analytics? Yeah, I mean where I sought first surfacing was, you know, there were software technology and companies are adding, hey, we do ai and and A. Prioritize leads and opportunities, right, so you focus your effort on where you can predictively get the most good. The other is prospects. Right, once we figure out predictively what your best customer is or what they look like, then we'll go out and we'll use that same predictive score to apply it to others that look like that, you know, having tent like that and that may purpose from you. So that's been the really the foundation area is, how do you prioritize leads and opportunities? How do you prioritize prospects and go after very, very, very targeted prospects? But some of the some of the...

...exciting stuff that we've been playing around with and I've seen a lot of others doing, is optimizing sales activities. You know, Ta could taken a look at where you know what channel people are communicating in, what their preferences are, how their behavior changes based on how you communicate to them. The other is the message itself. What our message resonates. And similar to marketing, right, you know what email message, what content resonates. You do the same thing. And sales conversation, right. What what key phrases? What key phrases help you win? What key traises help you lose? And so to me that's a really, really exciting area because then a salesperson can focus on the sale and the conversation, not on all the busy work leading up to it, in a north of efficiency game. For several years now I've been using a graphic design service called Capa Ninety nine. You've probably heard me talk about them on the show before, because it's one of the best investments we've ever made in our business. For just three hundred ninet nine bucks a...

...month you can get unlimited graphic design. Seriously, it really is unlimited. You can have them design your infographics, case studies and all of your blog posts and social media graphics. They even do t shirt designs, email signatures and book layouts. So head over to BB growth showcom design and sign up for their fifteen day free trial. That's BB growth showcom slash design. You're going to love it. All right, let's get back to this interview. I was with the company out in San Francisco, it might have been close to a year ago, and they're telling me that they were just getting enormous amount of inbound leads because they had written some content that was ranked on the first page of Google and and so predictive for them, kind of the primary use case was, okay, we've got to figure out how to prioritize all the leads that are coming in. And as they are telling me that, I thought, man like, what a champagne problem to have, like you're getting so many leads coming in that you're having to prioritize. You know which...

...ones to go after. So I love on that you that you shared a couple other use cases of how you can use predictive because I know a lot of folks listening to this don't have that champagne problem of having too many leads come in, but knowing that they can use predictive to figure out the messaging that's resonating and, along with you know, a whole host of things that predicted can be used for, not not necessarily just, you know, lead prioritization. Absolutely and and you know for every client that has because a lot of times clients a hire us to sell a new product into a new marketer, to go downstream. But but we the first thing they asked for is two years of sales date and what we do with that sales state is we pinned any third party data we can get on demographics and things they've done in the past, and we have some companies that can provide us with a digital footprint and how much it's growing over time, and we bring all these things in and then we have a richer. This is what your ideal customer is right. It's not just, you know, size of business, number of...

...employees, vertical sold, you know the the king five. It's literally like they're seventy or eighty things that we can use to say this is what your best customers look like, but then you use that to go back and apply to find prospects and that's that's there. There's several companies and software companies that are focusing on that. But the thing is is, with that level of variability of data, that's where the art comes in and that's where I've kind of struggled a little bit with some software companies, because you still need to figure out what are those attributes and do that predictive intent on what your best customer is, and those data sources vary depending on the industry you sell in and and what kind of thing you're selling, and so I've fallen down a little bit just on the surface buying software that says, Hey, we will use predictive intent and machine learning to prioritize your leads and prospects. Because of that art that has to happen, pulling the data together and then creating the model, and I've kind of stopped looking to software to do that. I've actually got a team of analysts and for clients we do it based on a...

...model that the analysts create based on what clients have available and what we can procure for them, and I've better results than trying to plug in the software to do lead and opportunity prioritization. Interesting, okay, so as because I've seen a lot of software companies pop up that claim to do this. But in your experience variants going having having an actual team inhouse that's that's doing that has allowed you to see better results. Well, it's a value, right, it's a value conversation. A lot of that. You know. That's how much your pain per month, per rep, how many reps you have involved? And then what is the efficiency gain you'll get in prioritization versus, you know, have an analyst that runs lots of different models, lots of different things. And Yeah, I've I found higher predictive intent using inhand house analysts, not because they're better than the software companies, but the software companies are trying to go against the broad based right and anytime you've that level variability...

...of, you know, the data that you can collect and who you're selling into, it makes it tough. Now, software company who've had really, really good success with is a company called Gong, and I think the reason why we've had such good success with them is because they have a really stable data set right they do speech to text and then they know that that is the data that they're going to run machine learning on in an aion, and so they can build that into their software and we've had a lot of success using them to enable the man to find places to coach Reps. Hmm, I love it. Okay, sov on, this has been super helpful in kind of what folks should be looking for before they get started, what they can do with predictive and once they do get started, or really almost like knowing what you're going to want to solve, the business case for what you're wanting to do predictive with before you ever get started in the first place, and then also being careful looking at software to do predictive when you might be able to get better results because of the more granular kind of data set that...

...you're you're looking at by, you know, doing something in house. Is there anything else can related to predictive analytics von that you think listeners should understand before I let you go today? Yeah, I'd say predictive analytics is, I don't know, I always use colorful terms. It's it's kind of like, you know, light the fire with matches, right, you know, it's not quite the Stone Age, which is which I would consider segmentation and all those fun kinds of things. But the thing about predictive is if you get the predictive right and you can start using correlation, then when you get to the point of having massive data and using machine learning, that's like using a flamethrower. So to me it's a journey of getting your data right, running some predictive correlation and then you get into ai and machine learning, which is, you know, they can take hundreds of different attributes. You don't have to have an analyst that boils it down to a few and you can. You can find correlations that you would never think of. Love it.

For me. That's that's the exciting frontier, which is, you know, once you get your data at least where's interconnected, then it doesn't matter how much you have because machine learning will be able to handle it. Love it on. This has been fantastic. If there's somebody listening to this, they want to stay connected with you, they want to learn more about market star. What's the best way for them to go about doing that? Yeah, solutely connect with me on Linke Dan. It's Bond Vau Thhn dot off. Yeah, I'd love to hear from you if anyone's got some projects go on. Our things are interesting. Er adviceyed one. I've definitely been kicked in the teeth a few times. Lad hears that therey and I'd love to have to help out. Love it wonderful. Vold will, thank you so much for your time today. This has been fantastic. I really appreciate it. There are lots of ways to build a community and we've chosen to build the bed growth community through this podcast. But because of the way podcasts work, it's really hard to engage with our listeners, and without engagement it's tough to...

...build a great community. So here's what we've decided to do. We're organizing small dinners across the country with our listeners and guests. No sales pitches, no agenda, just great conversations with likeminded people. Will Talk Business, will talk family, will talk goals and dreams, will build friendships. So if you'd like to be a part of a BB growth dinner in a sitting near you, go to be to be growth dinnerscom. That's Bob Growth dinnerscom thank you so much for listening. Until next time.

In-Stream Audio Search

NEW

Search across all episodes within this podcast

Episodes (1802)