Davis Giedt: Welcome everyone. My name is Davis Giedt, and I’m the leader of the Benchmarking and Analytics practice here at the Alexander Group. On the phone we have Remen Okoruwa, co-founder of StatusQuota, a firm focused on using predictive analytics to drive visibility into sales and marketing outcomes. Remen and StatusQuota have worked with many Fortune 500 companies to help them better utilize large volumes of pipeline, customer and transaction data to help improve sales outcomes. Today, we’re going to talk a little bit about how to uncover and capitalize on the best revenue opportunities. This touches on pillar one of Alexander Group’s Revenue Growth Model. Pillar one is called revenue segments. The Revenue Growth Model as a whole was developed by Alexander Group as a framework for understanding and organizing the levers that drive top-line revenue growth. Visually, the model is made up of nine key drivers or pillars, with the first pillar focusing on better understanding your buyer journey by segment of Persona – one; and two, quantifying revenue potential by segment; and three predicting how much of that potential can be captured by your sales force. Getting a better understanding of your revenue segments is crucial to improving your go-to-customer model. How you prioritize revenue opportunities influences the way your sellers deliver value propositions to customers, which channels are most advantageous for you to sell through? And, of course, which roles are best suited to cover your accounts? At the core of it, uncovering and scoring revenue opportunities involves gaining a deeper understanding of who your best customers are and capitalizing on those opportunities. So, Remen, will you tell us a little bit about why you’re interested in this topic and what your perspective is on it?
Remen Okoruwa: Sure. Happy to Davis, and thanks for having me on. So a little bit of background on me. I really cut my teeth in sales on the sales team at HubSpot, which was a really fantastic experience. The company was and continues to be a real growth rocket. And, you know, prior to doing sales, I actually worked as a management consultant serving a broad set of clients. With those two experiences, I was always interested in how you can use data and insights to sell smarter, not just as a sales rep, but as a sales organization more broadly. And that’s what led me to found StatusQuota where we take a really data-driven perspective to helping tech clients grow faster, focusing on improved sales, marketing and retention efforts. Now, over the past few years, the team here at StatusQuota, we’ve built a number of custom revenue opportunity scoring models to help large technology companies with really complicated lead funnels, as well as tons of information to synthesize. We frequently see a few things are going on once we look at the data under the hood. One. Your best customers aren’t who you think they are, and then two, you may be targeting the right people, but with the wrong messaging. In both of these situations, there are big opportunities to drive more revenue growth by working a little bit smarter.
Davis Giedt: So tell us a little bit about what common issues you see companies make when they’re looking to build lead scoring models for themselves.
Remen Okoruwa: So a few of the common issues we see when we work with a new client is one – you’d be surprised, but there’s a lot of companies that have no lead scoring model set up in the first place. We’re going in, it’s completely greenfield, which is actually exciting for us because it means that there’s an opportunity to create a lot of value. The second thing that we’re actually going to see, and I think this is a lot more common, especially when you get to larger companies, is that there is a scoring model in place, but the scoring is all manual and it’s being based on arbitrary rules. So think of an example where downloading a whitepaper is worth five points and opening an email associated with a campaign is worth one point. At some point, the marketing operations team decides that if you’ve earned enough of these points doing different activities across the website, you become a qualified lead. Well, that’s better than nothing. What that misses out on is the fact that when you do predictive analysis, you can actually choose to value certain actions not based on an arbitrary rule, but based on the actions likelihood to actually lead to a conversion down the road or purchase in the future. And that’s really where we think there’s an opportunity to take a more predictive approach. Now the third thing we oftentimes are going to see when we come into a client situation is that they might have a predictive model, but it’s one predictive model and it’s trying to predict conversions across the entire sales organization and it’s not being segment specific. If you’re thinking about best practice here, you’re always going to want to have a different model for, let’s say, your SMB segment versus your middle market and your enterprise. If that’s the way that you slice your universe of leads and opportunities, the behaviors of different companies of different scales with different value propositions, it’s going to be different and you want to make sure that your analytics reflect that reality.
Davis Giedt: Ok, so what are some of the first steps I should take to start building a good model?
Remen Okoruwa: In our process, we like to break this down into seven steps now. The first step to get to your question, though, is just data warehousing. Really getting a hold and centralizing all of the data needed to do the project. This is going to include stuff like data cleansing. Every client we work with initially apologizes for the quality of data coming out of Salesforce. This is a really common issue. We’ve all been there and a big part of our process is doing an initial data cleanse to make sure that all the data coming out is usable in a predictive model. We don’t want a garbage in, garbage out situation. Now the second step is benchmarking the funnel. We want to be comparing funnel metrics, think conversion rates, deal sizes, the time different opportunities are spending in different stages to fingerprint the unique sales process for each client. And once we have those benchmarks in place, the next step is choosing the question to answer. Most people that we’re working with, they’re going to care about two things. A – how do we score the likelihood of a deal coming to a close? And then B – how do we identify just how big the deal will be when it closes. Two really common questions that most sales leaders care quite a bit about. The next step for us is building the initial predictive model. That process is really just taking all the data that we’ve collected and using some algorithms to rank the most important traits that are predicting a desired outcome. As the fifth step in the process, we want to go take all these initial results and share them with the client because oftentimes we’re going to find a few things pop out when they see that. One, there’s predictable insights coming out of the model that are really important but predictable. Two, there are oftentimes some really big aha moments that the client didn’t realize and can significantly transform the way they think about a go-to-market strategy. And then three, you know what? We’re data, people. Sometimes the data we’re using can actually give us a little bit of a funky result. Sitting down with the client allows us to figure out which data we should exclude for the next round of iteration so that we can get an even better model output. Going to the sixth step. We run the model again based on the feedback given to us by the client. And then finally, we hand off the model to the client with tactical recommendations for how they should readjust their sales and marketing resources to tackle the opportunities that we’ve spotted.
Davis Giedt: So let’s say I’ve gone through the seven steps correctly, what sort of impact can I expect from my new model?
Remen Okoruwa: This actually reminds me of a recent client situation we had. We’re working with a pretty big tech company that was struggling to apply which deals were most likely to close and why. That meant they were wasting marketing and sales resources on low likelihood opportunities. Their big sales challenge was getting insights from thousands of Salesforce and eloquent records, which they attempted to review in the past with little success. So after we create a predictive model, we found one of those aha moments. For this company, that aha was that one of the industries they are selling into, the services industry, was a huge winner for the client. It was actually 47 times more predictive of deal success than any other industry trade that we could see. The client had no idea they were sitting on such a massive opportunity, given that they sell to nearly every kind of company that you can imagine. Once we built the final model, our data back testing showed that it was delivering a 51% improvement on lead screening, and that’s being scaled across more than a billion dollars of new sales pipeline each month. That’s huge improvement on what was being done before.
Davis Giedt: Yeah, thanks very much for the insights. It’s been great conversation so far. So as you’ve seen, predictive analytics can be real powerful for uncovering and scoring revenue opportunities and ultimately for reducing wasted sales and marketing efforts. This is a very important concept for Alexander Group and for many of its clients as getting it right, as we’ve shown here, impacts a lot of elements farther downstream in the go-to-market model. And the reality is it doesn’t necessarily take a Herculean effort to change the status quo. You just need the right approach to get started. Just to wrap things up, feel free to contact us at agiinsightsexp.wpengine.com to learn more about how you can take steps to improve your revenue opportunity scoring approach.
Remen Okoruwa: You can visit us at StatusQuota SEO to learn a little bit more about our analytics work and how we help companies optimize their revenue processes.
Davis Giedt: Thanks for listening!
