The right location can make all the difference to the success or failure of a store. And the more stores you manage, the more complex your site selection process becomes. You want to ensure that they not only have every chance to succeed in their own right, but also won’t cannibalize your other locations.
Predictive analytics helps retailers find the optimal location for each of their stores by analyzing a wide range of factors within a trade area, creating sales forecasts, and pinpointing sites that are most likely to succeed. Typically, businesses turn to third-party solutions like Tango to inform their location decisions.
But while there are numerous types of site selection software and services, not every solution takes the same approach to predictive analytics. Broadly speaking, the main approaches to real estate predictive analytics include the traditionalist and the bleeding edge, each of which comes with pros and cons. And the third option—Tango’s approach—brings together the best of both worlds to give you personalized service, the most accurate forecasts, and the ideal locations for your business to thrive.
This article explains what sets traditionalist and bleeding-edge approaches apart, going over how they operate and what that means for your business. Then we show you how Tango’s approach offers the best benefits of each.
Companies like Buxton, Intalytics, Forum Analytics, and Sitewise tend to take a similar approach to predictive analytics. They’ve been around the predictive analytics space for decades and have well-established processes for analyzing trade areas and selecting locations. While each of these solutions will have its own benefits and drawbacks, some commonalities include the following.
When you work with a traditional predictive analytics provider, you’ll generally be working with retail experts who will help you along each step of the process, offering a personalized approach and providing customizations as needed. These providers bring years of experience to the table, and are typically invested in making businesses feel valued and cared for.
Since the traditionalist approach has you working in direct consultation with actual people, they can provide a thorough explanation for what drives their model. You’ll not only gain access to the numbers you need, but you’ll be able to explain where they come from and defend those numbers to executives, stakeholders, and anyone else who needs a justification for why decisions were made.
However, this approach to predictive analytics has drawbacks as well. When it comes to the success or failure of your stores, reliability is even more important than a personal touch.
In site modeling, a single overlooked variable can create drastic discrepancies between a sales forecast and a store’s performance. So it’s crucial that you take every relevant factor into account in order to get the most accurate forecasts.
It’s here that the people-driven, traditionalist approach to predictive analytics comes up short. Humans simply can’t process large amounts of data the way computers can. Even the best experts fail to recognize patterns and misjudge the significance of individual variables. You can only get so far by adding one variable at a time into a regression model. And when people are responsible for inputting and calculating numbers, it introduces the possibility of human error into the equation.
Additionally, most traditional predictive analytics providers have only a single algorithm that they use for every company they work with. Whatever the business type—whether the client is a diner, a healthcare provider, an apparel store, or anything else—they’ll all get the same algorithm.
This means that some businesses—the ones whose needs happen to align well with that algorithm—will get accurate and useful data. But others will be faced with glaring holes from an algorithm that wasn’t designed to factor for the particularities of their industry. Crucial elements can be left out that may make all the difference in the success of a given location.
Some of this is offset by the personalized service businesses receive. That provides the opportunity to offer certain customizations, fill some of those gaps, and end up with results that are closer to the ideal. But the underlying algorithm that drives the model will always be the same no matter what. And it simply won’t work as well for some businesses as it will for others.
Getting started with a traditionalist approach can be quite a process. All the back and forth of the personalized approach, combined with manual research and data entry, and you’re looking at an initial build that can take as long as 12 to 16 weeks to complete.
And no matter which approach you choose, you’ll eventually need to update your site-selection model to account for changes in market conditions and/or your own business. With a traditionalist approach, the update tends to take just as long (and cost just as much) as it did to create it in the first place. That’s a huge use of time and resources every few years.
The bleeding-edge approach to predictive analytics is exemplified by companies like SiteZeus and Qlik Sense. For better—and for worse—they offer pretty much the opposite approach to that of the traditionalists.
When you work with a bleeding-edge predictive analytics provider, you’re essentially just getting a license to use their software. And that isn’t necessarily a bad thing in itself.
The fact that you’re only dealing with software means it can handle large amounts of data without missing things or introducing human error. They may also use elements of AI and/or machine learning to improve their accuracy.
You won’t have to go through a lengthy setup process before seeing results. And you likely won’t have to deal directly with people, so you won’t have anyone else holding up your process or influencing your decisions. But this self-guided approach may also involve spending more time waiting for in-depth support.
Everything is automated and streamlined. You plug in your data, and you get results. It’s simple and straightforward, and it’ll do the job just fine for some businesses. But for others, that personal touch is more important. Especially when it comes time to explain your results and advocate for your choices.
With a software-centric approach, you don’t get nearly the level of personalized consultation that you get with a traditionalist approach. You’ll have technical support like you would with any other software provider, but you won’t have retail experts guiding you through the process, providing advice, or creating customizations.
You’re getting a product not a partner.
And when it comes to explaining what drives their model, you’re typically faced with a black-box problem. You put numbers in, and you get numbers out, but what happens in-between is a proprietary algorithm you don’t have access to. They won’t give you detailed insights into how they arrived at your model.
If the results they provide lead to good site selection, that’s great, but you won’t be able to show your work or demonstrate why you chose the sites you did—other than to point at the software and hope that it got it right.
And whether or not the software got it right is also in question because—much like traditionalists—many bleeding-edge providers use a single algorithm for every kind of business. Without retail real estate experts working with you to build the models, these solutions lack the insights needed to understand the nuances of your stores. So you’ll still be faced with the fact that their model works much better for some businesses than others, and it may not weigh your relevant variables properly. And the only way to know for sure whether it works well for yours is to gamble on it.
Tango’s approach to predictive analytics bridges the gap between the traditionalist and bleeding-edge approaches to give you a compromise-free solution that is accurate, explainable, and personalized.
You’ll be working with real people who will personally lead you through the whole process, providing expert advice and allowing for individualized customizations. But unlike with the traditionalist approach, you won’t rely on a single predictive model regardless of your business.
Tango employs a variety of different models and algorithms depending on the specific needs of your company. And we use advanced AI and machine learning to determine which algorithms to use and how to layer, combine, and/or weight them to find the best fit for your data. No other solution does this.
Our unique process results in a 40% increase in accuracy versus traditional methods, giving you forecasts you can trust. And you won’t have to trust them blindly. Unlike the bleeding-edge approach, there’s no black-box problem with Tango. We provide you with a full, in-depth explanation of your model’s output.
Traditionalist and bleeding-edge approaches each come with pros and cons, and for some businesses, they work well enough. But only Tango’s approach gives you the very best accuracy, with a personalized touch, and with zero compromises.
Ready to see what Tango can do for your business? Request a demo today.