Prachi Gupta, Editor and Tech Blogger at July Systems details how a predictive location analytics platform can provide insights into customer churn across channels and help consolidate relationships with wavering customers.
Customer churn is the clearest sign of a brand’s diminishing value. But it is extremely hard to find out the root cause of customer churn. For a long time now, retailers have been trying to improve customer retention by trying to identify the most profitable sales channel. They have been embroiled in the online versus offline crisis. But what they don’t realise is that these two channels are not divergent, one is not necessarily weaker than the other.
If you want to improve customer retention, the digital versus physical approach to retail sales needs to be abandoned in favour of a more integrated approach because today’s customers are dynamic, empowered and shop across multiple channels. To truly understand customers and the reasons they churn, retailers need to analyse the digital and physical behaviours of customers to identify patterns and then use them to predict future trends.
Fortunately, by using a predictive analytics tool, businesses can gain insights into the reasons of customer churn. A predictive location analytics tool is not really the technological equivalent of a crystal ball; rather it uses data such as – precise customer location, their in-venue behaviour and their dwell times to identify which customers are likely to leave, much before they actually leave. Location analytics can be implemented as a stand-alone system but its predictive capability increases remarkably when combined with CRM and other data sources. This multi-pronged approach allows customers to be segmented and profiled more accurately, meaning that communication and interaction with customers can be made more relevant and therefore much more likely to be acted upon.
For example, the WiFi analytics platform deployed at a shopping mall can tell you that a particular visitor has been observed visiting three furniture outlets recently without making a purchase. The odds are excellent that he is contemplating buying a desk, lounge suite or some similar product but is probably still on the fence about making an actual purchase. Perhaps because the price is a deterrent or maybe he needs to seek his wife’s opinion, who is not accompanying him. Now, based on these insights if you configure your analytics platform to send an SMS with a discount coupon to one of these furniture stores, he is very likely to respond positively because price is no longer the barrier to consumption. Or you could email a catalogue of the latest product line to him, which he can go over with his wife at home and come back to the store to buy it. Basically, you are not confining sales to any single channel. In fact, you are creating every opportunity for the customer to access your product on any channel he wishes to purchase from.
The digitally connected consumer can be very demanding when it comes to what kind of content he’s served. Instead of annoying a potential customer with a blanket marketing strategy, tailored messages can be seen by the recipient as helpful suggestions. Sending coupons or other promotions only to those who are most likely to make use of them is not only the most efficient use of available marketing resources, it will also be appreciated by the customer and will lead to better engagement over the longer term. Known as propensity modelling, analytics can help determine what kind of topic will grab a customer’s attention, how a message regarding it should be structured and which delivery channel will yield the highest conversion rate. For instance, sending someone a discount coupon by email for a restaurant he’s passing by will be futile. Because by the time he reads the email, he’s already at home having dinner. In such a case, an SMS or in-app notification will be more appropriate. Multiple engagement channels are available to the modern retailer: social media and in-store approaches, for example.
Knowing how to use each most appropriately and effectively can make the difference between taking marketing to the next level and failing at effective customer engagement. Predictive location analytics can assist in finding correlations between variables in large data sets with ease. Let’s try and understand how that works: suppose a certain downtown mall, which once experienced great footfall on weekends suddenly finds itself struggling to draw in visitors. The mall owner can’t explain the reason of such drastic customer churn. All he has are educated guesses, but they aren’t helping him lure the customers back to the mall. Had he deployed a WiFi analytics platform, he would have seen that a certain restaurant in his shopping mall typically sees an uptick in business between 6pm and 8pm on Saturday evenings, which also corresponds with the release of a new movie every weekend. And to top it all, the parking lot was always full. If all three of these conditions are met at the same time, a human observer might not be able to explain the amount of crowding, much less predict it. But with a predictive location analytics platform, he could have anticipated this crisis and made better preparations to deal with a high volume of visitors and cars. For example, the mall management could send coupons for ride pooling services to help visitors avoid parking troubles. Or run a special offer for guests who arrive between 2pm to 5pm.
Sometimes customers leave because they can’t find what they are looking for. An in-store predictive analytics tool can address such issues. By using a location analytics platform, your store staff can instantly view the purchasing history of individual customers, identify cross-selling opportunities in real time, gain visibility into store inventory and track the location of specific items. If a customer cannot locate a particular item, the floor assistant can immediately identify another store where the item is available and can redirect the customer there.
Since a lot of customers are already informed about the products they want to purchase, they do not want the store staff to deliver a rehearsed sales pitch. The staff will be better utilised if they can actually ‘engage’ the customers and act as shopping advisors. Floor assistants with access to real-time analytics can be more responsive, better informed and more efficient at completing their tasks. Customer attitudes, offerings from the competition and market conditions are constantly shifting. Keeping track of your client base can, at times, feel like trying to nail jelly to the ceiling. More accurate profiling and segmentation of the customer base can make this much easier, enabling effort to be directed at the types of customers with whom it will show the greatest benefit.
“Predictive location analytics can assist in finding correlations between variables in large data sets with ease”
Of obvious interest are regular big spenders, but also those customers who are exhibiting specific behaviours associated with decreasing engagement. For instance, if someone used to visit a location once a week but hasn’t returned for more than a month, a targeted promotional message might just be what’s needed to stop him from churning. Similarly, predictive analytics allow underperforming customers to be targeted, or be used to point out other types of marketing opportunities. Multiple different channels can be used not only to gather information, but also to reach out to specific customers, or those fitting a specific set of characteristics. Choosing the correct channels and engagement strategy becomes much easier when customer profiling is supported by predictive analytics.
Predictive location analytics has just about reached that point on the technology maturation curve where the early adopters have made all the mistakes the rest of us need to learn from, while people are still busy figuring out all the ways in which it can be used to improve the bottom line. In short, right now is the perfect moment to invest in such a system.