These Consumers AIn’t Loyal: Using AI to get to really Know Your Customer
Brands and consumers used to be in long-term relationships. Now they’re casually dating with no commitments.
This week, Tesco angered shoppers by announcing that it will be reducing the value of its popular Clubcard reward scheme from June. From the company’s point of view, augmenting the mechanism with which it rewards its consumers for purchases will help protect its bottom line in times of economic uncertainty. Clubcard users, however, were not happy as they feel they would be getting rewarded less for their money at a time when the supermarket chain should value its most loyal customers more than ever.
Questions that emerge are:
How would this decision impact both short-term profitability and long-term brand perception?
What is the best way for a company to structure such an announcement?
In the past, the only way to know the outcome of a decision was to ‘just do it’. Historical analysis was then the go-to way to assess the strategy a company chose to undertake, but there is one obvious problem with this approach: if unsuccessful, the fallback from the wrong decision has already occurred.
At present, simulations can be used as an approach to augment your customer analytics to better understand how to retain existing clients and attract new ones. By mapping out the entire market in a digital twin form, two key areas the model could then focus on are what drives customer loyalty and how marketing channels interact along a customer journey to drive conversion.
A digital twin simulation gives companies the option to test strategies with the help of AI on a virtual audience with accurate results without any real-life implications.
There’s plenty of data, too.
CRM databases on customer interactions, loyalty card data on customer purchase patterns, media habits and exposures for different types of customers, and survey data from consumer panels: All of these data sources provide ample opportunities for exploring and learning more about strategies to employ to help a brand grow.
Just looking at this data can help characterize and identify a number of descriptive characteristics about customers. Some examples of these types of descriptive analytics include answering the following questions:
What percentage of our current customer base shops both online and in-store?
How many people saw your latest TV campaign and about what percentage reacted favorably to it?
How frequently do our customers visit our site? Are they likely buying competitor products as well?
What are customers saying about our latest product feature?
Mining and analyzing a large range of these data sets can provide a lot of useful insight for decision-makers. However, another challenge altogether is molding all of this customer data into a strategic plan for attracting and retaining customers.
Simply describing the customer is not sufficient to answer key strategic decisions.
As a basic example, you might know what percentage of the population saw your brand’s latest TV campaign, but that does not by itself provide insight into the return on investment of that particular campaign. Answering that question requires the integration of information on marketing, consumers, and sales. Such integration is not carried out by simply looking at descriptive components of customer analytics. However, simulation can be used to augment customer analytics and data to answer broader and more strategic questions.
Customer analytics may describe what percentage of your customers are loyal to your brand, product, or service. However, it does not describe why customers engage in a repeat purchase. The following section describes how simulation can be used to understand drivers of loyalty:
Data on individuals’ purchasing behavior has become more widely available over the past decade. This individual-level data has provided many analytic opportunities to further explore consumer behavior. Tying insights into consumer behavior to aggregate sales outcomes will allow marketers to better understand both the short-term and long-term effects of marketing.
One key factor that impacts the long-term effects of marketing is the degree of loyalty that consumers exhibit: how much do consumers tend to repurchase the same brand as opposed to going astray and switching between brands? However, simply quantifying repurchases and switching percentages may not be sufficient to guide a marketing strategy.
It’s important to identify what factors are driving consumers to repurchase a particular brand if the goal is to disrupt the market and change that behavior (get consumers to switch to your brand) or to preserve that behavior in a shifting marketplace (keep consumers purchasing your brand). The first step is to acknowledge that “loyalty” may result from a range of consumer behaviors and mindsets, which are not always related to a consumer’s love or affinity for a particular brand.
Below is a list of various factors that may drive repurchase behavior in a market:
1. Inertia — defaulting to a brand out of habit. Day-to-day consumers are faced with many decisions, but everyone has limited time and energy to apply cognitive effort to choose. So oftentimes, consumers take the decision-making shortcut of simply following a routine and taking the path they usually do.
2. Awareness — buying a brand due to the absence of alternatives in mind. In some cases, consumers may simply take the path because they do not know of any other way.
3. Pricing — choosing the lowest cost alternative. Many consumers are hunting for the best deal and may choose the most competitively priced option.
4. Perception — staying loyal to a brand that is held in high regard. Consumers may continue to buy a brand because they think it is the best of the alternatives available.
Clearly, there may be much more to “consumer loyalty” than warm and fuzzy feelings about a brand.
Perhaps it is human nature that makes many gravitate toward the fourth factor on the list. Most dog owners would hope that their beloved animals find their way home each day out of love as opposed to out of laziness, convenience, or simply because there is no other option. Some brand managers may hold to the same hope about consumers who do not stray from their brand.
Hope is not a strategy though.
To get the most out of customer analytics, managers need to acknowledge that in reality consumer “loyalty” is driven by a range of factors that may constantly evolve in a dynamic market that will allow a marketer to succeed in a competitive environment.