Marginal Value Exchange
New Wealth through Anonymous Engagement
Intuition
It began as a modification of a COVID Early Warning (CVEW) system to protect the privacy of business employees. It ends with a way to implement CVEW more straightforwardly and effectively.
Once we could anonymously link individuals with networks of timestamped values, we realized that this privacy-preserving technique solves a fundamental obstacle to adopting conventional customer loyalty reward (CLR) systems.
A subsequent intuition realized that this network of timestamped values could yield both tactical CLR interventions and opportunities to nudge people in directions that could enhance the prosperity and livability of communities worldwide.
We named our enabling platform AEgis.
Motivation
These intuitions motivated us to translate ideas into technology and protocols. In other words, our first intuition enabled producers to capture the (majority of) consumers who would not be willing to exchange their personal information (such as email or mobile phone number) for potential product or service discounts. While measurable and substantial, traditional buy-until-you’re-earned-it rewards are still blunt motivation instruments.
We realized that once we could measure how many times a consumer exchanged value with a producer, we had a new and previously hidden piece of field intelligence about consumer behavior. That intel was still based on anonymity.
For example, AEgis could identify the most loyal customer in real-time (because that distinction can move among one or more individuals constantly). We could measure how many consumers exchanged value repeatedly and how often they acted. We could measure how long consumers took (ceteris paribus) to get to any level of loyalty. We could measure changes in the frequency of repeat business, as well.
We came to name this new field intel as Marginal Value Exchange. To clarify this name, consider the stream of consumers who do business with a producer only once (perhaps because they pass through town and stop at your restaurant for no reason). They have an MVE of 1. Now consider a local patronizing the producer twice (the second time because they enjoyed your product or service). This local is more than simply more valuable (by the second purchase) than the passer-through; they are more valuable because the probability of returning a third or more times is much higher.
Affecting this probability of returned business can be a function of tactical interventions enabled by the AEgis ability to identify the relevant and non-personal data about their past consumer behavior. In other words, AEgis observations can reduce the uncertainty of specific consumer behavior. Less uncertainty invariably yields higher returns. This is how AEgis creates wealth otherwise unavailable to businesses and the government.
Justification
The next problem we faced was how to measure these behavioral observations. How do we give a value to Marginal Value Exchange?
We almost accidentally displayed ten months of AEgis data from a Singapore restaurant in a radar graph. We sought an alternative to our first choice, the standard bar chart. The fundamental issue with the bar chart is that it has no right-side limit. This means it’s harder to garner relative value from the bars alone.
A radar graph, on the other hand, is a circle with a known limiting value: 360 degrees. Now we can fit as many consumers as possible into n/360 degrees of an area in the chart. Further, the radar chart nicely broke into as many concentric circles as our data had for the maximum return visits. So the “radius” of each circle is the number of repeat visits for any consumer.
By treating each consumer as an equal proportion of all other consumers, we could measure the “area” of the radar chart for
- each consumer and
- for consumers en masse.
Ideally, if, for example, the entire business with six return visits for all consumers was all the establishment could handle, the radar graph would only have the circumference with a radius of 6. The area of the circle would be a little over 113 MVEs.
In the meantime, each consumer is part of a subgroup with a slice of the pie. The simplest case is R=1 (the center of the radar graph. The area of this first circle is 3.1415. If there are 700 one-time visitors, each contributes 700/total consumers of that area. The Singapore data, for instance, had 73% of all customers being one-time visitors. This yields a contribution of 2.303 out of 3.1415 MVEs.
We follow the same mathematics for each of the remaining circles. Finally, we sum up all these sub-circle areas to find what percentage of available wealth we can claim. As of March 2023, the Singapore restaurant has a mere 4.79 MVEs. In other words, the restaurant leaves 95.77% of potential wealth on the table.
See the appendix for an Excel model of this data and a details description of the mathematics.
Implementation
At this stage, we realized that most businesses have neither the time nor the inclination to get into the technical details of MVE and AEgis. But we immediately realized such engagement was unnecessary if we offered AEgis/MVE as a service. Here we provide the producer with two things: 1) a simple set of numbers that tell them how they are doing, and 2) specific tactical interventions that reduce behavioral uncertainty the most.