WebProNews

Recency: Predictive Marketing

The concept of Recency as a human behavioral metric goes back to the early 1900's.

A whole body of research was conducted, building on the work done by Pavlov and his famous dogs. Just before feeding time, Pavlov would ring a bell, then feed the dogs. By repeating this over and over, the dogs began to associate the ringing of the bell with being fed, to the point of salivating when the bell rang - even if there was no food present. Stimulus and response; the bell rings and the dog salivates. Edward Lee Thorndike took this idea a step further with humans and proposed the Law of Effect - the response to any particular situation, if followed by a rewarding experience, will become the likely response to the situation. Now, you may think to yourself, "Wow, how brilliant - duh" but he was the first guy to say it, so there.

J.B. Watson, building on the work of Pavlov and Thorndike, was the first to formalize the idea of Recency. He noted a man was more likely to get up from his chair when a woman entered the room if he had done so Recently; the more Recently he had done this, the more likely it was he would do it again. Apparently, he was able to get a hold of some bums who didn't stand up when a lady walked into the room for comparison. The point is, these "behaviorists," as they were called, studied human behavior and watched for patterns. They then used these patterns to make predictions about human behavior.

Recency found its way into the psychological literature in many ways, because it simply kept popping up on the radar screen. When studying the ability of a subject to remember a list of words, two effects were found: Primacy and Recency. When a list of words was spoken, subjects had far better luck remembering the first and last words spoken than ones in the middle. Primacy refers to words at the beginning of the list - those spoken first. Recency refers to words at the end of the list - those spoken most Recently, or last.

Why is there a Recency effect? One theory: there is simply something about human behavior and decision-making that draws us to the familiar - that which has happened to us Recently. These events are fresh in our minds; we understand all the details and implications of them, and they tend to strongly influence our behavior going forward - positively or negatively. A Recent bad experience can be just as powerful as a Recent good experience - right?

Here is the way I look at Recency in the context of marketing: Recency simply reflects the power of the Customer LifeCycle. As customers pass through different stages of their relationship with you their needs change, and keep changing until they don't need you any more. At this point, the LifeCycle ends and customer value accrual (either positive or negative) stops.

Note: A LifeCycle does not cover a human lifetime, and LifeTime Value is based on the LifeCycle, not a human lifetime. Until you know the LifeCycle of a customer, you cannot compute LifeTime Value, because you don't know where the "LifeTime" ends. So please stop worrying about LifeTime Value, and concentrate on first things first - understanding the customer LifeCycle.

But I digress; where was I? Yes... Recency.

Think of it this way. If I have just engaged in a transaction with your business, something has changed or occurred in my life that caused me to need your products or services and to reach out for them. Some event has taken place that changed my outlook or need for your business. Perhaps something I own needed to be replaced. Perhaps I got a new job and can buy more stuff. Perhaps I moved and now need a different set of services than before.

The point is, there was a triggering event and it changed my behavior. Going forward, there is no reason to believe this change is not permanent until through my behavior I tell you things have changed again. If this event caused me to need your products or services, it is likely I will need them again and again, unless something changes. And the point where this need is greatest is the point closest to the triggering event. Right after this event, I need your products or services the most. As time goes on, I need your products or services less and less, because I either have what I need or I find substitutes - better quality for the price, a higher level of service for the price, and so on.

If the above scenario is true, then the more Recently I have engaged in a transaction with you, the more likely I am to engage in another transaction with you - relative to those who have not engaged in a transaction as Recently as I have. People who have not transacted with you for a long time are simply less likely to transact with you again, relative to those who have transacted with you Recently. Please note the use of the word "relative." We are not talking about absolutes here; we are talking about comparisons of one customer to another. A customer who has transacted with you Recently is more likely to transact with you again relative to a customer who has not transacted with you in some time. In fact, customers can be ranked by their Recency (number of days or weeks since the last transaction); this ranking in effect sorts all your customers by their likelihood to transact with you. Those in the top 20% of the Recency ranking are far more likely to transact with you than those in the bottom 20% of the Recency ranking.

If Recent customers are more likely to transact, then two other ideas follow:

1. The more Recent they are, the more likely they are to respond to promotions

2. The more Recent they are, the higher their potential value to the business

The first idea is just common sense. If a customer is are more likely to transact they are more likely to respond when contacted about a transaction - you are in effect pushing customers who are already predisposed to transact right off the fence. Like shootin' fish in a barrel, so to speak. If you are more likely to go rock climbing than somebody else, if somebody asks both of you if you want to go rock climbing, you are a lot more likely to say yes than the other person. No science to that, just logic. This assumes, of course, you can actually tell who is more likely to go rock climbing. Oh, but you can, remember? The person who has gone rock climbing more Recently is the one more likely to go rock climbing again. This does not mean the person will go, just that they are more likely to go rock climbing relative to the other person.

The second idea is just as logical, but perhaps a bit mysterious at first. Customer transactions normally increase customer value. A customer who is more likely to transact and more likely to respond has the potential to contribute more value to your company by transacting than a customer who is less likely to transact and respond. So it follows that the more Recent the customer relative to others, the higher potential value they have, because their likelihood to transact and contribute value is higher than others less Recent.

Next Article: Recency: Visit Behavior Predicts Visitor Value

This article is taken from the book Drilling Down: Turning Customer Data into Profits with a Spreadsheet. The first article in this series can be found here.

Jim Novo has nearly 20 years of experience using customer data to increase profits. He is co-author of The Guide to Web Analytics and author of Drilling Down:Turning Customer Data into Profits with a Spreadsheet. If you want more visitors to take action on your web site, try using the free conversion metrics calculator, downloadable here. If you need to sell more to customers while reducing marketing expenses, get the first nine chapters of the Drilling Down book free at http://www.drillingdownbook.com.
Ask Jim a Traffic Analysis Question!