Email marketing workflows that are used in various marketing solutions like HubSpot or Salesforce Marketing Cloud (formerly known as ExactTarget) are often confused with customer journeys. However, workflows and customer journeys are not the same thing. As will be shown below, workflows and customer journeys share a similar relationship to that of landscapes and maps. The former exists whether or not you map it, while the latter seeks to describe the former. Appreciating this distinction is useful for marketers who want to engage customers in ways that are more in tune with those customers’ preferences.
Workflows Are Based On Generalized Models Which May Or May Not Reflect Customer Behavior
Workflows are based on models that attempt to describe customer behavior to varying degrees of accuracy and validity. Below are a few common approaches to modelling customer behavior.
The Classic Purchase Funnel
A classic model of customer behavior is the purchase funnel, which moves from brand awareness through opinion, consideration, and preference to purchase. In recent years, there has been criticism of this model. Critics assert that it does not accurately describe customer behavior. In its place new approaches have become popular.
Discover, Explore, Buy, Engage
In his article for Forbes, Steven Noble, a senior analyst at Forrester Research, puts forward a new model of customer behavior. He argues that the customer lifecycle is more accurately described by four distinct phases — discover, explore, buy and engage — than it is by the classic sales funnel. Let’s look at an example to see how this works.
Suppose a person is in the the market for a new laptop. According to this model, she would:
- Discover: The buyer-to-be does research and discovers various models and brands of laptops, such as Lenovo and Macs.
- Explore: The customer weighs her options in terms of price and quality.
- Buy: After reviewing her options, the customer purchases a Lenovo.
- Engage: The relationship doesn’t end after purchase as the customer continues to receive information from the retailer from which she bought the laptop, which helps keep the brand top of mind.
The above models have their uses. However, both of these models seek to generalize customer behavior rather than understand and predict the unique behavior of individual customers. For this reason, both models will be a far from perfect fit at the individual level.
Workflows Cannot In of Themselves Account for Each Individual Customer
Many platforms such as Marketo and HubSpot offer solutions that enable marketers to create automated workflows that control outreach on various channels like email and social. These solutions use automated sequences and triggers which alter those sequences in order to adapt to the behavior of customers. However, while such systems do indeed respond to individual actions, their responses are more or less canned.
For instance, an automated workflow may send a win-back email to customers at risk of attrition after they reach a certain point of non-activity. While such a workflow is responding to a customer’s action (or lack thereof), it’s not necessarily responding in such a way that reflects that person’s individual tastes, nor does it account for the entirety of the customer’s interaction with the retailer. Continuing with the example, the win-back email may feature a best-seller at a discount price–rather than an offer more closely tailored to the taste of the individual customer. In order to adapt to the customer’s personal behavior, marketers need to use a personalization system of some kind, such as customer journey optimization.
Customer Journeys By Definition Reflect Customer Behavior
A customer is going to interact with your company (and others) of their own volition. Your marketing no doubt affects how they will interact with your company, but ultimately their actions are their own. More specifically, customers will browse and purchase what they like and disregard that which they don’t. We call the environment of products and content in which they do so a ‘transactional space’.
Each customer moves through the transactional space along a certain trajectory. Their trajectory intersects with various items, such as products and content, within that space. This means that they are on a course to consume those items at some point in time. Beyond trajectory, there are other dynamics that influence the paths customers take through the transactional space. These dynamics can be described by two principles:
The Coherence Principle
People have tastes that are unique to them. We call this sphere of tastes a comfort zone.
The Consumption Principle
As customers travel on their journeys, they fluctuate between states of satisfaction and need according to their particular tastes. Meeting a need can satisfy that taste for a certain period of time before the customer will desire a product again. By gently nudging people slightly outside of their comfort zone, you can introduce them to products they would not normally purchase, resulting in increased revenue.
You Need Machine Learning to Map and Optimize Customer Journeys
Marketers can and do create very sophisticated workflows that model customer behavior and drive outreach across channels. However, such workflows, though complex, can still be understood and controlled by a sufficiently competent human.
In contrast, it’s impossible for even a really smart and hard working human to set up workflows that model the journeys of thousands or millions of individual customers. The sheer number of paths each customer takes as well as the dynamics influencing those paths are so overwhelmingly complex that they necessitate advanced data science coupled with automated machine learning to interpret and act upon in an efficient and effective manner.
At Coherent Path we use a type of math called hyperbolic geometry to map out transactional spaces and plot customer journeys through these spaces. Our system then determines how to guide customers towards certain outcomes — such as more frequent purchases or cross-sells — by leading them along a personalized sequence of products in the transactional space.
Many people speak of customer journeys and workflows as though they were the synonymous, but they are not. Workflows are essentially models that attempt to describe generalized customer behavior. They can approach a certain degree of personalization through the use of automated triggers, but by these means alone are not capable of responding in a unique manner to the behavior of individual customers.
In contrast, customer journeys are similar to natural phenomena in that they occur regardless of whether or not they have been modelled. They are also unique to each customer. Through the use of advanced data science and machine learning, customer journeys can be mapped out and influenced.