Journey Builder is designed for creating 1:1 personalised marketing journeys for your prospect and customer communications.
Thanks to its ability to bring together several communication channels within a single interface in Marketing Cloud, Journey Builder is a constantly evolving tool that enables more actions to be automated as part of customer journeys.
Path Optimizer came with the May 2020 release, and is part of the ‘Flow Control’ family, a group of Journey activities that determine which branch is the most efficient when a path is set up in Journey Builder. We can say that Path Optimizer falls between ‘Join’ and ‘Einstein Split’ activities.
Why is Path Optimizer Needed?
Until now, flow control activities have either waited for something (by duration, by attribute or until a specific date) and allowed you to split a Journey using decision splits (based on subscriber data, eg. city is ‘Michigan), engagement splits (based on opens, clicks, bounces), Einstein Splits (using AI insights), or random splits (based on a %).
So, why add Path Optimizer?
These Flow controls met the need to test and experiment with content, sending frequency and channel type. Shouldn’t the engagement split determine which of the different branches a contact should go down according to these engagement metrics?
Source: Trailhead “Create a Path Optimizer Activity”
That’s where Path Optimizer comes in!
A Path Optimizer activity helps you identify the most effective way to reach your customers by automatically (or manually) selecting a winning branch, and then sending the message to all contacts entering that path of the Journey. Think about it this way: you can test the pieces of the journey prior to the final design of it!
How to Add a Path Optimizer
Here’s how to add a path optimizer with only a few clicks.
Just like any other activity, you can drag and drop the Path Optimizer into a Journey.
Source: Trailhead “Create a Path Optimizer Activity”
In the Path Optimizer settings, you have to fill the following elements:
- Activity Name: enter a short and meaningful name to identify the optimization.
- Activity Description: enter a description to explain the purpose of the path optimizer. It helps make it understandable to your team.
Above: Path Optimizer settings. Source: Trailhead “Create a Path Optimizer Activity”
- Winner Evaluation: there are two options here
o “Email Engagement”: automatically decides which path based on engagement data (click, open, engagement) in order to choose a winning path.
o “Manual Selection”: a marketer can select the winning path that seems most relevant, using their own judgement.
Above: winner evaluation. Source: Trailhead “Create a Path Optimizer Activity”
- Split: used to define the contact allocation (in %) to be included in each branch (up to and including 10 branches).
- Holback: valid for Data Extension entry source with a run once schedule type, selection of the people to be integrated in the test group, others are kept in a group in order to integrate the journey when a winner will be defined.
You can let Journey Builder automatically define the winner based on opening, clicks and unsubscribe (up to you to choose the right one) or manually select one. You can also select a waiting time before a winner is defined.
Use Case for the Path Optimizer
Let’s take the example of a welcome campaign in which I want to perform a multivariate test on the same email in order to identify which email subject or version allows me to have the highest opening rate.
I can design a Journey by splitting the sending of four different email versions.
Example of a Welcome journey
By doing so, I allow Journey Builder to automatically define a winner (i.e “Email Engagement”) and to inject the selected contacts into the holback.
With this feature, Salesforce simplifies the use of multivariate tests and addresses the following issues:
- A limited test volume that can’t allow actionable results → with the Path Optimizer, you can add up to 10 paths for testing with a distribution that equals 100%.
- A difficulty in achieving statistical reliability for decision support → the choice Path Optimizer makes to select a winner automatically or manually is done through simple engagement criteria.
- Wasting time not knowing what to test → Test time is limited with the Path Optimizer, and is equal to the sum of the evaluation period combined to the length of the longest path including wait activities.
Summary
When a major weakness of multivariate tests is uncertainty around what exactly to test, Journey Builder tries to reconcile marketers’ doubts – kind of an iteration inside another.
Personally, I see it as a way to facilitate processing and analysis time to improve Journeys almost automatically, in a short time and without any technical skills needed. One thing’s for sure, Path optimizer lives up to its name!
Not convinced? Want to get more information? Go to this specific Trailhead Module