What they are not telling you — Hightouch Journeys

Three years ago, we launched our first customer journey at Castled.io built on top of the data warehouse. As a customer engagement platform, this was a crucial feature for us. It also held special significance as we initially believed it would be incredibly challenging to build a journey on top of a data warehouse with so many moving parts.

Since then, it has become the most widely used feature in Castled. So needless to say, I was intrigued when Hightouch launched customer journeys on top of the DWH last week. I have always believed that customer journeys are a useful addition to any CDP—composable or not—as a multi-destination orchestration tool.

So, why am I writing this post?

Since its launch, I have received numerous DMs asking if Hightouch Journeys could replace the customer journeys in Castled.io , MessageGears or even Braze and Iterable. I've also noticed several LinkedIn posts purely focused on solving cross-channel customer engagement use cases using Hightouch Journeys.

Having spent countless hours watching our customers create complex customer journeys, I immediately knew that much remains untold.

Why CDPs Cannot Replace Customer Journeys in a CEP (Customer Engagement Platform)

By definition, the role of a CDP is to be a single source of truth for your customer data and then sync only the "relevant" customer data to the third-party tools in your stack. For example, a CDP at best adds users to an audience in Braze/Iterable and does not compose email, push and in-app messages to send to your customers.

In short, a CDP lacks a comprehensive context of the actual use cases customers are solving with their data and hence fails to pull many key levers in multi-channel orchestration. Allow me to illustrate this with a straightforward example from a real-life customer engagement scenario.

Now, let’s explore how this journey can be implemented using Hightouch Journeys alongside a dedicated customer engagement platform - lets say Braze?

Challenge 1: Campaign Hell

To craft the engagement workflow mentioned above using Hightouch, we must create a new campaign in Braze for each messaging step—Email (I guess Hightouch has native support for email, so perhaps we can skip this), WhatsApp, Push, and in-app channels.

Please note that this is a simplified journey I created for the ease of explanation. However, it's very very common for marketers to create customer journeys with around 20-25 steps. Don’t believe me? This is an example of one a customer created recently on Castled.

Now, imagine the complexity of creating approximately 20 new campaigns in Braze for each messaging step of a customer journey, mapping them to corresponding steps in Hightouch, and maintaining them continuously.

No marketer I know would ever be okay with this.

Challenge 2: Inability to Create Basic Branching Conditions

More than 85% of the customer journeys created in Castled have branching conditions based on channel events — similar to the push clicked/discarded branching condition in our reference journey. I understand that most stats are made up, this one is not - I personally ran a query on our data warehouse few minutes ago to fetch this data.

Remember, Hightouch is just adding users to an audience in Braze and not calling a "send push" API that returns a callback when the message is actually clicked/discarded. Therefore, there is no way Hightouch or any other CDP can access the delivery events and, more importantly, map them accurately to the journey step that sent the message. As a result, they fail to create the most basic branching conditions growth teams create 85% of the time while building a journey.

Sounds like a deal breaker?

Challenge 3: Lack of Basic Step Level Metrics

All marketers who create customer journeys need step-level stats— for example, the number of clicks, opens, reads, conversions, etc —to see how their campaigns are performing and better optimise them for the future. Here are some of the stats we capture for every step of a customer journey.

Again, due to the inability of a CDP to accurately map the channel events back to the delivery step, it becomes near impossible for marketers to track step-level attribution. The only way is for them to track the actual campaigns in Braze against each step and check the stats there.

This is literally an attribution nightmare, and no marketer I have talked to in the last three years would build a journey without native step-level metrics.

I hope this clarifies that an effective multi-channel customer journey can only be fully realised by platforms that natively manage messaging channels. This leads us to the billion-dollar question.

Will Hightouch or other CDPs build native messaging themselves?

100%. For those closely monitoring the industry, the signs have always been clear. For those new to this discussion, here it is : CDPs and CEPs are on the verge of consolidation, or at the very least, significant efforts are being made towards making this possible.

However, I firmly believe that customer engagement is a deep dark forest, possibly more complex than an entire CDP. Many of us have been building it for years or even decades and still have not cracked it 100%. Like many other features, this cannot be built as a feature of a CDP or vice versa.

More on this in my next substack post. Thanks for reading ??