We have often heard that campaign and journey visual builders (e.g., Braze’s Canvas feature) are attractive features in conventional CRM and CEM platforms. But having paid close attention for a long time, we’ve observed that the most enthusiastic marketers are the ones that haven’t used the Canvas-type builders extensively. More importantly, we noticed that lifecycle teams at larger companies seem to increasingly experience those Canvas builders as a burden. They are A) Difficult to maintain and B) Very difficult to move beyond - the extensive canvases make innovation in a dynamic world extremely difficult.

I’ve been analyzing how large enterprises use these canvas builders - we can use the specific example of Braze’s Canvases, which is how Braze lets you set up repeatable journeys - they're reusable messages that get sent under certain conditions.

I’ll use the canvases of one enterprise app as an example. This app operates in 6 different countries. The CRM teams has 30-50 Braze canvases per country—over 200 in total. For comparison, this is less than one-fifth of what we've seen with other customers, so it’s a lot, but nowhere close to the most complex 

I looked at the users those canvases were sent to over the course of a week and calculated the number of users each pair of canvases had in common.

Look at the relationships between different canvases.

All the canvases from one customer.

Notice a few things:

1. The countries are clearly represented in clusters (different colors = different countries).

2. The dots with black rings around them are transactional messages - things a user just gets as a matter of course if they do something on the app. For example, you get a "purchase completed" notification if you buy something.

3. Three of the six countries don't have those standard transaction messages. (These kinds of gaps are common - it’s hard to keep track of everything going on, and sometimes teams just aren’t able to find the time to ensure 100% coverage on even their existing journeys).

4. Most of the non-transactional messages for each country basically go to the same users. These are different flows, but the tight clustering shows that the same people are going through those flows.

5. The few canvases that don't fall within the central cluster for each country aren't all of the same number and the same distance from that central cluster. Even when a canvas does depart from the normal patterns, it does so differently for each country. If it's a good enough practice to implement at all, it should be good enough to let all users benefit from it.

Maintaining this colossal inventory of canvases is a nightmare.

Most of a canvas-centric approach results in a sprawl of nearly identical canvases. That sprawl makes it difficult to see where you've failed to implement an important flow for a particular subset of users. But more than that: the sprawl is so hard to manage that it traps you into making small improvements (or even just maintaining a static state) for a small subset of flows, when you should focus on discovering opportunities for new flows.

Sure, an abandoned cart flow is important, but every other app has an abandoned cart flow. So just having one isn’t enough. When you have more things to manage and maintain, more of your energy goes into managing and maintaining those basic things, and less of your energy can be spent on differentiation and other more important things.

One last thing: all of these flows are the same every time. Every time you abandon a cart, you get the same message. Because a canvas forces you to spend all your efforts just managing the flow, rather than building a rich messaging inventory.

Legacy manual user journey builders (like the Braze Canvas) force you to focus all of your efforts on maintaining the administrative elements of automations, and that creates blind spots. They require you to manage complexity rather than reduce it.

Want to see the alternative to elaborate and complex canvases you have to manually manage? Learn more about Aampe!