Hey Smart Marketers!
Welcome to these Summer Series about Data Cloud!
Before jumping in, I just wanted to let you know that I’m conducting a survey about Salesforce Data Cloud to validate/deny some assumptions I have. I would really be grateful if you could take 5 minutes to fill it out. Here’s the survey link.
Now let’s get back to today’s topic!
It seems like Agile has won the war…
Can’t think of a company I met recently that doesn’t have its daily meetings, product owners, scrum masters and so on…
As if everything could be done incrementally.
We already know that Salesforce Marketing Cloud Engagement is not really Agile-friendly… but let me tell you that Data Cloud is worse.
When it comes to implementing Salesforce Data Cloud, Agile is like trying to fit a square peg into a round hole.
Let’s dive in!
Salesforce Data Cloud is Not Your Average Project
The whole point of Data Cloud is to build a 360 view of your Customer Data.
We’re talking about pulling together information from across your entire organization, cleaning it up (before injecting it to Data Cloud…), governing it, and then using it in real-time to make smart decisions.
This isn’t something you can just “iterate” your way through.
Take a look at the official Salesforce implementation guide…
Trying to cram this into Agile’s framework of two-week sprints and Jira tickets is just a recipe for disaster.
Why Agile doesn’t cut it
- 2-week sprints are for tactics not strategy:
When you start your Salesforce Data Cloud project, you need a crystal clear strategy.
You need to define how your data will flow, who will access it, and how it will be used across your entire organization. Agile’s incremental, “we’ll figure it out as we go” approach doesn’t work here.
If you don’t get the big picture right from the start, you’ll end up with a disjointed mess of data that no one trusts.
- Data Governance isn’t Agile
Handling customer data is serious business.
With regulations like GDPR breathing down your neck, you can’t afford to wing it. Data governance needs a meticulous, well-thought-out plan.
- Integration is Complex
And the keyword here is Integration. Salesforce Data Cloud is here to break data silos. You can’t do this with teams working in silos on their little piece of the puzzle.
You need tight coordination.
You’re not just pulling data from a couple of sources: you’re integrating across systems, departments, and even 3rd-party platforms. If you don’t want to set yourself up for a technical Frankenstein, you need proper alignment.
- Change Management needs more than a Backlog
You think working with Unified Profiles and setting Identity Resolution rules will just smoothly become part of your marketer’s life? Wake up!
People need to be trained, processes need to be updated, and everyone needs to get on the same page.
You need a real strategy for change aligning with your overall strategy for this project. This goes beyond user stories.
The pragmatic way
So, what’s the alternative?
Let’s be real here: Data Cloud is new and evolving as hell from one release to another. The truth of today might change tomorrow.
But here’s a more pragmatic approach:
- Start with a Solid Plan (Yes, Like Waterfall)
First, lay down a clear, detailed roadmap: Define your data strategy, map out your governance framework, and figure out how everything will fit together.
This isn’t the time for “we’ll figure it out later.” Get it right from the start.
- Use Agile Where It Makes Sense
Once your strategy is in place, use Agile for the parts of the project that benefit from iteration—like creating Calculated Insights or adding external Activation Targets.
Remember: there will always be work to do on Salesforce Data Cloud so your first Agile Sprints are laying down methodology and task types that will be used once in Run phase.
- Regularly Sync Up and Test
Integration isn’t a one-and-done task. Set up regular checkpoints where all the pieces come together, and test, test, test.
I know we hear a lot the word “real-time” when it comes to Data Cloud but truth is: Idendity resolution takes time., ingesting data takes time, segmenting takes time… Bringing all pieces together may not go as well as you thought… speaking of this, you might want to take a look at this free appexchange to orchestrate actions.
- Change Management is Critical. Handle it accordingly.
The Bottom Line
I’m still building my framework for this kind of projects so your feedback would be appreciated. Also, Salesforce is running Implementation cohorts which I’ll attend and maybe it’ll be enough… not sure though.
See you next week!
Other ways I can help you: