Choosing a movie to watch with friends can be a daunting task, filled with debates over streaming services, genres, and specific titles. This scenario mirrors the challenges of implementing Customer Journey Analytics (CJA), where the abundance of data can overwhelm users. To ensure successful CJA adoption, it's crucial to start with end-user needs, design mockups of necessary dashboards, identify required dimensions and metrics, understand data sources, and link them effectively. By curating data carefully and planning thoroughly, organizations can make CJA a powerful tool for improving customer experiences and driving business success.
“Let’s pick a movie to watch.”
If you’re anything like me, these words strike terror into your heart. Yes, it’s lovely to sit down with some movie snacks and enjoy some quality time with friends and loved ones, but you know what’s coming. First, there’s the argument over which streaming service to open. Will it be Netflix? Max? Wait, you actually have a login for Criterion?
Once you settle that, then the argument over genre selection begins. Everyone has to shout down that one person who keeps insisting on a horror movie, and the two people who insist that any time of year is good for a Christmas movie. Even if you agree on a comedy, then there’s a lively discussion about what exactly qualifies as a comedy, and Dave, I swear if you bring up how technically Dante’s Inferno is a comedy again, you will not be invited back.
Now you have to scroll through the comedy section and it’s endless. Someone is arguing for a classic Hollywood comedy, another person is yelling for a Brat Pack movie, and someone is loudly insisting on an obscure absurdist comedy put together by veterans of a barely known improv group that had two seasons late at night on MTV in the ‘90s. Me. That’s me. I’m insisting on that.
And now you know why CJA implementations can get into trouble. Just like there are now overwhelming numbers of movies competing for your attention, some CJA implementations are ingesting what turns out to be an overwhelming amount of data.
CJA is an amazing tool. I love it. I’ve been on teams that carried out migrations, pitched more, and worked with CJA on completed migrations, and the possibilities are endless. There are no more fixed limitations on the number of dimensions or metrics. You aren’t just restricted to clickstream data. You can pull in datasets from anywhere! This is the value proposition, but this is also a temptation that can lead you to having a tool no one can effectively use, because while the metrics and dimensions are vastly expanded, the attention span of your team is still limited.
In my years of working with Adobe Analytics, I’ve trained hundreds of people. I can teach them to create a good dashboard in a few hours. They always feel very capable. It’s easy! The Analysis Workspace UI is simple and intuitive. That feeling of ease only lasts a little bit, which is why I hold office hours daily for two weeks after training is complete. The questions that follow are never about using the Analysis Workspace UI. They’re always questions about what dimension or metric they need to build their freeform tables. The ones who master Adobe Analytics are the ones who obsessively reference the data dictionary as they’re learning, which is a hard task.
That’s one data set. That’s one data set that’s in their area of domain expertise, the single implementation they focus on and have been working at for eight hours a day for years.
You are now expecting CJA users to incorporate far more. It’s a good expectation. If they can do it, they’ll improve the customer experience, reduce calls and chats at your contact center, make a ton more sales, personalize more effectively, and achieve all kinds of things! They will, but only if you make it easier for them to adopt CJA.
To return to the metaphor we started with, you have to be a bit more like an old-school video store and curate your data carefully based on demand (and keep doing that with even more carefully curated data views for different groups of users).
The first thought most people have is to design an implementation like this:
- Pick data sources
- Find or implement a connecting identifier
- Pull data in
- Create dashboards
This is a totally reasonable method that will have you bringing in every potentially useful piece of data, including one of your dev’s Spotify playlists. Instead, work backwards from the identified use cases, following this advice:
Put together mockups of the dashboards end users will need.
Starting with the end users to design is always the way to go. They’re the ones who are using this and will make your investment in CJA successful or not. Giving them a chance to think through how they’ll use the data also plants the seeds for successful use after implementation occurs. This also gives them a hand in designing the implementation so it’s their implementation, rather than a surprise sprung on them.
Determine what dimensions and metrics are needed to create those dashboards.
Now you’re ready to build your data dictionary. You’ll put in the names that your end users are expecting and have it ready for review by them. Again, this is the hardest part of becoming a Workspace expert. You’ll give them a head start on it before implementation even begins.
See and understand where the source data comes from.
This sounds obvious, but without the steps before this, there’s a good chance you’ll pull in a data source that is duplicative of another or that everyone thought had useful data, but actually doesn’t. If you decide on the data sources before determining the data needed for end users, you won’t know until you’ve burned through a lot of hours that you made a mistake. Any company that can use CJA effectively usually has so much data that has been gathered for different purposes over the course of so many years by so many people that no one will know all of it.
Figure out how to link data sources together through identifiers that exist or will be implemented.
This is the final step of the planning process. Figure out how to get everything linked up. Again, it’s pretty obvious, but at the same time, if you haven’t figured out the data you need and where it lives beforehand, you’re looking at rework or wasted work.
Now that you’ve got your CJA implementation well planned, you’ll be ready for your end users to hit the ground running. You’ll eliminate a lot of unneeded data. You’ll save yourself a ton of time, which you can use to argue over what movie to watch.