Welcome to the new data platform

Something new is coming to Einstein Analytics. Something you should pay attention to as it will change the way that you work with data. The most visible part of this new data platform is a new Data Prep editor. It’s more approachable and powerful featuring machine learning capabilities. Hints have been dropped at Dreamforce and other places, but now the time has come to welcome the new Data Prep editor as the Open Beta is coming in Summer 20, which is only weeks away.

Today, Einstein Analytics has a few options for generating your datasets; dataflows, the dataset builder and recipes. For those of us that have been working in the Data Manager and know it well, know when to use what tool and accept some inconsistencies in the dataflow. But when you are new to Einstein Analytics the Data Manager seems daunting; where do you start, what do the nodes do and the list of questions continues. Just the other day I had a call with a new user where I was trying to help with a few nodes that extended into debugging the dataflow – in this conversation I heard “I don’t understand” several times. Without an experienced user’s help, this could have taken hours and that is what all new users are facing. The Einstein Analytics product team has acknowledged that:

1. there are gaps in the data prep tools,
2. the entry for new users is difficult and long.

This is why the Data Prep 3.0 project came about and you will see it in the Data Manager as “Data Prep (Beta)”. The project is aiming to bring a more powerful and robust tool to the Einstein Analytics users as well as making it more approachable for new users to get their data just right for their dashboards. Oh, and in time of course have just one editor for all things data prep and dataflow.

What is coming?

It won’t be a big bang and the new Data Prep replaces recipes and dataflows, instead, it’s a phased approach starting in Summer 20 with the Open Beta. What this means is everyone with Einstein Analytics can use the new editor when settings have been enabled by your administrator. It also means that some kinks are probably still being worked out and the team is looking for your feedback – so make sure to use the feedback button in the editor! It also means that not all the features and functions from dataflows will be available from day one, but more will be added with every release. Of course, you can do everything that you do in Data Prep today and then add some more cool things, too like sentiment analysis and an output connector. Some highlights to expect that a lot of users have asked for is:

• A new graph layout that combines the best from the recipe and dataflow
• A flow overview with clear icons and colors that represent the different nodes
• Preview of the data as it moves through the transformations (computed or calculated values)
• The flow provides an easy overview (source and output) and lets you add new nodes without the high risk of errors
• Packaged machine learning capabilities ready to be used on your data

Data Prep won’t include everything from the beginning, so you probably have a lot of questions around features and timeline. Below you can see the high-level timeline including the features for the new Data Prep editor (remember the forward-looking statement applies*).

What does this mean for users?

Obviously, we are looking at a big change in Einstein Analytics, and changes mean we all have to adapt. So as an Einstein Analytics user, you should consider the following:

• Read the release notes carefully when they become available to understand the details of what is changing.
• Evaluate if new requirements can be delivered with data prep, or if you are missing key features from the recipe or dataflow. Eventually, dataflows will open in the new Data Prep editor as well, so why not start leveraging Data Prep when possible?
• Accept that some requirements should be approached in a different way than you are used to. For instance, if you are used to working with case statements in computeExpressions you might experience that it can be solved with a bucketing transformation.
• Constantly evaluate if dataflows should be migrated to Data Prep once a migration tool for dataflows becomes available. In a matter of time Data Prep will be more powerful than any of the other data prep options, make sure you are not missing out.
• Use the feedback button within the Data Prep to let the product team know what you think or if you want to start a discussion use the Trailblazer Community.

Stay tuned, there is more to come

Summer 20 is just around the corner, so Data Prep is a near reality for Einstein Analytics. There are so many things to be said about Data Prep, which cannot be covered in one single blog, so stay tuned as we do a deep dive in the Data Prep editor covering (subscribe to the blog on the right, if you want to be notified):

• Data Prep feature deep dive
• Native Machine Learning capabilities
• What Data Prep means for different users
• Behind the scenes of the new data platform
• End-to-end walkthrough of a use-case

Finally, comment below if there are specific topics or questions covered, let’s see if we can’t answer them in the coming blogs.

*Forward-looking statement

This content contains forward-looking statements that involve risks, uncertainties, and assumptions. If any such uncertainties materialize or if any of the assumptions proved incorrect, the results of salesforce.com, inc. could differ materially from the results expressed or implied by the forward-looking statements we make.

Any unreleased services or features referenced in this document or other presentations, press releases or public statements are not currently available and may not be delivered on time or at all. Customers who purchase our services should make the purchase decisions based upon features that are currently available. Salesforce.com, inc. assumes no obligation and does not intend to update these forward-looking statements.

Written by