How To Pass the Einstein Analytics & Discovery Consultant Exam

Einstein Analytics has been an area where I was very reluctant to tread. I don’t have a computer science background and I never took a data science class or studied statistical analysis. I did have to takes one math class while earning my Bachelors of Music Performance and the class I took was “Math for the Everyday World”.

Hearing all of that you might be surprised that I've made it this far in the tech world. But I'm also really stubborn...which tends to come in handy in my learning process. So while it took me a bit longer than I had hoped to become Einstein Analytics proficient, I’m so glad that I persisted.

More than a year ago, a few of my team members and I went through several introductory training classes to try to "learn Einstein". Unfortunately, our main take away was that the interface was very foreign and that Einstein Analytics was much more complex than we had expected.  I knew that we didn't have the skill set to leverage Einstein at that point, but it went on my list as something I wanted to learn in the not too distant future. And thanks to Trailhead and a number of excellent community resources I passed the Einstein Analytics and Discovery Consultant exam last month.
​

About the Exam

Here are the subject areas that are covered on the exam. Do not skip over this part! Knowing how the different sections are weighted and the specifics of what you need to study in each subject area is the most important step in preparing for this type of exam. The full exam guide can be found here. 
​

Data Layer: 24%

• Given data sources, use Data Manager to extract and load the data into the Einstein Analytics application to create datasets. Describe how the Salesforce platform features map to the Model-View-Controller (MVC) pattern.
• Given business needs and consolidated data, implement refreshes, data sync (replication), and/or recipes to appropriately solve the basic business need. Identify the common scenarios for extending an application's capabilities using the AppExchange.
• Given a situation, demonstrate knowledge of what can be accomplished with the Einstein Analytics API
• Given a scenario, use Einstein Analytics to design a solution that accommodates dataflow limits.

Security: 11%

• Given governance and Einstein Analytics asset security requirements, implement necessary security settings including users, groups, and profiles.
• Given row-based security requirements and security predicates, implement the appropriate dataset security settings.
• Implement App sharing based on user, role, and group requirements.

Admin: 9%

• Using change management strategies, manage migration from sandbox to production orgs.
• Given user requirements or ease of use strategies, manage dataset extended metadata (XMD) by affecting labels, values, and colors.
• Given a scenario, improve dashboard performance by restructuring the dataset and/or data using lenses, pages, and filters.

Analytics Dashboard Design: 19%

• Given a customer situation, determine and define their dashboarding needs.
• Given customer requirements, create meaningful and relevant dashboards through the application of user experience (UX) design principles and Einstein Analytics best practices.
• Given business requirements, customize existing Einstein Analytics template apps to meet the business needs.

Analytics Dashboard Implementation: 18%

• Given business requirements, define lens visualizations such as charts to use and dimensions and measures to display.
• Given customer business requirements, develop selection and results bindings with static queries.
• Given business expectations, create a regression time series.
• Given customer requirements, develop dynamic calculations using compare tables.
• Given business requirements that are beyond the standard user interface (UI), use Salesforce Analytics Query Language (SAQL) to build lenses, configure joins, or connect data sources.

Einstein Discovery Story Design: 19%

• Given a dataset, use Einstein Discovery to prepare data for story output by accessing data and adjusting outputs.
• Given initial customer expectations, analyze the story results and determine suggested improvements that can be presented to the customer.
• Given derived results and insights, adjust data parameters, add/remove data, and rerun story as needed.
• Describe the process to perform writebacks to Salesforce objects.

​My Study AppRoach 

​Studying for this exam was a bit different for me than other exams. I didn’t have much of an existing framework to hang the different bits of knowledge I needed to absorb. It was truly like learning a new piece of software! So I started in the best place I can think of...Trailhead. But you’ll need more than just Trailhead practice to pass this exam (and to complete the Einstien Superbadges for that matter). Luckily there are some amazing community resources out there...and I’ll share my favorites with you in just a bit. 

I also participated in a 3-day instructor-led course taught by Salesforce that covered some of the topics I still was struggling with including loading and and transforming Salesforce (and non-Salesforce) data, modifying dashboard queries in JSON and leveraging  Salesforce Analytics Query Language (SAQL). Sometimes people ask me how I manage to study for exams in addition to my full-time job. This was a literal example of how I manage to “make it work”, with the 8-hour a day classes starting at 4AM so I could minimize the hours I would be out of the office. 

Could I have passed this exam without the instructor-led training? I’m sure I could have. But this helped fast-track my confidence around this relatively foreign tool. I also really lucked out in getting a great teacher who made the 4AM wake-up call for the class worth it. 
​

Where To Focus

• Data Flow vs. Recipe - what is possible with each?
• Edgemart vs Sfdc Digest - what's the difference?
• COMPARE TABLES ARE IMPORTANT! Not just from an exam perspective, but also from a functional perspective. They are often the starting point for the chart or dashboard you want to create. 
• Dashboard design....seriously. Do you know when to use a Heat Map vs a Bubble Map? What about a Treemap Chart? Brush up on all the chart types, because you will need to know this for the exam. 
• What can be modified via the Dashboard JSON file (you know you can get there via CME+E or CTRL+E...right?) vs. the Dataset XMD? 
• Oppty Stages (how would you go about transforming the order of stage names in EA)?
• GINI Coefficient...what is it?
• Einstein Discovery Writeback - what are the steps to set it up? What is it used for?
• Know how you can control sharing in EA (security predicates, flattening role hierarchy etc)
• Talking about security...what about encryption? Can you BYOK? Are datasets encrypted at rest?
• Do all datasets count against your dataset limit? What if they only run for a short amount of time?
• Sharing inheritance...how does it work? What are the limitations?

​My Top Tips

• Don’t start with the superbadge, especially if you are new to EA.
• Always start with the UI, even if you will need to use SAQL, get the skeleton for your SAQL query by configuration first.
• Don’t stress too much about SAQL - there aren't too many exam questions about it and in the real world, you'll often be copying and pasting the syntax. Plus, their is more and more that you can do without SAQL with each new release
• Formatting matters, know what you can customize via the UI and what you can't.
• EA is changing at a more rapid speed than the rest of the platform. This is an area to stay on top of with release notes. 

​My Favorite Resources

• Rikke Hovgaard is the founder of the #LondonDataTribe and her blog, salesforceblogger.com has a WEALTH of information about Einstein Analytics and Einstein Discovery. My EA teacher recommended this blog to me, and I'm so glad she did. 
• Do you prefer to watch videos? Peter Lyons, a fellow Salesforce MVP has you covered. His YouTube channel Let's Play Salesforce really helped me understand some of the core concepts. Warning: some of the SAQL videos are AMAZING but contain outdated syntax. I hope he will update these at some point because they are very useful. 
• My fellow co-lead Charly Prinsloo has a great guest post on her blog that reviews an exhaustive list of Einstein Analytics resources (some of them are only available to Partners) but it's a great reference.

​In Conclusion

Einstein Analytics is an incredibly powerful tool that continues to change and evolve much faster than the core Salesforce platform. While there is a bit of a barrier to entry for folks who are brand new, the team continues to make enhancements to the Einstein Analytics platform that make it easier to use and learn. It took me two attempts...but I am so glad I persisted to earn the Einstein Analytics & Discovery Consultant credential.