(All views expressed here are personal. These are general tips for this exam. Please do not ask for any questions)
This is my first blog post since joining Salesforce in April 2018. It has been a life changing experience so far to say the least – but that is a story for another time.
I have had the good fortune of getting some exposure to Einstein Analytics (EA) and Discovery (ED) over the last few months. But as an architect, this involved understanding these offerings at the architecture level and not at the solution level.
Also, I had to learn Einstein Discovery before I learned Analytics in detail. But as you know, Discovery is now available in Einstein Analytics and the older product is called Discovery Classic.
So I was really stoked to see a new consultant exam for Analytics and Discovery and was really super keen to study well and be one of the first ones to get it thereby also ending my self imposed certification drought.
It is really a great feeling to get a brand new cert by navigating uncharted waters. I did it for most of the designer certs previously and I really missed that challenge.
My background in this area was about 1-2 months of high level experience in Einstein Discovery followed by a lot of self teaching by reading articles and doing hands on exercises in Trailhead which I will list below.
I had to learn Einstein Analytics from the ground up and in hindsight, that was really good in getting a well rounded understanding of this topic. I spent around 1 week on this by spending time after work and the weekends burning through the trailhead modules (total of 2-3 days) while my wife and cat (The lovely Aida pictured below! ) helped me keep my sanity.
At the end of it I did rack up a lot of badges and points so that was great.
I have to say that I really enjoyed working through the trailmix and realized that I love the analytics offering. I plan to continue to learn more about this topic and perhaps specialize in it once I achieve my CTA (That holy grail I am after! #JourneyToCTA!)
So, here is the official study guide for this exciting new certification.
As most of the other exams, there are 60 multiple choice questions. Usually, there a few additional questions added which do not affect your score but in my exam, there were only 60 questions. With a passing score of 68%, you need to get about 41 questions right. That might not seem too bad but I need to warn you that this exam was a little tricky and I nearly took the allotted time to answer all questions and do 1 round of review. At the end, I was somewhat confident of passing and I was so relieved to see that I did!
So if you do not have experience in this topic then it will be a challenge to pass this. Therefore, study and get a lot of hands on experience.
Luckily, Trailhead has all the info you need and you can also get a special Einstein Analytics(EA) developer edition sandbox where you can access all the cool features and try them out.
So here are the main resources for the exam.
- The quintessential Trailmix – Please work through this. Maybe do all the modules first and then attempt the projects. Based on my experience, if you do the trailmix and understand all the concepts well then you will pass and that’s that!
- The superbadges in the trailmix above is optional. I have not done them yet but will do so whenever I get the time.
- There is also another learning resource where all the relevant articles and trails have been organised in a learning map which can be accessed here. I highly suggest following this to understand the end to end flow of data in EA.
http://www.einsteinanalyticslearningmap.com/ - Understand how data is brought into Analytics – Dataflows, Datasets, Dataset builder, full sync(What is data drift???) , incremental sync, options to integrate with Salesforce/other systems.
- Data flow transformations are super important!!! – Some of the key ones to know are: augment, computeExpression, computeRelative, digest, edgemart, flatten, sfdcDigest, sfdcRegister. These would seem a bit hard to understand at first but keep at it and you will be fine.
- Understand the Analytics External Data API and its use cases.
- Understand the two ways of managing security in EA (Sharing inheritance v/s predicate expression to implement row level security)
- Know to set up EA and access it in Salesforce. Learn basics of creating an app and a dashboard, lens etc. You will need to know the basics of setting these up.
https://help.salesforce.com/articleView?id=bi_help_setup.htm&type=5 - Learn about the license and permission set assignments for EA. These have undergone changes last year so pay attention to this.
- Learn about EA and ED Limits.
- You will need to know the different templates available with analytics and how you can package and distribute your analytics apps.
- Learn about the options to deploy EA changes.
- Read up on the best practices for building your own dashboards
- Look up “Progressive Disclosure” in EA
- You will need to know SAQL. Just the basics should be fine. Know where SAQL queries can be issued or modified in EA.
- Pay special attention to the time series SAQL function with all its arguments.
- Bindings!!!! I have to say that I struggled with this topic. But spend the time on this and learn the two types of bindings (Selection v/s result binding) and when to use each.
- Also for bindings, understand what a step query is and where is the filter specified (source step or target step?)
- Spend time reviewing this for Bindings and read through the JSON step examples.
https://developer.salesforce.com/docs/atlas.en-us.bi_dev_guide_bindings.meta/bi_dev_guide_bindings/bi_dashboard_bindings_wave_designer_use_case_query_parts.htm - XMD is another important but easy topic. Learn what XMD could be used for.
- Learn about the different types of Charts and when to use which one. There are tons of charts, so you will need to remember quite a few of these. Create a few of these and use the projects in the trailmix to help you.
- What is the purpose of the dashboard inspector?
- Learn Einstein Discovery – Stories, Models,Limits, Optimizing the data set for better predictions
- How do we deploy models to Salesforce and display them in an object.
- How to improve your discovery models?
- Read about GINI co-efficient wrt to ED.
Tips:
- Please do not attempt this exam unless you have good experience with EA. You have to go through all the important topics and try them out in your DE.
- Study smart by focusing on the exam outline. Based on the weighing below, pick and choose the topics that you are most comfortable with.
- Data Layer (24%),
- Security(11%),
- Admin(9%),
- Design(19%) ,
- Implementation (18%),
- Discovery Story Design (19%)
- Focus on Discovery but keep it for the last as its a much smaller topic than EA.
- One of the things that worked for me is this little known resource. Everybody has their own study style that suits them. For me, it is to quickly read through the awesome Salesforce documentation for a topic that I need to learn. This is a little outdated but I used this to quickly revise all topics the day before the exam and it helped me immensely. It seems like a lot but you will realize that you can skim through this very quickly.
https://resources.docs.salesforce.com/214/latest/en-us/sfdc/pdf/analytics.pdf - For the EA Trailhead modules, it is very important to not just read the content but also do it in your EA developer sandbox.
I feel that since the resurces on Trailhead are so awesome now, it will be easy to pass this exam once you finish going through that and cover the topics listed below.
So all the best with your exam and please do let me know if you have any general questions about it. I would also love to hear if this helped you in your preparation. So definitely do reach out to me on Twitter or LinkedIn!
Live long and prosper!!! May the force be with you!!!
Wait! Did I just mention two separate franchises?
