Einstein Prediction Builder and Einstein Discovery – What is the difference, and which one is for me?

Guest post by my esteemed colleague, Arun Arunagiri.

Salesforce’s Einstein adds intelligence to its CRM via a suite of products and they continue to evolve on what is essentially a machine learning platform behind the scenes. A variety of products such as Einstein Next Best Action, Einstein Voice, Einstein Bots, etc., on top of Salesforce clouds/apps that are already intelligent, have made Salesforce not only the #1 CRM, but also the smartest CRM. Einstein Prediction Builder and Einstein Discovery are two key products in their suite that exist for different reasons. Or do they?

Einstein Prediction Builder

Picture 1

Image credit: Salesforce.com

Einstein Prediction Builder is a declarative tool that helps fast-track predictions based on Salesforce fields. This will help a Salesforce admin to custom-build predictions on any object via few clicks on a visual interface and power workflows and apps using the newly-found AI.  If your company has no in-house expertise nor the budget to invest in machine learning, Einstein Prediction builder is your best friend. You don’t need a data analyst nor scientist to bring intelligent insights and enable the business to make smarter decisions based on future outcomes.

Two typical use cases for Einstein Prediction Builder include predicting churn rate – the percentage of customers who are likely to stop using your company’s product or service – and predicting the housing prices in the volatile Australian real estate market. The calculated scores can be easily seen, embedded in lightning pages and actions based on them performed via workflows.

Einstein Discovery

Picture 2

Image credit: Salesforce.com

If Prediction Builder is meant for Salesforce admins to fast-track predictions, Einstein Discovery (ED) will delight a data expert by crunching millions of data combinations in minutes, helping to gain deeper insights with unbiased explanations, predictions and recommendations, at a granular level of visibility of model accuracy. ED works on Einstein Analytics datasets which means that one can digest data from other popular databases such as SQL, SAP, Hadoop, etc., in addition to simple CSVs.  The ED Story, which is a comprehensive statistical analysis of the data, will help one stay on top of the game by gaining detailed insights and fully understanding what the data is telling us: what is happening (Descriptive), why is it happening? (Diagnostic), what is likely to happen (Predictive) and what do I need to do (Prescriptive)? It is normal for an ED Story build to be an iterative process where refinements and improvements are made and the Story is rerun.

A typical use case scenario is of a business CFO looking to make meaning out of a sudden plunge in profit margins and a CEO urgently looking to remediate the situation, with a large volume of data at hand with complex interdependencies between variables impossible to be analyzed manually. ED comes to the rescue!

4 Types of Data Analytics (descriptive etc)

Image credit: www.principa.co.za

Data is the key challenge

There is a common challenge in both prediction builder and ED – ensuring data is fit for the purpose; not just in volume, but also in other characteristics (unbiased, balanced, etc). If data is an issue, it is back to the drawing board as Business Intelligence (BI) should streamline the data collection and analysis before Artificial Intelligence (AI) can make predictions. By far, majority of the time spent in prediction initiatives will be in collating and cleansing data – it cannot be overemphasized how crucial this is. Garbage in, Garbage out!

What the future holds

Salesforce is doing its best to help Admins and developers leverage the AI capabilities without breaking a sweat. Point-and-click is the trend and there will be less coding. Their array of products intend to make every part of the business smarter and bring actionable insights and smart decision-making capability without the need for hiring data specialists/scientists. At the time the Tableau acquisition was announced, Marc Benioff made a statement that summed up their vision and commitment to AI and Data:  “In 2020 the world will generate 50-times the amount of data it did in 2011… tapping into this massive data flow creates huge opportunity for economic and human advancement.”

References

Einstein Prediction Builder and Discovery Best Practices (video): https://www.youtube.com/watch?v=yFQVfCU1j6Y

Discover deep insights with Salesforce Einstein Analytics and Discovery (Slides): https://newdelhisfdcdug.com/salesforce-einstein-analytics-and-discovery/

Artificial Intelligence and Salesforce Einstein — Part 2 (Products and Use Cases): https://medium.com/@jayantkjoshi/artificial-intelligence-and-salesforce-einstein-part-2-products-and-use-cases-72148bc1b9f

Trailheads:

https://trailhead.salesforce.com/en/content/learn/modules/wave_exploration_smart_data_discovery_basics/wave_smart_data_discovery_your_data_scientist

https://trailhead.salesforce.com/en/content/learn/projects/prediction_builder/prediction_builder_prediction

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s