Salesforce November 6, 2017 – Everything around us is getting smarter, but that is not necessarily true for our business processes. Employees need to be freed from one-size-fits all tools that leave them looking for answers and to be freed from the repetitive tasks that take up their days. However, companies are hindered by the infrastructure costs, lack of expertise and the resources required to optimize their workflow with AI. That all changes today with the announcement of myEinstein, which empowers admins and developers of all skill levels to build AI-powered apps with clicks, not code. With myEinstein, the employees who are actually managing and driving business processes have the power to build and customize AI apps to fit their specific needs, paving the way for everyone to be smarter and more productive.

Builder

With simple clicks, the declarative setup guide walks users through building, training and deploying AI models using structured and unstructured Salesforce data. It sounds easy — and it is — but behind the scenes, myEinstein automates the model building and data scoring process in creating custom predictive models and bots. Admins can then easily embedded directly into Salesforce workflows. Models automatically learn and improve as they’re used, delivering accurate, personalized recommendations and predictions in the context of business. myEinstein services include:

  • Einstein Prediction Builder allows admins to create custom AI models on any custom Salesforce field or object to predict business outcomes, such as a customer’s likelihood to churn or the lifetime value of an account. For example, an admin at a financial services company can build a predictive attrition model to understand which customers are most likely to churn based on all Salesforce fields related to the account, such as the number of external accounts they link to their checking account and history of customer support calls. Using a declarative, point-and-click setup tool, admins define the prediction, identify the field to build the model on and which data to use. The attrition score based on these signals can be directly embedded into the Salesforce customer account page and create a task, automatically alerting reps to high-risk customers so they can deliver personalized, one-on-one interactions to reduce attrition risk.
  • Einstein Bots enable admins to build, train and deploy customized service chatbots with just clicks. Using natural language processing, Einstein Bots can communicate with customers, freeing up customer service agents to focus on more complex inquiries. Einstein Bots are connected to customer service workflows and trained with historical service and CRM data to quickly and accurately respond to common customer inquiries, resolve routine issues, and seamlessly handoff to customer service agents, if required. Both agents and customers have the ability to provide feedback to the chatbot models, which continuously learn and improve to deliver more accurate and relevant information. For instance, a Salesforce admin at an ecommerce company can deploy Einstein Bots to handle common service issues like tracking order status, resetting a lost password or initiating a return. The chatbots understand what the customer is asking and can collect basic information, initiate the appropriate workflows and deliver personalized, relevant information to the customer.
  • myEinstein Extends the Power of AI to Developers As the demand for intelligent apps increase, the speed at which developers need to build skyrockets. myEinstein also includes Einstein Language and Einstein Vision, a powerful set of APIs launched in June 2017 to empower developers of all skill levels to build AI-powered CRM apps that business users need. With Einstein Language for Intent developers can train a model to classify the underlying intent of customer inquiries in order to automatically route leads, escalate service cases, personalize marketing campaigns. Einstein Language for Sentiment enables developers to classify the tone of any text as positive, negative or neutral so that companies can quickly gain insight into customer attitudes and then take appropriate actions. Developers can also leverage computer vision with Einstein Vision for Image Classification, which brings visual search, brand detection and product identification capability to CRM apps; and Object Detection, which enables developers to train models to recognize multiple unique objects within a single image, as well as the location, size and quantities of those objects.We can’t wait to see what our trailblazer community builds with myEinstein. You can learn more about these new features this week at Dreamforce 2017 or by taking the Einstein Prediction Builder trail.