Salesforce Hacks
And Advice from Twistellar
[OCTOBER 2021]

Salesforce is going to deactivate Workflow & Process Builder automations in the Winter '23 release, which means that most declarative business rules will have to be migrated to the Flows representation.

To facilitate this change, Salesforce is going to introduce a couple of migration tools in the next year so all customers have time to update their configuration before the final retirement.

The current flexibility and usability level of Salesforce Flows already allows building more complex and sophisticated business automations. It can combine multiple flows and subflows to implement tricky and multi-choice execution scenarios, manage and render screens for system users to avoid building custom components, send notifications and invoke Apex for managing specialized actions like making API calls and running scheduled asynchronous jobs on large data sets.

We recommend spending some time evaluating existing automations in your system and defining the best transition combining actions for your Workflows and Process Builder Flows. Feel free to contact us if you think you need professional help with these activities or want to get the most out of your system setup.

This retirement is just another chance to review and optimize your business logic, so go with the Flow! Please find the decision table and the link to learning materials below:

Sometimes it can be required to identify text on the documents of your customers or partners, such as PDFs, scans, or pictures. In this case, you have to decide which service to utilize and how to connect it to your system as seamlessly as possible.

With Salesforce Einstein OCR, this task can be resolved in a few simple steps. So if its accuracy is enough for handling a chosen case, it becomes the best fit for the system as an OCR extension.

Let's examine it with a real example. We delivered the extension for the Salesforce Mobile application during one of the previous system implementations. It let the agents upload photos of driving licenses and unformatted document pieces that needed to be attached to the business cards of their customers and spare parts orders.

Einstein is a great solution for such a case because it has predefined models for recognizing data on such documents as driving licenses, passports (few countries supported), resident cards of the US, and employee pay records.

It supports tables and free text, so you can transform images into text lines and automatically import them into the core system, e.g. the Salesforce Sales/Service Cloud. After that, you can transform and distribute them between other clouds like Marketing or B2C Commerce.

Einstein OCR works with PDFs as well, so you don't need to manually copy text from your PDF files and put it into the searchable data.

This solution might be a good fit instead of JS-based OCR solutions or pricy third parties.

Learn: