Saravana Amuthu

Hi, I'm Saravan Amudu. I have a total of 6 years of experience working in the Salesforce ecosystem. In my previous experience, I've worked with companies like Accenture and Cognizant on multiple projects. I've worked on Salesforce configuration and went into customization as well. I've worked on building objects, fields, other automation tools like workflow process builder and flows. I've also worked on security settings like object-level security, field-level security, and all-level security. In one of my client's projects, I was involved in maintenance releases. I've successfully deployed 25 plus releases. These maintenance releases are nothing but small enhancements, which involve configuration as well as customization. So, what I do in those maintenance releases is that I start by analyzing the requirement from the client. We have a deck prepared by the business analyst. I go through the deck, and I understand the requirement. Then I specify the components that are needed to be built along with the efforts needed to build it. So, after that, I build. I push it to QA and UAT using AutoRabbit. After that, I work closely with the user for user acceptance testing, which is UAT. Then I use AutoRabbit for deployment. In my current role with the company, I'm mostly into the support side of Salesforce. We daily deal with requests and queries or some issues from other teams which are using Salesforce within Folio, like the sales team, finance team, etc. We handle the cases raised by them or the emails put to us. We reply to the emails. We have a separate case management system within Salesforce. Apart from that, I'm involved in minor enhancements, which follow a joint methodology. So, I'm also involved in solving CPQ-related questions as well. I've provisioned prospecting tools like Outreach, LeadIQ, ZoomInfo for the sales team. Currently, I'm based out of Bangalore. I work mostly in APJ timing, but I'm comfortable with any kind of job. Thank you.

Okay, so to handle customer queries including email, chat, and phone calls, for emails, we can use email to case, which is already in Salesforce. We can enable it, and we can put a deal on that. And whenever the customer sends an email to that particular deal which has been set up, it gets created as a Salesforce case. For chat, we can use the chatter function, which is already available in Salesforce. We can assign the user with the chatter license. And for phone calls, I'm sure that there is something available over Service Cloud, which requires a license, which I am not sure about, which I have not worked on. So I'll need to look more into the phone calls.

I haven't really worked with Salesforce integration with third-party systems. I'll need to look into the Salesforce APIs.

Okay, so whenever there is a new sales process or release, we'll have the release notes, and we can circulate those with the internal Salesforce team first and come up with any blockers with the existing implementation. If there's something that's going to block the business, we can send out a communication to the stakeholders saying there's something new implemented from the Salesforce side, and we'll need a certain number of days to get it changed or upgraded. And after that, we can break the entire requirement into user stories and assign it to the team and work on it, which goes for your agile process or whatever the existing process is. It should be done first in the dev sandbox, then in UAT, and finally deployed to production.

So I used to work with the finance team, and their major requirement is to track the approvals and the time needed for each and every approval. So what is the time lag between each and every approval? In my company, there's a department that handles credit-related issues. So they're mostly looking forward to the approval process. But what they really look for is the approval process and the time required for each step. What's the time required to complete it, and where is the delay? They're looking forward to cutting off the delay. In this case, the approval process is related to a credit check, but I can provide the approval process steps and dates for the number of days elapsed between each step. But I can give you the credit check number or the record number specifically. However, I cannot give you all the information of the related record. Let's say the credit check in this case. I created a report with the steps, one for each step, and the days it took for the user to complete the step with the corresponding credit check. This report was a little helpful, though it didn't match the exact requirement. I told them that they could look forward to the days elapsed, which was the major priority for them than the information present in the credit check. The users were okay with that, and it reduced their time by at least 60%. Previously, they were manually exporting all the records and comparing them to their own Excel sheet, and everything. Using this report, it reduced their time by 60%.

Okay, process builder. I would suggest going with flows would be a better option, or using an approval process separately would be a better option because process builder is kind of getting in the way. So flows would be a better option when we are going for multistep approval, or we can use a standard approval process for discount requests. In my company, we use Aptis approval request, which is more helpful for multistep discounts in coming to billing or sending out order forms to the customer. But coming to some things related to credit checks or so when the customer's looking to buy some product on credit, we use a standard approval process.

Okay. So we can either do via profile or permission check. So for new hires into the sales department, we usually have a process where when they are onboarded, by default, they get Salesforce. And after that, to set up correct permissions within Salesforce, we compare their portion over Workday and take the permissions of their existing coworker and copy the same permissions for them. So this would involve copying the same profile for the new hire who is getting onboarded to the same team and also other permissions that are already existing. So we take a different user who is already existing in the team and copy the same permissions for the new user who is getting onboard.

Profiles and roles are completely different. Profiles are mostly into object and field level security, and roles are into record level security. So, to give a visibility of a field or control over an object, we use profile. For example, there is an object on account, and there are 10 fields. A user needs control over 5 fields, but should be able to edit only those 5 fields and have read access over the other fields. Then we'll assign read access to all the fields through profile and give edit access to those 5 fields. For instance, let's say there is a manager and there are 5 subordinates, and the manager should have access to edit all the fields in the particular record and also see their subordinate's records. In this case, the manager will have a manager profile, and the subordinates will have a team member's profile. Team member's profile will have read access and edit access to certain fields, which they are supposed to input their inputs from their side. And the manager can also edit and see. We go for two different profiles. For roles, the manager will have a manager role, and the team's subordinates will have a team member's role. So, team member's role for that particular object will be set to private for team member's role, and the manager will have access via hierarchy. He can see every team member's record, but the team member will see only their own records because it is set private. So, by this way, roles and profiles are helpful.

Okay, just getting updates and deletion while preserving the tag. Okay, so coming to updates, basically, we use tools like data loader or data inspector. So we would always have multiple demo operations going on in objects. So whenever something is changing a record, depending on the condition, it could be from flow or process builder or it could be from triggers. So it's always wise to reduce the batch size so that we don't try to update the record again and again and get into multiple DML error issues and complete the operation. Coming to deletion, it really depends on what all data I would like to delete. Maybe the incomplete data can be deleted. So this is my understanding. I'm not really sure about this.

Mitigate government limits. While doing data migration, the government limit, which we mostly get, is that there are too many SQL queries, which would be the most issue we would come across. So, in that case, it's very good to optimize the DML operations, which are already set up in the system. This could include effects and flows, and whichever automation tool requires collecting data. So the gate operation, which is in Apex, and flows also, we have a get records operation. And also, we have update operations and create a record function. So first is to optimize those. Recently, we have been going through an optimization of a flow project. So, we identified several flows that are trying to update a particular record based on certain conditions. And whichever DML operations are happening separately, we had put them into the same flow so that they don't fall across two flows. So, in that case, the DML operations can be reduced.

Formula fields are something which comes in early in the order of execution. So formula fields are to be optimized in such a way that it can handle raw volume of records without impacting the system. For that, I really need to think about that. Maybe we can put more filters on based on what criteria this formula should be calculated. Or, it's better not to have much form of things.