The documentation goes over this pretty well. There are three options for selecting an instance in dynamic mode. Which one you choose depends on your situation. To answer your question if you specify a specific instance manually (string literal), then more than likely when it runs in each environment its going to look for that Graph Mail instance, if it exists in that environment, it'll use it, if it doesn't exist, then it'll probably fail.

Option one: Use placeholders in the specify instance name field

If your instance names follow a predictable pattern, you can use alert placeholders to define which instance to use. The following example uses the alert placeholder in the Specify instance name field. Two instances are defined under the Active Directory integration: ActiveDirectory_UK and ActiveDirectory_US. The ingested alert contains a field called location. In order to make use of this field, you can use the [alert.location] placeholder in the Specify instance name field. In this scenario, within the Active Directory playbook step, you would use ActiveDirectory_[alert.location] in the Specify instance name field, and if the alert comes from America and therefore the value of location = US, then the platform chooses the instance.

Option two: Use entity placeholders in the specify instance name field

and you need a way to return the correct entity only. This is achieved by using both the entity placeholder and the Siemplify Power Ups Tool Buffer action to scope the correct placeholder and return one result.
The following procedure shows how to set up retrieving the correct entity.

1. In the Tools_Buffer action, in the entities drop-down, select the scope that you want to use; for example, Destination users.


2. In the Result Value field, insert the placeholder [Entity.location]. The result of this action will be the placeholder [Entity.location] scoped to Destination users only.


3. In the next playbook step, for example, VirusTotal_Enrich Hash , select Dynamic mode and in the Specify instance name field, select VirusTotal_[Tools_Buffer_1.ScriptResult]from the placeholder options.

Option Three: Use static instance name and the Flow condition step

If your instances names don't follow a predictable pattern, you can still dynamically select an instance based on parameters of your alert using the conditions step. The following example shows how to set up different conditional branches to return the correct instance to be used. The example uses the Alert Rule Generator condition and relies on two instances having been set up under the Email integration named Email_1 and Email_2. Based on the condition result, the playbook will run on different branches and therefore choose the correct instance. So if the Alert Rule Generator equals Cloud Email, the playbook will run on the first branch which uses the instance named Email_1