Best practices for Status-Based Capacity with Omni-Channel

This post is to describe how and when to setup and configure Status-Based capacity in Salesforce, when using Omni-Channel routing, including some common pitfalls and how to avoid them.

One of the big issues we hear a lot is that admins will configure agents to have really high capacity numbers, so that they are always busy with cases that are in progress – maybe a capacity as high as 100 cases at once. This works OK initially, but then an agent goes on leave for a couple of weeks and when they come back online they get 100 brand new cases right away they need to action – PANIC!

I’ll talk about how to avoid this problem down below, but first – let’s start with the basics.

What is status-based capacity?

Within the Omni-Channel setup you have two options for capacity management – ‘Tab-based’ or ‘Status-based’. This gets set at the ‘Service Channel’ level, so for each object type that is routed in Omni-Channel, you can use a different model (e.g. VoiceCall versus MessagingSession).

This influences how and when work items are consuming an Omni-Channel users capacity, and therefore if and when they receive new work.

So what’s the difference? I’ll start with Tab-based, this is the default capacity model and the simplest. When an Omni-Channel user accepts a piece of work, that work is opened in a new console tab. That work will continue to consume capacity until the user closes that tab, or logs out of Omni-Channel.

Status-based works differently – capacity consumption is based on a the value of a picklist field on the record that is being routed. In the Channel Settings, the admin will map which picklist field to use on the object and then map each of that picklists’ values to an Omni-Channel status – either ‘In-Progress’, ‘Completed’ or (new in Winter ’24) ‘Paused’. When the appropriate field on the record is updated, the Omni-Channel status may change, and result in capacity being released. For example, in the configuration below, if the ‘Status’ field on this record was updated from ‘Working’ to ‘Waiting for Customer’, capacity will be released as the work will move to a ‘Paused’ state – similarly if it moves back to ‘Working’ it will start consuming capacity again.

When to use status-based capacity?

Status-based capacity is best suited to long running asynchronous work, where the work may pause and resume throughout it’s life cycle – perhaps you need to wait on the customer to get back to you or a back office team to respond with more information – these are perfect candidates for Status-based capacity. Most often, this is an object like the Case or Lead object.

Today, unfortunately it isn’t supported on the Chat, Voice or Messaging channels, but we’re working on fixing those last two, stay tuned!

How should I configure status-based capacity?

I’ve sold you on status-based capacity – great!! The basics you can find in our help docs, but in this post I want to go into more details of the ins and outs of this.

What we have often found, is that admins will only map the ‘Completed’ status to fields that map to the work being fully complete, rather than also when the agent is unable to do more work on it right now, e.g. Waiting on a customer. To work around this, they will set the agents’ capacity to a really high number – which works OK initially, until a new agent or an agent who was on holiday for a while comes back and they get assigned 50 work items at once! Oh no!!

To avoid this, admins should always ensure that the capacity matches what a user is actually capable of handling at any one moment, so more like 3 or 4 cases at once for instance. There are then 2 more steps: at once.

  1. Map or create any mid-states for the work, where work is not actively happening on the work right now, but will should resume later, to a ‘Paused’ or ‘Completed’ status, to ensure work in these states do not consume capacity
  2. Ensure that users are updating the records appropriately to release capacity, or ideally use automation to update the record as appropriate for the user.

When to use Paused or Completed? What’s the difference?

In Winter ’24 we enhanced Status-based capacity to add a new ‘Paused’ status. The purpose of this is to give users visibility into work that they currently own, and will need to resume at some point. The purpose of this is to give us

The difference between paused and completed is two-fold – the first part is the visibility mentioned above, and being able to see their paused work in the Agent Inbox. The second part is related to when the work resumes – paused work will always go back to the record owner, regardless of their current status or capacity, meaning they could be offline when that work resumes. Work that resumes from the ‘Completed’ state, can go through some extra logic to determine who the work should be assigned to – for instance it can check availability or capacity and return it to the original queue if the record owner isn’t available right now.

Which you want to use are up to you and your business case – is a quick response more important? Then use Completed. Is a consistent owner and the customer relationship more important ? Then use Paused. Both work great for different use cases.

Just now there is no easy way to initially ‘Paused’, and then change to the ‘Completed’ behavior (e.g. at the end of your shift or before going on leave) – but a way to ‘Leave’ the work, or relinquish ownership will be coming soon (goal for Spring ’24!) – stay tuned!

Are there any downsides? What’s the catch?

Once you have status-based capacity configured correctly, there are very few downsides, however there are one or two things to be aware of:

  1. You are now more dependent on agents to end their work and look after their workload a lot more, so they will need to stay on top of their work and backlog to update it as appropriate (or add automation to auto-close work as needed).
  2. The ‘Active Time’ field on the AgentWork record doesn’t get recorded – this is on our backlog to resolve, so keep a lookout each release.
  3. Messaging and Voice don’t support this capacity model today (yet!)

We will continue to invest in enhancing Status-based capacity, but don’t worry – Tab-based isn’t going anywhere!

I hope you found this post useful and please let me know in the comments if you have any questions!