Articles in the series
- Messaging deployment guide: Introduction
- Part 1: About conversational support with messaging
- Part 2: Designing your conversational messaging workflow
- Part 3: Planning your staffing and operational requirements
- Part 4: Rolling out conversational messaging support
When it comes to workforce management, messaging works best with operational processes and procedures that differ from those used with more traditional channels and support structures. Whether you're coming to messaging from traditional phone- and email-based support structures or a newer chatbot structure, you might need to shift the way you think about staffing, workflows, and performance evaluations.
In this article, you'll learn about the staffing and operational changes you should consider for messaging.
Finding the right balance between supporting customers, meeting service level agreements, achieving high customer satisfaction scores, and managing your budget is key to building an effective support staff. While designing your conversational messaging workflow, you should have considered the number of staff you'll need and how best to configure messaging workflows to maximize effectiveness and efficiency.
The Zendesk website has useful calculators to help you estimate the number of agents you'll need when implementing messaging.
Considering how messaging triage and routing affects staffing
How you configure the flow of messaging conversations is a key part of a successful conversational support strategy. It also has a big impact on your staffing needs.
We recommend keeping your ticket triage and routing process simple to maximize agent time and improve the level of service provided. In addition to carefully choosing your routing method, you can also help agents manage their conversational messaging workloads successfully by doing two things:
- Setting conversation limits or capacity rules
- Automatically closing inactive conversations
Setting agents up for success
- Roll out messaging in phases.
First, consider taking a phased approach to the development and deployment of your messaging workflows. You can start with a live chat-like setup that your agents might be more familiar with, then introduce more asynchronous messaging functionality over time.
- Train your agents ahead of time and follow up after the
Training your agents is equally important as giving agents time to adapt to different support systems and behaviors. Before you roll out your first conversational messaging workflow, notify your agents and tell them what to expect. Then, continue to do so for each change you make to your conversational support model.
The following resources can help:
- Zendesk messaging training course
- Receiving and sending messages in the Zendesk Agent Workspace
- Using notification lists to manage conversations
- Explain how work will be routed to them in the new workflow:
- Help them understand their role within the workflow if you're using autoreply or conversational bots.
Understanding your goals for conversational messaging support can help you configure the optimal workflows and customer experience for your unique needs.
After agents have been trained and messaging has been activated, it's important to expect an adjustment period while you learn to compare messaging volume to volumes you previously experienced from email, chat, or calls.
You're likely to see request deflection rates changing for the better soon after deploying conversational messaging support. With fewer customers needing live support from an agent, a lot of statistics you used to evaluate agent success and requirements will need to change. Give yourself ample time to observe these changes, and then apply them to your volume and staffing forecasts. Be prepared to work with the admins managing your bots and automation on an ongoing basis to maintain this. See Measuring success when migrating from live chat to messaging.
Additionally, it's normal to experience a temporary increase in request volume while your customers and agents adjust to the new experience. Listen for trends in your customer feedback that might indicate the need for additional agent training or even increased staffing to prevent customer frustration and increased wait times. Remember that messaging volume per channel should ultimately be lower compared to chat due to the higher likelihood of self-solved tickets and the exclusion of disconnected support requests from the queue.
Finally, abandoned conversations are an invalid condition for messaging tickets. Therefore, you should remove them from your forecast.
Changing your metrics and SLAs
After your agents settle into your new conversational messaging support process, you can start focusing on updating your workload and productivity expectations, optimizing your bots and automation, and performing root-cause analysis of your tickets. All of these should lead to a long-term decrease in ticket volume and increases in agent productivity.
You'll likely need to reconfigure your workflow management tools connected to Zendesk. This normally happens due to introducing new skills, groups, or metadata that you want to use to build your forecasts.
Note how the handling of tickets changes after you deploy messaging. Are there more ticket touches or fewer? Are those ticket touches shorter and over a longer period of time? Questions like these can help you plan your staffing and operations requirements to fully realize your organization's potential.