Ask yourself these questions when considering chat capabilities for your contact center:
1. Will we have a high volume of chat, or might we start with a trickle? Or do we just not know at this point? If you are uncertain of your volume or it's low, dedicating a group can be an expensive proposition. Consider training a few "blended" agents to start, and go from there. Another option is to consider an outsourcer.
2. How long are our calls? How long do we expect chats will take? (Perhaps do some internal pilots or role plays to assess this.) If your calls are long, and chats will be too, be careful about dedicated groups. This combined with low volume can be a killer on your efficiency and queue stats.
3. How complex are the questions we handle? More complex questions make it more likely you'll need to avoid multiple chat sessions. The agent needs to focus on one interaction at a time.
4. What is the risk of providing the wrong answer to the wrong person? Sound like a crazy question? We've seen situations where the company was considering multiple chats but the agents knew that could lead to them pushing the wrong answer to the wrong person, with some dangerous consequences. Avoid that situation!
5. Do we have a lot of down time today? Is our agent utilization low? If you have down time and opportunities to boost utilization, having blended agents might help (the fewer silos the better).
6. Do we have times when we're hammered with peaks, and the queues back up? If so, look at other reasons you should or shouldn't do multiple chats - they can certainly help with peaks.
7. Are there times when our agents are waiting on the phone for the customer to do something? If so, this creates an opportunity for multiple chat sessions.
These are just some examples of things that will point you in one direction or another. You also need to consider how the chat option is being marketed, and what the business goals are with it. A full analysis of your specific situation is worth the time - make some pros/cons tables, and do a cost/benefit analysis. And absolutely do a pilot, with time to adequately analyze the results and an open mind to change course if necessary.
Click here to read Part I of Lori's expert answer