New & Improved Chat Metrics
Multiple systems, no unified view
Support teams have to accept that some customers will always want to use their favourite channel to get in touch. Offering support through phone, email, chat, social media and, more recently, messaging apps has become the status quo. Your agents on the frontline are now multi-tasking professionals; experts in omnichannel communication.
But management teams need to track the activity on each of these channels to be able to cope with the growing demand and ensure the same quality of service delivery.
When you’re trying to give your customer ‘best in class’ support, you may choose to work with a variety of tech partners that manage your chat and phone channels, for example. You have the best technology stack (the one that suits your needs the best) this way, but the issue is that your data is scattered between applications: getting the full picture is a real challenge.
Each application is equipped with its own reporting solution, so managers struggle to get a unified view of all their channels to answer the big, high-level questions:
How are customers distributed (across channels) by contact reason?
What % of tickets come from channel X or Y?
What's the average reply time per channel?
Where do we need to put more resources to cope with our growing demand?
Too often, managers manually export the data from each system into a spreadsheet to get a holistic view of their operations. The more fortunate teams have data or BI teams to build and maintain static reports for them - when they have time to… It’s not always possible and it’s still all manual work.
It could be much simpler.
Centralize all your omnichannel data
Chat conversations that take place in external chat applications may have their transcript sent to the ticketing system, but a lot of valuable information is lost. It might create a ticket that looks something like this in Zendesk:
This is a transcription of a conversation that took place through Zendesk Chat but, because it’s a transcript, there are lots of metrics that you can’t calculate; one example is Average # of Replies (the number of interactions in a conversation).
In the ticket above, there are several replies back and forth, but there’s no differentiation between what the agent said and what the customer said. Let’s see how your metrics used to look in Miuros:
Here, the Average # of Replies was 0.72 for chat. Since chats usually have multiple replies per message, this is potentially misleading because you don’t have the full story.
Metrics you were missing:
First Reply Time
Average # of Replies
Average # of Messages/Interaction
Customer Waiting Time/Reply
Total Time to Solve
Technically, they’re not “missing”... but they’re wrong - in the sense that don't necessarily reflect the truth. These metrics are calculated ‘wrongly’ by any tool that only looks at a ticketing system’s data.
Miuros Reconstructs each ticket
With our latest release, we’re artificially reconstructing each conversation to accurately calculate the real metrics. For example, we scan the timestamps of each message to calculate metrics like Next Reply Time or Customer Waiting Time.
Or, using our previous example, calculate how many interactions there were to give you the Average # of Replies/Average # of Messages/Interaction:
The Average # of Replies was 0.72… Now it’s 6.83, which is far more accurate and gives a much clearer picture of the chat channel as a whole.
It’s now so much easier for them now that it’s all on one dashboard with a flexible view. Before, they couldn’t do it without a data/BI team that spent hours a week collecting all this data and delivering the results.
We’re calculating metrics for these and many more... See what else Miuros Insights can do here!
