I have worked with customer service and Zendesk since 2013. Much of my work has focused on making customer enquiries easier to handle. Better workflows, faster responses, more automation and better reporting.
But there is one thing I think we talk about far too little: Why do we receive all these enquiries in the first place?
If customer service answers the same question a hundred times, perhaps response time is not what we should be trying to improve.
When efficiency improvements are not enough
Imagine you run an online shop selling bicycles. Your customer service team keeps getting questions about which colours a particular bicycle model comes in.
You could create a macro that makes it easy for agents to respond. You could put the information in a knowledge base, or perhaps let an AI agent handle the question automatically.
All these solutions could make customer service more efficient.
But what if the real problem is that the available colours are not shown clearly enough on the product page?
Perhaps that is where we should start.
This is one of the things I find most interesting about customer service. We spend a lot of time improving how we handle enquiries, but there is also huge potential in using what those enquiries tell us to prevent the same problems from happening again.
Customer service often knows more than we think
Customer service agents speak to customers every day. They know which questions keep coming up, which products people struggle with and what causes frustration.
That information is valuable to the rest of the business, too.
If customer service receives numerous complaints about delayed deliveries, perhaps the logistics department should investigate what is happening. If the same product generates lots of complaints about defects, it may be something the product development or quality control teams need to look into.
And if customers repeatedly ask for information that ought to be available on the website, perhaps someone should update the website.
Of course, these are not necessarily problems that customer service can solve on its own. But the team can help uncover them.
I think customer service departments should aim to reduce the need for their own existence. That might sound a little extreme, and they will never become entirely unnecessary. But surely the goal should be for customers to experience as few problems as possible, rather than for us to become world champions at answering the same questions?
What do we really need all those ticket fields for?
This is where Zendesk plays an important role.
With the right setup, we can categorise enquiries, identify recurring problems and track developments over time. But that depends on thinking carefully about what information we need and why.
I have been involved in many discussions about reporting, and it is easy to get carried away when you see all the possibilities. Suddenly, everyone wants twelve new ticket fields because it would be interesting to report on even more details.
The problem is that someone has to fill in those fields.
If an agent spends an extra 10 seconds per ticket recording information, that might not sound like much. But across 100,000 tickets, it adds up to almost 280 working hours.
So we should be fairly certain that the information actually has value.
I am not saying we should stop recording information. Quite the opposite. But before adding another field, we should ask a few questions: What will we use this information for? Who needs it? And how detailed does it really need to be?
If the product department needs to know which product models generate the most complaints about defects, perhaps that is exactly what we should report on. We do not necessarily need to record every tiny detail if nobody intends to use it.
Zendesk gives us many possibilities, but that does not mean we need to use them all.
The people recording the information should also see the results
This is something I care particularly strongly about, perhaps because I have worked as a customer service agent myself.
When you spend all day handling tickets, it can be frustrating to keep entering information without knowing what it will be used for.
I think we should do a better job of showing agents what their reporting actually leads to.
Imagine how much more meaningful it would feel to record a contact reason if you later found out that the information had revealed a problem with a supplier, and that the problem had now been resolved.
Or that customer feedback had led to better product information on the website.
Then you can see that your work has value beyond resolving an individual ticket.
It also affects the quality of reporting. If agents see data entry as unnecessary extra work, it is hardly surprising if the quality deteriorates over time.
Can AI help us gain better insights?
One development in Zendesk that I find interesting is the opportunity to automate more of the categorisation of enquiries.
Traditionally, we have relied on agents to record contact reasons and other information themselves. That takes time, and it is also fairly subjective.
Two agents may read the same enquiry and arrive at different contact reasons. Perhaps the issue involves both a damaged item and a delayed delivery. Which category should they choose?
Features such as Intelligent Triage in Zendesk can make it possible to categorise enquiries more consistently without agents having to do all the work manually.
It is not perfect, and the quality of the categorisation still needs to be assessed. But I think it is a good example of how AI can be useful without necessarily being the one that responds to the customer.
For me, one of the most exciting possibilities of AI is precisely its ability to remove time-consuming, repetitive tasks from agents' working days.
While also helping us get better information from the tickets we handle.
Resolving more tickets does not always mean we are doing a better job
We are used to measuring customer service by response times, resolution times, the number of tickets handled and customer satisfaction. These are useful metrics, and I absolutely think we should continue using them.
But I think we should also ask ourselves some other questions.
Have we managed to reduce the number of enquiries about a known issue? Have we used customer feedback to improve our products or services? Are there questions customers no longer need to ask?
These improvements may not be as easy to measure. But they can make a big difference to both the customer experience and the workload of the customer service team.
I enjoy helping businesses get more out of Zendesk, whether that means automation, reporting or making agents' working days easier.
But perhaps one of the most important things we can achieve is making sure the business actually learns from the enquiries it receives.
Because the best customer service department may not be the one that resolves the most tickets.
It is the one that helps prevent tickets from arising in the first place.
Would you like to learn more about how Zendesk works, or do you need help with your existing setup? Feel free to contact us for a no-obligation conversation.