This article was originally published on CustomerThink and is being reposted with permission from the editor.
A customer contacts us with an apparent product issue. With a bit of digging, we discover a larger bug or problem that may need to be addressed by engineering. Before reporting to engineering we get a sense of how many customers are impacted by the issue. Is this a one-off or a system-wide problem? Upon realizing the wide impact, the question that inevitably arises is:
How did it take the customer support team so long to discover this issue?
Ugh! This isn’t my favorite question to answer and yes, I’ve had to answer it many times during my contact center career. In retrospect, there’s always a trail of clues but somehow we missed it or caught it long after many customers were impacted — only increasing the resulting backlash.
As contact center professionals, we literally listen to the voice of the customer all day every day. We have no excuse but to be intimately attuned to our customers and their issues, right? The problem is that we listen to one call at a time or read one conversation at a time.
The question is, how do we get better at connecting the dots and catching those customer-impacting issues faster, minimizing the impact on our customers and our business?
The problem with traditional methods of spotting trends
So how do we typically spot these major trends and issues? Here are some of the methods by which I’ve discovered customer-impacting issues in the past:
A customer responds negatively to a customer satisfaction survey. After some investigation, we find that the system wasn’t working as it should.
The phone or ticket queue is triple its normal size — a clear indicator of a larger problem.
Customers blast the company on social media because of some failure of the product or service.
An analyst reviews trends using speech or text analytics software discovering many support tickets about the same issue.
For those of us without speech analytics, a quick search in the ticketing system yields multiple support tickets about the issue.
While these are all valid means of discovering customer-impacting trends and issues, I can’t help but think that some of these detection methods occur a bit late in the process. Furthermore, these are means by which managers find issues. Agents typically know about issues before managers do. The challenge is in getting them to recognize that it’s a larger issue.
It’s important to note, in the case of increased call or ticket volume that some issues, like a server outage, may result in a quick, drastic increase. But that billing bug that was introduced in last week’s release will instead cause a gradual increase in ticket volume as customers become aware of the issue.
How do we improve the speed at which we identify customer-impacting issues of all shapes and sizes? In the remainder of this article, I’ll pose six key questions that will assess the readiness of your contact center to detect issues.
Are you ready to catch problems when they happen?
Before we begin this assessment, allow me to give you the answers. In short, the answer to each of these questions should be “yes.” If you do these things in your contact center, you’ll catch and correct customer-impacting issues faster — sometimes before your customers experience them.
1. Are your agents empowered and given the time to thoroughly test an issue or are you micromanaging their average handle time (AHT) or after-call work (ACW)?
I think it’s fairly rare for contact centers to ask their agents to spend more time during and after calls. Sure, some contact centers may pressure their agents more than others. But time metrics are easy targets for managers because they can be tied to reduced costs with relative ease.
While some focus on time management is important, added pressure on agents to take less time with each customer, means they will be less likely to expend the additional effort to identify customer-impacting issues and trends. In the interest of spotting issues more quickly, perhaps it’s time to encourage agents to spend the time they need (within reason) with each customer.
2. Are your agents trained to recognize when the system isn’t doing what it’s supposed to?
Agents should be able to connect the dots and realize that if the system isn’t working correctly for one customer, it’s possible that others are impacted. If they don’t understand how the system works in the first place, that’s a training issue. The ability to recognize potential issues is an important reason to constantly improve documentation and training.
It might sound like we can improve training with the wave of a hand but it’s more complicated than that. We are constantly working to better document how systems, like billing, work so that our team is more and more educated. Training also happens when agents go to supervisors with questions, hopefully increasing the likelihood that someone will discover a larger issue.
3. Do your agents have the tools necessary to test your product?
Agents should have access to the software or hardware that customers are using and they should have an environment through which they can test. All of our employees have access to our software and mobile application to test customer issues.
In past roles where IOT (Internet of things) devices were used, we successfully created a “lab” within our contact center where agents could get their hands on the devices they were supporting.
4. Do you involve your agents in testing new releases?
This one is a major challenge in a contact center environment. After all, we need our agents to interact with customers, right? As your engineering team is releasing new features and updates, it’s important that the contact center is involved as much as possible in the testing process. Create a means by which they can report bugs and raise feature suggestions.
At our company, our engineering team is begging for as much feedback as they can get which puts the onus on contact center leadership to carve out the time for agents to be involved. This familiarity with the product or service pre-release is invaluable if and when issues arise post-release.
5. Do agents have an effective means of collaborating with one another?
In a contact center environment, there’s incredible value in one agent being able to turn to another agent and say, “Hey, this customer is experiencing an issue. Have you encountered this before?” When multiple agents say, “I’ve seen a few tickets (or taken a few calls) about this today” a trend has been discovered.
In the modern, hybrid contact center where the team is distributed between an office and remote work, a collaboration tool like Microsoft Teams or Slack is the new way these discussions happen. Leaders must create a culture where agents can talk and collaborate. Sure, some of the conversations might seem like a waste of time, but when issues arise, you’re going to want them to speak up as soon as they notice that something isn’t right.
6. When team members raise a concern, is someone listening, filtering, and quantifying?
Finally, the challenge for supervisors and managers is to understand an issue and determine the full impact before raising it to the engineering team. Lest we be accused of crying wolf, this is where we quickly work to quantify the gravity of the situation. As mentioned before here are some ways we can piece together the clues:
Resulting negative customer satisfaction: It’s not uncommon for us to talk about negative customer satisfaction results or complaints on social media and review sites about an issue.
Overall ticket volume: Speech and text analytics can be invaluable tools to figure out how many customers have contacted us about an issue. Create alerts for certain key words or phrases that might indicate a problem. For those of us that don’t have these tools, the search function in our ticketing system is helpful. For known issues, we’ll often have our agents apply specific tags to tickets so we can more easily track them.
Review the queue as a whole: It’s easy to focus on one customer interaction at a time but someone should always be tasked with looking at the entire queue. For example, at our company, if we see multiple tickets with a subject stating “Unable to send text messages” we dig deeper.
Database query: Our management team has access to query our customer database. If a customer reports that they were able to do something out of the ordinary, we’re able to quickly search for other customers with the same issue. This level of access sometimes allows us to catch problems before customers do.
Once we’ve completed this investigation, we can confidently proceed to the engineering team with a full report of the issue and an understanding of the priority level.
Fostering a culture of open communication
I’ve never known a contact center agent who was unwilling to talk about the issues impacting customers. At their very core, I find that customer support professionals are passionate about solving customer problems. They derive a sense of purpose and accomplishment by improving the customer experience. They also don’t love interacting with upset customers– at least I don’t.
In environments where agents either don’t feel heard or are made to feel that their concerns are invalid, self-preservation kicks in and they will stop sharing. It’s critical that we, as leaders never allow this to happen. We must always be listening and working to understand the depth and breadth of issues. And when agents raise issues, our response should always be, “Thank you” — even when it doesn’t end up being a large issue.
If the contact center wants to be a good partner to the rest of the organization and take better care of customers, we must hone our ability to recognize and report customer-impacting issues as quickly as possible. Try this assessment on for size and let me know how it goes.
