When your customer contacts your support center via call or chat here are the top 9 pain points they face*:
- Long Wait Times (Before reaching an agent)
- Slow Service (When an agent is reached)
- Self Service is difficult to use
- Rude agents
- Agents aren’t knowledgeable
- Repeating info when a call is transferred
- Cannot get service in the preferred channel
- Service isn’t personalized
Support centers are high-stress environments. Most incoming calls are complaints, and so understandably, the people calling aren’t in the best of moods. It isn’t easy to motivate agents to thrive in this environment. And since many customer support centers are still treated as cost centers, budgets are tight. Calls often exceed the number of agents available, leading to high call volumes. Often agents handle calls back to back. And sometimes even managing calls and chats simultaneously.
Yet all the problems listed above are solve-able. Over the past decade of working with contact centers, we’ve learned some of the best practices for dealing with these problems. And of course, we have worked hard to provide technology to help.
Let’s delve into each problem and see how it can be solved in efficient, cost-effective ways:
Long Wait Times (Before Reaching an agent)
What’s the issue? When customers call in with complaints, most likely they’re already facing a problem. What they want, more than anything else at that moment, is for the problem to disappear fast. This should be—and probably is— your support center’s most important focus. Reducing the time taken to reach your contact center is the best place to begin.
Why does it happen? There are two main reasons why your customer cannot reach a live agent fast:
- Agents are unavailable to take calls or chats leading to long queue times
- IVR is too long.
How to measure it? Average time in Queue. And Average Time to reach a Live Agent.
According to our research across 160 million calls: in 2018, customers spent an average of 75 seconds in the queue. A long IVR menu may further delay a customer. On chats, your chatbot may waste their time. A measure of around 10 seconds to reach a live agent is far more ideal. And it’s do-able too.
How to fix it?
- Plan staff/ Forecast regularly. Your data is a gold mine. Use it to plan your staff more efficiently. If you go into your call reports regularly, you can see what time of day or day of week call traffic is high, and when it isn’t. Use this to plan your agent rosters accordingly. If the metrics show a need for increased staff, you can use the data to quantify exactly what your increased budget requirements are. (Also Read: Call Queue Management: Plan Staffing)
- Create Fallbacks. Take your metrics seriously. If calls do not connect with a live agent within the stipulated time, you should provide fallbacks.
- Another skill group: Your fallback could be another department or support center. Or it could also be a group of trainees or a support crew you hire on a work-from-home basis. You could also set your fallback to a voice bot/chatbot/answering machine or voice mail. (Also Read: Call Queue Management: Create Fallbacks) .
- Call back. When agents are too tied up to answer a call (or chat), don’t keep your customer waiting. It is disrespectful. Instead, do the decent thing— let them hang up, and promise a callback. And then deliver on that promise. To do this, you can use IVR/voice bot or a simple voicemail to take down their complaint. Use the Abandoned Call Alert feature to send alerts to your agent or managers’ inbox. Or automatically schedule a callback into your dialer. (Also read: How to automate callbacks)
- Voice Mails: Your IVR can ask callers to leave a voicemail. The callback from the agent will then take place once s/he has understood the problem. This will increase the chances of a faster resolution. Best practices suggest that you should use this fallback only if you can ensure a callback within a minimum of 15 mins.
- Prioritize. When a lot of callers are waiting in the queue, you can prioritize your customers as opposed to your non-customers. A CRM integration can be used to recognize your customers while diverting casual callers to an IVR/voice bot/chatbot. You can use Dynamic Queue Reprioritisation to divert important callers to the top of the queue. (Also Read: How to treat your customer like a VIP)
- Improve your IVR. Wherever possible, automate. Use your CTI integrations to dip into your CRM/Ticketing software data. Bypass the IVR when possible. For example, customers with unresolved complaints, or high-value customers can be directly connected with a well-trained, live agent. Prospects can automatically connect with agents they last spoke to. (Also read: Sticky Agents)
- Keep IVR menus short. If a long IVR menu can’t be avoided, consider using voice bots, or conversational IVR. (Also Read: Modern IVR)
- Consider Self Service. Analyze what calls your customer support center gets. If you get a lot of “how-to” questions, consider creating a bot or a conversational IVR to answer these. If you get a lot of transactional calls—such as rescheduling meetings, making bookings or cancellations—consider diverting these to self-service. It will free up your call queues for tasks where a live agent is more essential.
Slow Service (When an agent is reached)
What’s the issue? It isn’t enough for customers to connect with a live agent. The agent should be equipped with the training and information needed to handle the task. If your agents themselves are fumbling around looking for information, it will only frustrate your customer.
Why does it happen? The reasons for slow service can be many and varied. It could include:
- Poor connections. Poor call quality can lead to voice lags, and both agents and customers may waste precious time repeating information.
- Poor training. Agents may not be trained to handle the problem.
- Disconnected Systems. When your CRM and telephone solution is not integrated, it’s unnecessarily time-consuming for agents to search for information.
How to measure it? Average Handle Time + Csat.
In 2018, according to our research, the Average Handle Times were 3 minutes per call. Your average handle times, however, will depend upon your product/industry. Instead of considering your average handle times in isolation, correlate it with your customer satisfaction scores, before drawing any conclusions.
How to fix it? If indeed, customer dissatisfaction coordinates with long handle times, then you may need to consider these methods to reduce your handle times.
- Good voice quality (Also read: Insider tips for ensuring high voice quality in your cloud telephony solution)
- Skill Hunting. Dividing your agents into skills can help you train your staff more easily. A group of agents can specialize in handling specific problems. Routing calls to the right skills will then lead to a corresponding increase in Csat. (Also Read: A guide to creating skills)
- CTI Integrations. Are your agents getting all the information they require? A good CTI integration with your CRM or Ticketing Software will ensure your agents have all the information they require at their fingertips, leading to far faster resolutions. (Also read: Applications of Call Center CTI)
Self Service is difficult to use
What’s the issue? It’s not enough to implement self-service, you need to ensure it is fast, convenient and easy to use. You need to carefully track user experience and satisfaction levels. And continuously improve your system.
Why does it happen? Few reasons for bad self-service include:
- It is badly designed. A badly designed self-service is difficult to use. Ideally, a UX designer should be involved in creating a good self-service. The benchmark to use is that 50% of calls should be handled by self-service within 6 months of implementation. If this is not happening, its best to remove the self service option and review what did not work.
- It’s time-consuming. A caller may get frustrated if a bot/IVR is asking too many questions before providing a service/giving information.
- The task isn’t suited to self-service. Self-service should be selected for repetitive, transactional tasks that can be completed faster via self-service than by a live agent. For example, a cancellation. Complex tasks like handling irate customer complaints should be handled by a human agent.
- The customer doesn’t like self-service. Some customers like self-service and others don’t. Some may need a live agent to guide them for the first few times. Understanding your customer profiles can help you decide which customers to divert to self-service, and which to onboard more gradually.
How to measure it? CES And FCR.
First call resolution (FCR) is a good measure to apply to your self-service. If your self-service results in an escalation or a repeat call, consider it a minus point. Also, use surveys to measure Customer effort (Customer Effort Score or CES) at the end of each transaction.
How to fix it?
- Keep a live (human) agent handy. Implementing a perfect self-service loop takes time. Defining fallback rules for your self-service help. For example, if your bot/conversational IVR doesn’t understand the customer request after o1-2 tries, it should immediately connect to a live agent.
- Ensure that it’s self-learning. An AI-powered bot will learn from its mistakes. For example, if the caller wants to know how to fix a product and the bot connects it to the wrong resource, they will express some dissatisfaction. For that moment, the bot will divert to a live agent and this agent will connect to the correct resource. Meanwhile, in the background, your bot will learn from this mistake, and next time connect the user to the right resource.
(Also Read: Guidelines for Implementing Self Service)
What’s the issue? This one is self-explanatory. Rudeness obviously leads to unhappy customers
Why does it happen? Overworked agents may be rude. Untrained agents may come across as rude. And connecting customers belonging to one culture with agents belonging to another culture can lead to the perception of rudeness, even if none was intended.
How to measure it? Csat.
The simplest way to measure dissatisfaction is to route a customer to an IVR/survey at the end of a call or chat. Rude agents can be cited as an option under “cause for dissatisfaction”, to help pinpoint if this is what is making your customer see red. Alternately Speech analytics can be used to catch any rude words/phrases used by the agent during the call.
How to fix it?
- Creating a workplace that values and appreciates good customer service is important.
- Ensuring your agents get enough breaks, especially after handling difficult calls is necessary.
- Cultural sensitivity and training can help improve agent performance.
- Digital assistants can use speech analytics to cue agents when they are being rude, or alert managers.
- Barge in. Managers can use speech analytics to monitor calls and barge-in when necessary.
(Also Read: Tools and tips for training customer support agents)
Agents aren’t knowledgeable.
What’s the issue? Often agents don’t have access to the knowledge they need to solve a problem.
Why does it happen? The issue could be a lack of training or disconnected/difficult to use knowledge bases.
How to measure it? FCR
When agents don’t have the knowledge needed to solve the problem, they will need to escalate the issue, reducing the First call resolution or FCR.
How to fix it?
- Skill-based routing. As I mentioned before, dividing agents into skills allows them to get skilled at handling certain issues. Skill-based routing can be used to divert customers to the right skill for the job. (Also Read: Call Routing Tips)
- Smart Attendants. A smart IVR or auto-attendant can gauge what a customer’s problem is and divert them to the agent best trained at handling that problem.
- Digital Assistants. Digital assistants are an intuitive, effortless way to access knowledge. They can give your contact center agents quick access to the knowledge needed to solve problems, in much the same way as you get smart reply suggestions on your Gmail. (Also Read: Contact Center Trends 2019)
Repeating info when the call is transferred
What’s the issue? When a call is transferred, the agent transferring the call should brief the agent about everything the customer has already told them, so that the caller doesn’t have to repeat themselves. Or the next agent should receive an automated transcript of the previous conversation when s/he is connected.
Why does it happen? Using cold transfers means that the agent whom the call is transferred to has no idea of call context and has to ask the customer to repeat anything they said before.
How to measure it? Average Handle Times and Csat
Asking customers to repeat themselves increases average handle times and decreases Csat.
How to fix it?
Warm Transfers: It must be made a practice that whenever a customer has given an agent any information that isn’t mentioned on their ticketing history or previous call disposition—the agent only transfers calls using a consultative hold or warm transfers. (Also read: A guide to warm transfers: how, when and why to use them)
Cannot get service in the preferred channel
What’s the issue? Today’s customer is spoilt for choice. They expect to be able to connect to your customer support via social media, messaging service, email, phone or chat, whatever is most convenient at that moment. The time isn’t far when a customer will use their digital assistants to place calls while driving, cooking or walking.
How to fix it?
Use a contact center solution that is omnichannel. An open platform that can integrate with any new third-party applications with ease. It should have social media integrations, email integrations, SMS and text message plugins built-in.
(Also Read: Why your contact center needs to integrate facebook messaging, A Cloud telephony WhatsApp integration)
Service isn’t personalized
What’s the issue? Customers have got used to personalized service, and have come to expect it in all their interactions. Greeting customers by their name, knowing their caller/ticket/purchase history and their various preferences are all part of personalizing the experience for them.
Why does it happen? Disintegrated systems. It is important that your systems: your CRM, your social media, live chat and voice interactions—can all seamlessly communicate with each other to build a unified customer experience.
How to measure it? Average Handle Times and Csat, NPS
Asking customers questions increases handle times. Non-personalised service decreases Csat and Net Promoter Scores.
How to fix it?
Customer journey mapping.
Omnichannel presence. (Also Read: How can your contact center deliver Omnichannel Service)
Customer support isn’t easy. But in an increasingly digital world, it is often the only human voice and personal interaction people have with your brand. It is important to get it done right. I hope this article was able to illustrate that with the right tools, training, and motivation, creating a happy, successful customer care center is an achievable goal.