AI in Contact Centers

AI in Contact Centers

First, I thought that launching a newsletter would be difficult. But now, I think that second edition is quite difficult for me. What to write about? What topics do I pick for this? Mostly I talk about customer service, contact centers, human factors and people.

?

And the hottest debate going on is how to use AI in any function. So, for service, customer service, contact centers. I am all for human interactions.

Human to human interactions will always, so far, will remain priority for me. I wouldn't recommend anyone to jump on the bandwagon, copy-paste what others are doing, and jump in the AI well, or what do you call it, AI ocean, something like that.

Technologies evolve, but the use, the utilization should be very pragmatic & planned. Present era is a disruption era, but to use any technology head-on is not the solution.

  • It will not improve your customer service, or to use the buzzword, customer experience.
  • It may frustrate the customers.

I continue to see bad experiences; customer had to surf between the channels, from chatbots to IVRs, to going here and there. After 40 minutes, customer happens to speak to a human agent, and the problem was solved in 15 seconds.

Anything from lengthy product-based IVRs to pushing customer, forcing customer to use automated channels, instead of giving them the upfront option of talking to a human agent.

For IVRs, to continue providing positive, memorable customer experience, is to give them options, but do not make them surf the IVR unnecessarily, after lengthy wait time, they can talk to a human. Give them all the options upfront.

For example, in banking, it's good to use automated banking, self-service banking, if it's a routine transaction.

But if there's a problem, for example, customer wants to discuss an unauthorized transaction, it's better that they speak to a human, and get a complaint logged.

They get a proper answer, assurance, some sort of input on how long it will take to resolve. And, of course, the back-end process should be very efficient, and complaints should be resolved within the minimum possible time.

Back to AI usage. I think, instead of chatbots or machine answering on your WhatsApp channel or your social media channel, or your chat, or maybe an email, or even a voice AI agent, I would suggest, and I think it's better to implement or utilize AI in a manner where it should not impact the customer negatively.

The recommendation is, use it for workforce management.

Schedule agents & team leads, depending on the workload. The data of contact center goes into a machine, AI processes it, throws up a schedule based on at least a year's data. It can identify very easily what are the peaks, when most agents are needed, when not , where to schedule a vacation, or how to manage during holidays and public holidays.

When this schedule is circulated, there's a room for change. Work force managers will know immediately and tweak it where required

The second is, quality assurance. It's impossible and humanly impossible to listen to at least 5% of all the agents' calls, evaluate them, prepare results, arrange the coaching sessions, take corrective actions, and so many things. AI can help here.

AI can do the analysis. AI can check for certain defined parameters. Identify the words agents should or should not use, the language, the accuracy of information provided.

It can be trained to gauge the tone and the solution. It will take a lot of input from historical data and some input in terms of how the process is defined for AI, what to look for, and how to evaluate or mark or gauge the response of agent.

Human QA can review that; if they think it needs a second review, they can listen to the call themselves. In the long run, AI can later provide better responses without actually impacting the customers.

QA can spend more quality time on improving the processes, identifying the weak areas, training and coaching the agents, developing learning modules, and so many other things. The other area where AI can be used is to fetch data from your systems and prepare an agent scorecard based on defined process about the talk time, the HT, any other matrix, maybe the login time, the non-productive time, their utilization.

AI can easily prepare a scorecard.

The other area I recommend is if contact center is offering multilingual support and if a customer calls to speak to someone in French, but at that time, an agent who can speak French is not available. What AI can do is, when system routes the second-best agent, AI can help them. Maybe a customer can speak in French, AI can translate into English, and that agent can respond in English, and AI can translate and play it back for the customer.

Essentially, the response is human. The only problem could be that if the customer is irritated, not happy, or does not understand the solution, it can be improved over time.

Run an internal test with select few having French or Spanish as a second language but are fluent in English, or maybe their first language is Arabic and they are fluent in English.

At a lot of local, regional level, these could be helpful for customers and agents. The same principle can be applied to chats, WhatsApp channels, your social media interaction, where a customer wishes to put forward their query in, say, Arabic. A chatbot can actually convert it for a standard set of FAQs, kind of a thing.

When it comes to real conversations, I still suggest humans, humans, humans. Humans can make or break. Don't let machines rule you, but humans should take care of human interactions.

?And there is always big, big room for improvement, innovation, changes, continuous monitoring, which brings out the best.

Oh, by the way, another area. If call center is heavily loaded. SLA is negative and customers have to wait long. Ask AI to fetch data and identify what are the top five reasons customers are calling . Start improving the process & training around those.

?

That’s for the September issue, three freebies below:

?

1-????? I am offering 4 free 30 minutes meeting on the topic of your choice from career coaching to contact center operations. Book yours now www.expertpin.com/reshma-abdullah-9249

2-????? Want to learn on more, subscribe to my channels

3-????? Check AI in 5 series by https://www.dhirubhai.net/in/simonkriss/

?

Asad Afridi

Aviation / Airlines IT Professional

6 个月

Excellent! Keep up the good work that you do.

Simon Kriss

Australia's leading voice on AI adoption | International Keynote Speaker | Author | Board & C-suite AI Mentor | Chief AI Officer | Senior Fellow AIFOD

6 个月

Thank you for the shout-out and listing of the AI in 5 minicast !!

要查看或添加评论,请登录

Reshma Abdullah的更多文章

社区洞察

其他会员也浏览了