A Guide to What IVR Technologies Are and How They Work (NLP & NLU Defined!)

A modern IVR system (Interactive Voice Response) relies on two critical components to be able to effectively ‘listen’ to your callers. These are:

• NLP (Natural Language Processing).
• And NLU (Natural Language Understanding).

Both of these components have two very distinct roles and both terms are used interchangeably. Given the fact that these IVR technologies are going to be critical for the next generation of automated call tools, it’s best that you fully understand the distinction between the two.

In this article we’re going to cover the following:

• What is Natural Language Processing and how does it work?
• What is Natural Language Understanding and how does it work?
• Are Interactive Voice Response systems worth investing in?

Let’s get started!

What is NLP (Natural Language Processing) and how does it work?

So, what is NLP?

NLP is a component of artificial intelligence that focuses on processing speech and turning it into structured data that can easily be identified by a machine.

In essence, NLP aims to make artificial intelligence more human-like, which is handy in an inbound call centre environment.

Given that speech is the fastest method of communication for humans, having a machine that can easily understand our speech is important. However, the problem that IVR systems face when using NLP is the tricky variations in human speech:

• Different regional accents
• Sentence structure
• Choice of word
• Disfluency
• Synonyms
• Homophones
• Dialect

There are many different variations in human speech – much of which we humans can struggle with, let alone a machine.

In any case, with the rate at which artificial intelligence is growing, it won’t be long before Natural Language Processing will be far more adept at understanding and processing a multitude of human language variations than the average person.

What is NLU (Natural Language Understanding) and how does it work?

So, what is NLU?

NLU refers to a useful subset of NLP (an important distinction). This is designed to look beyond word choice and determine what the intent is behind those words.

Ultimately, it allows the IVR system to better respond to various words and phrases with the appropriate actions.

So for example, if you call a speech-enabled IVR call centre and say: “I’d like to reschedule my engagement for May 8^th to the 10^th please” – the NLP will understand every word in the sentence, however, the NLU will be able to determine that “reschedule” is the key intention of the statement and route your call accordingly.

Are conversational Interactive Voice Response systems worth investing in?

The fact is, you can call practically any large call centre and they will almost certainly be using conversation IVR systems already. There will be many more companies following suit shortly.

But, just because everyone else is doing it, does it mean that you should too?

The easiest way to know for sure is by asking yourself the following questions about the current service that you offer:

Could you do with more customers using a self-service feature?

Self-service systems are traditionally rather poorly implemented. For example, the DTMF input (e.g., “please press #1 to check your balance”) is long-winded and can be the cause of much frustration for callers. These systems also force callers to listen to several irrelevant options which are tiresome (particularly when the option you need invariably comes up last).

The worst thing about this is the fact that, as a result, many call centres end up using their IVR system for call routing alone, as opposed to a self-service option – which is a big waste of such a valuable and versatile tool.

Rest assured, your customers are open to using conversational tools. Many of them don’t always need or necessarily want to speak to an agent, so having an IVR with NLP and NLU integrated can bring excellent value both to your business – and the user experience.

Is your current IVR system negatively impacting the customer experience?

If you are using an IVR system without conversational integration such as NLP and NLU, there’s a good chance that it is negatively impacting your customer experience.

Certainly, if your IVR was set up well it might not be, however, traditional IVRs do rank poorly with the vast majority of customers.

So, how can you tell if your customers aren’t happy with your current system?

• They’re constantly complaining in surveys and to your agents.
• They regularly ‘zero’ out of the IVR system in favour of an agent or terminate the call immediately.
• Your IVR has a low FCR rate (first contact resolution).

If any or all of the above apply to your situation, don’t panic; it’s entirely possible to fix your IVR system without scrapping it altogether. For example, if customers are abandoning specific pathways then it’s likely because they are being led in the wrong direction and reach a dead end before having their query addressed – in which case you must test and optimise your system to prevent it from happening in the future.

Of course, the only problem with this is the fact that it is time-consuming. The best alternative is to provide your customers with speech interface options in addition.

Read our blog about IVR and how it can help your call centre

Do you want to preserve your brand image as best as possible?

There’s nothing worse than a brand that doesn’t invest in its customer service. I won’t name-drop, but I can think of a dozen brands off the top of my head that I would never touch again because of how poorly their customer support systems were organised.

Don’t fall into that category.

Just as it’s important to market your business well and attract the right customers, so too is it to invest in a customer support system that serves them well.

The fact is, practically everyone (some 96% of us) agrees that customer service plays a huge role when we make a choice in which brands to use and remain loyal to.

As such, it’s important that you upgrade your dated DTMF systems and implement the latest technologies to streamline the self-service automation process. Otherwise, as more and more businesses implement a conversational platform in their customer service, you’ll start haemorrhaging valuable customers.

Will conversational IVR be able to cut costs for you?

Yes, conversational IVR is a sizeable investment, but the question is, will investing in it ultimately cut costs for you?

When it comes to reducing the number of agent-handled calls while simultaneously increasing the overall automation of your caller experience, the immense cash-saving potential that conversation IVR has to offer is irrefutable.

You’d be surprised by how many enterprise-scale businesses have managed to save millions of dollars a year by simply reducing the amount of strain that they put on their call centres.

That’s not to say that you should use conversation IVR as a barrier to prevent your customers from getting through to an agent if you need one. However, offering your customers an easy and user-friendly method of self-service when most enquiries can easily be resolved without an agent, will indeed save you money in the long run.

Conclusion

If the answer was yes when you asked yourself the four questions listed above, then it’s a clear indication that conversation IVR might just be the solution you’ve been looking for.

Still unconvinced? Why don’t you give us a call today and we can answer any questions that you may have?