Saturday morning, 9 a.m., the children are quietly playing and listening to some background music in the living room. I think it would be a great time to contact the customer service department of my cell phone company to negotiate my new plan. There is only one important factor to consider: I am aware that in order to get the best price, I will have to contact the customer retention department and that it is only available by phone.

I still decide to call. I get the line after a few minutes (it’s usually quick to get a line on a Saturday morning). However, the discussion drags on as the representative presents me more and more offers that are not to my satisfaction. After more than 30 minutes of discussion, my level of patience is getting dangerously low, especially considering that my children have been asking me to make them breakfast for several minutes already. After 45 minutes on the phone, my children start kicking and screaming and I know I will unfortunately have to call back later. So, I ask the call center employee to put the information related to our discussion in my file, even if I strongly suspect that I will have to start from scratch the next time I call (probably when my children are in bed on a weeknight).

Does it really have to be this way?

In an ideal world, here’s how things would have gone:

I would have contacted my provider via chat from a chat agent on their portal or by text message. Also, seeing as I would have contacted them from their secure portal or from my cell phone, my account security review would have been simplified and accelerated.

Moreover, I would not have needed to contact a representative. In fact, I often wonder why it is necessary to talk to a human to simply have them offer me packages that are “pre-made”. I know very well that the person on the other end of the phone has a list of options and is following a script. We could achieve exactly the same result, and with less frustration too, if the conversation took the form of a gamification process that would take into account my loyalty to my provider.

Even better, I would have had the opportunity to “pause” the process and resume later on the same channel or— even better!— on another communication channel of my choice. In the end, everything would have been activated automatically.

Review the journey to propose a multi-channel approach

The example above is a real-life case, and there exist many others. You probably have your own stories about the same subject.

Let me ask you a simple question: if you were to contact an organization’s customer service department, what would be your first reflex? I’m pretty sure it wouldn’t be to pick up the phone and call the company.

For many years, customers have been trained to call when they wanted to contact customer service. Times have since changed and the telephone is increasingly being abandoned as a means of communication. In 2020, the firm CommBox reports that the preferred means of communication is the written word. They have identified in order the most popular conversational agents: Facebook, WhatsApp, SMS, email, and Twitter.

In addition, customers are becoming increasingly demanding and insist on getting immediate access to information. Indeed, in 2016, Business Wire reported that 60% of Generation Z customers will hang up if, only after 45 seconds, no one answers. I haven’t found more recent data on the matter, but if I take a look at the people in my inner circle, I’d wager that the level of patience is not on the rise.

Put simply, multi-channel approach has become essential, because that’s what your clients demand. On the other hand, even though customers have been shifting their preference to new channels for several years, many companies yet to follow suit and as a result find themselves disconnected from their customers.

How can you build an effective multi-channel approach?

There are some fundamental elements to keep in mind when defining an effective multi-channel approach.

Element #1 – Human or Conversational Agent?

This is usually the first question to ask oneself when setting up a multi-channel path. There is no golden rule, but choosing a conversational agent is usually the best idea, as it allows for a consistent, fast, and efficient experience. For more information, I suggest you consult my previous blog on the subject.

Element #2 – Not all channels are the same.

Voice, SMS, emails, or even conversational agents are not identical means of communication. Each one has its strengths and weaknesses. It is therefore normal and important that the proposed journey be different for each of them.

For example, it is possible to perform certain actions with a conversational agent that it would not be possible to do by voice (e.g.: display a questionnaire or a web page). We must therefore keep this in mind when defining the journey or the functions that we propose to the user.

I have seen projects at large Canadian banks where the conversational agent proposed functions that could not be accomplished by the transactional site or by a call to the call center.

Element #3 – The choice of channel is not exclusive.

Since each user has their own preferences, it is important to avoid imposing on them the use of a particular channel, at least initially. Technology is so advanced today that it allows the user to move from one communication channel to another without penalizing the journey. I would advise you against using this type of functionality in a first iteration, but it is something to keep in mind when building the journey.

Indeed, it is possible to add new channels after a first iteration. For example, in a project, you could decide to set up a conversational agent and to then add voice, or Facebook. When adding a new channel, you need to analyze the features that are exclusive to it as well as the existing features that you can potentially reuse. At the end, you must look at your entire journey and determine if it would make sense to propose a change of channel to complete it.

One of the most popular track is voice to text. Think about it: almost everyone has a smartphone today. It therefore makes it quite simple to transfer a call to an SMS channel in order to complete certain requests.

Element #4 – There is no “best channel”.

As I often tell my clients: there is no best channel that stands out from the rest. However, some channels may be more or less relevant to your users. To determine the best choice, it is therefore important to take into consideration the user experience, to carefully identify when there will be interaction with the user, and to understand the needs we are trying to meet.

Element #5 – Setting up “one-to-one” channels

A common mistake when working on projects is to want to do everything right away. In this kind of project, where artificial intelligence is used, it is important to remember that the secret weapon is the training. Since each of the channels requires specific training to bring the machine learning model to an acceptable level, focusing on several channels at once will greatly increase the amount of effort required.

Similarly, the more components to train, the greater the chance of not achieving acceptable results. I have seen this type of project fail more than once for this very reason. Remember that it is better to create a “wow” effect on a small project than a “meh” effect on a large project.

Does it work for real?

In the last few months, I have worked on several projects that have proved me that it is in fact possible to do this kind of thing. However, I won’t hide the fact that there is a certain “trick”. Indeed, this kind of project is a classic example of digital transformation and, as I always say, there are several ingredients that will contribute to its success:

1. The tools – That’s the foundation. You have to have technological tools that can make your ideas come true and that are easy to integrate.

2. The method – That’s the secret. This kind of project requires skills and a proven approach to succeed. For our part, we use a method based on the IBM Garage approach (https://www.ibm.com/garage/method/).

Moreover, it is important to look at the human aspect of the project and to put extra effort into change management. It is often a neglected aspect but paying attention to it makes a great difference.

3. Expertise – It’s your guardian angel. I have no doubt that you are capable of doing it yourself, but you will surely spend more time and money than you had planned. I suggest that you get people who have already worked on such projects to help you, especially if this is your first project of the kind. It will reduce your risk as well as the amount of time you will take to set up a product that works.

Are you ready to innovate?

I’m passionate about this kind of project and I’d love to hear about your achievements, your good moves, and your failures. It is by sharing our common experiences and knowledge that we will make scientific advancement in this field.