Chat Now
4 min read

IVR Optimization in Multichannel: 9 Steps

Featured Image

In a lot of ways, whether you’re designing a single channel service or multi-channel service, the principles of user experience (UX) are not too different. For example:

  • Understanding and designing for user needs and expectations
  • Iterative design and evaluation
  • Evaluating designs with representative users
  • Multi-disciplinary approach

But the ‘multi-channel’ aspect of self-service does pose some additional challenges and points to consider.

In this post, I summarise 9 steps for ensuring early and continuous focus on UX for multi-channel service design…

1 Plan your UX integration

Good user experience doesn’t just happen. It takes planning, time and budget to make it happen. This is where having a structured approach to integrating the right UX activities within the project process is critical.

2 Ensure you have the right skills in place

In order to do the UX activities in the project plan, you need to right UX skill set in place. For example, if your multi-channel service includes an IVR, Smartphone and Web component, you’ll need a Visual Designer, Voice User Interface Designer and UX researcher to support the iterative design and evaluation process.

3 Understand the business goals

When planning the right approach to UX, it’s essential that you understand what the business goals are e.g. to reduce cost to serve? Reduce Agent Handling Time (AHT)? Improve CSAT or NPS (Net Promoter Scores)? Or is the business moving in a completely new direction?

All these will have a bearing on the kind of UX approach taken, UX measures collected, and how you define the multi-channel service to be a success.

4 Design for your customers

It’s critical that your business goals and customer needs don’t clash.

For example, while a business goal may be to reduce the cost to serve and reduce the number of calls into the call centre, it’s important that this isn’t at the detriment of what customers need or expect. For example, a task like adding credit to a pre-pay phone lends itself far better to automation, than say, reporting a lost or stolen phone. In this situation, the task is much more emotive and likely to need the support of an agent.

A successful multi-channel service is dependent on understanding customer needs and expectations, their tasks, context of use and in providing a suite of touchpoints that support the right kind of tasks using the most appropriate channels.

5 Selecting the right channels for your customers

Not all channels are right for all tasks. A well designed, multi-channel service will be based on an understanding of what customers want to do, the context in which they want to do it, and offering the most appropriate channel for supporting these tasks and contexts.

In a previous post ‘Is Facebook to right for Self-Service?’ I presented some considerations when deciding whether Facebook is right for self-service. For example, the relevance of Facebook as a self-service channel will be dependent on your customer demographic. Do your customers actually use Facebook? If not, then there may not be any benefit to spending time and money here.

And while it may be possible to provide automation within your IVR for customers reporting a lost or stolen phone, the reality is that in these kinds of scenarios customers want to talk to an agent.

The key message here is to make sure you really understand what your customers need to do, why, where and when, and use this insight when selecting the most appropriate channels for your multi-channel self-service design.

6 Don’t design in silo

Another factor that will determine the success of your multi-channel service is how well the different touch-points integrate and support a consistent user experience. This isn’t just about using the same branding and terminology, but really linking the touch-points and personalizing a customers experience with them.

For example, if an existing customer uses the web to get a renewal quote for their car insurance, and then calls the call centre the following day, a consistent user multi-channel experience would be one that could link these two interactions and provide a personalised welcome within the IVR, e.g. “Hi, I see you got a renewal quote from us yesterday. Is that what you’re calling about?”.

An holistic approach to all aspects of research, design and evaluation is essential. Visual Designers, VUI Designers, Developers and UX Researchers all need to be working closely together throughout the design and evaluation process.

7 Early and iterative design and evaluation

Not even the best designers will get things right first time. Prototyping allows you to test out early design concepts, identify and address usability issues early on and even compare alternative design solutions. Doing this before development begins reduces the risk of costly and lengthy rework at later stages of the project.

Prototyping can range from paper prototyping, to emulators and more functional simulators. Even starting with very basic paper prototypes can provide important early insights that can radically influence the design strategy.

8 Measure UX Objectively

It’s not uncommon to see ‘improve user experience’ in a business requirements specification. But what does this actually mean? And how do you know if user experience has actually been improved?

Benchmarking the existing UX at the start of a project will provide a baseline measure against which future designs can be compared.

And it’s not always necessary to have existing touch points in place to do this. Evaluating task-based CSAT for the channels customers currently use for carrying out their tasks can also provide useful measures and insight. For example, a baseline measure of tasks supported within an IVR can be used as useful way of indicating comparative experiences for the same tasks carried out using the new channels such as web and Smartphone.

Many companies already have CSAT or NPS (Net Promoter Score) processes in place that can be used to indicate general user experience. Repeating these surveys during the project process provides a consistent and tangible way of indicating the impacts on UX (though care needs to be taken to ensure that changes in CSAT or NPS aren’t due to other external factors). It’s recommended that this is done in addition to channel-specific UX measurement and qualitative research too.

These types of measures are useful in setting target UX metrics and goals, measuring ROI and in evaluating new products and services against these measures to ensure that the impact on UX is based on objective and tangible criteria.

And finally…

9 Ensure your UX strategy doesn’t stop at project release

When a project moves into the managed service phase, ongoing UX measurement and evaluation is just as important. In my next post, I provide 4 steps to ensure the optimisation of your multi-channel service during the managed service phase.