How to Gather Customer Friction Data in a Contact Center

Contact Drivers

Think about how many contacts your center handled last week. What did your customer say? There are a surprising number of call center leaders who cannot answer that question. Yet knowing the root cause of what drives customers to reach out to you is crucial to both your call center leadership team and your company overall.

Within the call center, your planning and budgeting teams can use this information to better forecast and staff to volume. Training teams can focus on the main questions your agents will face. Your leadership team can use data around customer friction to build bridges with other departments within the organization and help you get a seat at the table whenever the customer experience is being discussed.

When customer friction data is shared with the rest of the business, it gives measurable ways to determine overall customer experience. Once the data is understood, various departments will want to receive regular updates and may even chose to use some of your data in their department goals.

So what are the best ways to gather customer friction data? Below are what I consider two bad options and three good ones.

IVR Queues

You might be tempted to think you are getting this information through your IVR system. The problem here is that regardless of how sophisticated the IVR, some people will say anything to get to an agent quicker. Or they will select the dreaded “Other” option. This often results in “dirty queues”. For instance, a customer may have indicated they want to place an order when really they are calling to figure out why their technician hasn’t arrived yet.

Also, depending on your technology, your phone channel may be the only type of contact that has any type of IVR disposition. If integrated into your routing and skills structure, emails, social media contacts and chats often are their own queue or skill. Thus you have no way of knowing the reasons customers contacted you through those channels.

Frontline Feedback

A frontline contact center rep normally takes between 30 and 100 interactions per shift. That means within a week they talk to roughly 150-500 customers. So why don’t you just ask them why customers are calling? Well for several reasons.

  • Agents often fixate on one problem and exaggerate its volume and severity – “Customers always have problems logging into our site”
  • Most agents don’t speak in technical or business terms – “They had an issue checking out”
    • IT wants to know everything about the customer’s software and hardware.
    • The business wants to know how much money this problem is costing and what percentage of customers are impacted.
  • Depending on the size of your business, even talking to a dozen agents may be too small of a sample size to inspire action from those outside of customer care.

Speech Analytics

Many contact centers utilize speech analytics to help with Quality Monitoring. While this can be an excellent way to establish how your agents are treating customers, it should also help you identify why customers are reaching out in the first place. However, this approach is often costly as most models involve agent licenses or cost per minute to turn the speech into text. You also have to hire someone to manage and refine the system. The main advantage here is that you can analyze EVERYTHING that the customer told you with no additional work for your front line agents. Depending on the delay between the live interaction and you receiving the data, you can identify the root causes for spikes in volume. How would you like to be able to tell your digital team, “In the last half hour, we’ve received 642 customer inquiries about product # 56471. In all but 2 of those interactions, the customer said they wanted to add the product to their cart but could not.” Overall, I believe this is the best solution, but not everyone can get it into their technology budgets as it often requires additional fees and headcount.

The Call Path Solution

Some call centers utilize either a homegrown or purchased call tree that walks their employee through each interaction. Often the call tree will be linked to key systems that either will pop up at the correct moment or update behind the scenes. The main reason for using such a product is universalizing the experience and ensuring that necessary steps are always taken. Another benefit that is often overlooked in these systems is the customer friction data you can mine from them. Once again, there is a cost here and there needs to be someone managing the system. Also, compared to speech analytics, there is a chance that agents use the wrong path to get to the right outcome, thus skewing the data.

The Survey Solution

For call centers that do not have the resources to build their own program or the money to purchase speech analytics, there are inexpensive survey options. Most people are familiar with creating a survey to send to customers or even to employees asking about their experience. However, you can also create a survey that asks your agents why the customer reached out. This is relatively inexpensive and can be edited easily if new call drivers become known. Also, most survey tools allow you to create as extensive of survey as you want. If you want to know more about a specific call driver, you can always create another page of the survey for the agent to fill out. The main negatives here are agent time (it will impact AHT), accuracy in filling out the form and getting the agent to complete the form on every interaction. While not an ideal solution, it is one that is easily implemented at a low cost.


Regardless of how you gather customer friction data, it is crucial for you to both know why your customer is reaching out and to share it with the rest of the business. Contact centers who do this well are looked to as the Voice of the Customer and often have a greater impact on the overall customer experience. Your influence within any company claiming to be customer centric will grow when you can walk into a room and say, “We spoke with over 70,000 people last week. Here is what they said.”

Leave a Reply

Your email address will not be published. Required fields are marked *

I accept that my given data and my IP address is sent to a server in the USA only for the purpose of spam prevention through the Akismet program.More information on Akismet and GDPR.