Contact center analytics

Looking to reduce cost of operations and increase customer satisfaction? Tap into your contact center analytics for valuable insight.

Analytics have transformed the contact center from being just a service tool, to a key weapon that can influence future processes and marketing strategies. It’s a shift not dissimilar to Clark Kent turning into Superman. Minus the little red pants of course.

Every business leader knows it’s important to put the customer first. But how many are actually analyzing customer satisfaction, customer interactions and in the weeds looking at vital information?

And if we aren’t even regularly looking at contact center analytics, how can we manage resources more efficiently, or find ways to save time and money?

First, let’s look at what kind of data contact centers need to look at.

Defining the data needs of contact centers

Generally, contact center data can be loosely split into three areas.

1. Agent performance

Alongside things like average call length and number of calls taken, how are your agents actually performing on the call? If you don’t have call recordings and transcriptions, this can be hard to do because you won’t be able to see—err, hear—how the conversations go.

Some software can transcribe conversations (in real time!) between call center agents and customers and give contact center managers insight into how those calls are going, without them having to jump in to help.

That’s exactly what Dialpad Contact Center can do, thanks to its Voice Intelligence (Vi) technology. Not only can it transcribe calls more accurately than Google, it also shows you the sentiment of each live call your agents are on:



2. Customer needs

Finally, data sources can help identify customer needs (and even potential sales opportunities). It’s obviously important to keep an eye on customer happiness and make sure your contact center’s response rates to queries are good.

But you can also use this data to upsell, or update your marketing strategies in the future. For example, Dialpad’s Vi-powered Contact Center platform can pick out keywords or topics that keep coming up on customer calls—which you can then use to train your agents so they can answer these questions easily.

Namely, you can create Real-time Assist (RTA) cards with tips for each of these keyword topics, which will automatically pop up on your agents’ screens when those trigger words are spoken on a call! It’s a like cheat sheet for your contact center:


We think it’s pretty awesome, but don’t take our word for it… Here’s a review from G2:



3. Metadata

Firstly, let’s look at metadata.

This is the nitty-gritty info surrounding a customer call, including things like the time of day, length of wait time, and the duration of the conversation itself. This data can help managers budget their resources and schedule more efficiently so more agents are online at the busiest times. (It can also show up areas where your contact center needs to put in serious work.)

What can you measure with contact center analytics?

There’s so much data that can be measured with analytics. But every metric is meant to measure something specific—and not all of them are relevant to your goals.

Always start with a clear picture of what’s important to your contact center. Otherwise, you’ll just be chasing numbers that might not have an actual impact on your business.

As a starting point, let’s look at some common things contact centers typically analyze.

Abandoned call rate (ACR)

Abandoned calls are those where customers drop out of a queue before they reach an agent, or during the time the self-service (aka IVR) menu is being read out. It can also include calls you get during “closed” hours.

In Dialpad, you can easily see how many abandoned calls each agent is getting:



There can be many innocent reasons for an abandoned call of course—a customer might hang up simply because nature has called.

Because of this, you can also set a threshold in Dialpad so that if someone abandons a call unusually quickly (like after one or two seconds), it doesn’t affect your service levels:



But analytics can help identify trends in abandonment rates that are telling you that your system is not working efficiently. (Maybe your queue times are too long at certain points of the day, or your self-service menu is too long winded.)

Average handle time (AHT)

As you’re setting targets to match your customers’ expectations, you’re probably going to want to keep an eye on the average duration of your calls, or “handle time.”

You can see this in Dialpad Contact Center under your Analytics dashboard:



Reducing this can positively impact your other metrics too. The shorter your call handle times are, the less your other callers have to wait—which results in a better customer experience.

It’s not always bad if calls are taking a longer time! Yes, sometimes it’s a sign that your agents need extra coaching, but if you have a contact center that’s specifically for more complicated or difficult questions, this is not necessarily a good metric to judge your agents’ performance by.

👉Dialpad tip: Don’t look at analytics in a vacuum. (For example, a high AHT combined with great first call resolution rates tells you there’s a story to dig into behind the numbers.)

Average hold time (AHT)

The amount of time a customer is waiting in line for someone to pick up. There probably isn’t a single customer on Earth who enjoys even a single second of hold time, and this is a valuable stat to have at your disposal.

Having access to this data can help you analyze if your self-service menu is working effectively, if agents are taking too long over certain queries, or if your contact center is understaffed during busy periods.

If you’re worried about staffing, Dialpad also has a heatmap feature that shows your call volume trends, which you can use in addition to your AHT to make a case for hiring or staffing more:



Average transfer rate

This metric can show you how effective your routing system is. A high average transfer rate might indicate that your customers are more disgruntled (and if it correlates with your average call duration times going up or increased complaints about waiting times, then that’s a sign you need to make some changes).

You may need to cross-train agents so they’re empowered to deal with more queries, or improve your call routing system so less calls need to be transferred in the first place. First call resolution (FCR) is a good companion goal to look at here.

Customer satisfaction

How do you measure customer satisfaction? Well, it’s worth tracking this in a number of ways. As we mentioned earlier, Dialpad can show you customer sentiment on calls through its Vi technology, but if you want to go the extra mile, use a satisfaction survey.

You can do this in minutes right from your Dialpad dashboard:

From there, you can customize your CSAT script and questions:


👉Dialpad tip: If you’re going to do a CSAT survey, warn your callers in advance at the self-service menu stage that they’ll be offered a survey once their call is finished. And to respect their time, keep your survey relatively short. If you need answers to multiple questions, you can always break it up and run a different survey next month or quarter.

4 benefits of tracking contact center analytics

Now, let’s look at analytics in practice and three specific ways in which tracking this data can influence your bottom line.

1. It helps you reduce call volume (or staff well for high call volumes)

You can’t always reduce call volume. You can try, by making it easier for customers to find the information they need elsewhere, of course—but if you’re in the travel industry during Christmas, good luck reducing call volume there.

If you have analytics showing you when calls spike though, that can at least help you adjust your staffing so that your agents aren’t overwhelmed. Dialpad has heatmaps showing call volumes, like we mentioned above, but there’s also an option to see your average speed to answer:



This way, you can see not only when you’re getting more calls, but also how your team is responding and handling this increased call volume.

2. It can help you improve your average handle time

When you’re looking at agent performances and trying to find ways to empower them, one key thing to keep in mind is: Are you giving them the information they need to handle their calls—and making it accessible?

Many call center solutions let agents do the basics. Take calls, transfer calls, put people on hold.

A really good contact center solution will have other features designed to improve handle time, like integrations with popular CRM (customer relationship management) tools, or at the very least APIs (application programming interface) so that you can build these integrations yourself.

This opens the door to things like automation—for example, allowing them to take calls right in their Salesforce or Hubspot dashboard while seeing that customer’s information right in the same window:

👉Dialpad tip: One neat thing that Dialpad’s Vi technology can do is identify the excessive use of filler words and the speed of speech on a call—and pop up an alert to tell the agent to speak more slowly, for instance.


3. It helps you identify areas for coaching and training agents

A contact center is only as good as its agents. Are you giving your agents the skills and tools they need to do their jobs? Initial onboarding is one thing, but six months or a year down the line, things will probably have changed.

This is where having access to KPIs and metrics is essential. Specifically, if you notice that your agents are taking a long time on calls to address certain questions (maybe it’s a new product or feature), you can take those analytics and add that information to your training materials.

For example, as we mentioned earlier, Dialpad’s speech analytics can tell you when an RTA card might be useful to help agents talk about tricky topics like pricing or new features.

👉Dialpad tip: Numbers and stats are important, but combining call data with real agent experience gives you the full picture. Having insights to both sides of the coin allows you to offer praise as well as identify training needs.

4. It can help you boost in service-to-sales conversion

Most people think of a call center operation as a customer satisfaction play: just a way to keep customers happy. But of course, it can be a great sales tool too.

With a good call center software, you can create in-depth customer profiles that include their preferences, past products they bought, and more. Every query they’ve ever had can hold a nugget of insight that could become a new revenue opportunity.

FAQs about contact center analytics

What are “contact center analytics,” exactly?

Effective contact center analytics would generally cover the performance of every aspect of your operation. You don’t necessarily have to buy separate analytics software to get this data, which can include things like call volume, call duration, and other key performance indicators (KPIs).

Having these analytics is important for both large and growing contact centers, because it can give you crucial insights. (For example, it can show you areas where you can introduce more customer-led processes into your strategies or influence future decision making.)

What are some key contact analytics features to look for?

Common call center analytics tools will help you track metrics like the number of calls you receive, missed calls, time on hold, call duration and the number of calls taken by each agent.

And yes, this is essential data to have for allocating resources and getting an overview of the individual performance of each agent you employ.

But robust analytics solutions can do even more. Dialpad’s Contact Center, for example, can show you analytics of how often certain topics come up on calls (you can create “Custom Moments” to track as many keyword terms as you like):


This is essentially an automation that scans your calls for these important insights, which helps you identify the root cause of any issues—without lifting a finger.

Why do people have so many different names for contact center analytics? Aren’t they all just measuring similar things?

If you’re not familiar with analytics or customer service tools that have robust analytics built in, things can seem hazy at first. For a contact center especially, there are so many numbers and metrics you could keep track of and everyone seems like they use different lingo. (And customer satisfaction is definitely not the be all and end all!)

Here are a few main categories of analytics you might hear about:

1. Cross-channel analytics

A modern contact center needs to have an omnichannel approach to customer service. There’s no way around it. If a customer uses social media for their first point of contact, then goes on live chat and finally calls you, the same high level of service should be experienced throughout.

It’s essential that your data analytics platform is able to gather statistics from every channel, from social media to phone calls to live chat. Not only can this data be used to identify the most popular customer touch-points, but you’ll also be able to identify different levels of customer satisfaction within each channel, and then make the necessary improvements.

2. Speech analytics

Speech and interaction analytics can give business leaders a serious superpower, and the list of actionable insights you can gather from it is one of the most valuable types of information you can have.

What better way to understand how your contact center is performing than by looking at the conversations that are going on?

👉Dialpad tip: To get speech analytics in the first place, though, you’ll need some way of recording this speech—whether it’s through a call recording or transcript of the call.

Some contact center solutions, like Dialpad, can use speech analytics to help you coach agents, identify purchase trends, monitor customer engagement with certain products, and even identify future sales opportunities.

3. Predictive analytics

You’ve guessed it, predictive analytics are used to (try to) look into the future.

It can come in handy in a few different ways. You may be able to analyze patterns in calls with customers historically. With Dialpad, for example, you can create a Custom Moment of your competitor’s name. You’ll be able to see when your customers mention this competitor—and create some talking points for your agents!

Predictive customer analytics can also be used to identify sales opportunities. Maybe you’re running a sportswear company and a customer—who previously only purchased running shoes—mentions that they’re going rock climbing. This info can help you create targeted offers in future (in this case, for climbing equipment).

4. Self-service interaction

Traditional contact centers routed all queries through to human agents. Customers would have to wait the same length of time for a complex call as for one that required a five-second answer. Not very efficient.

Modern contact center software lets you do things like create an interactive voice response (IVR) menu, which lets callers find answers on their own: self-service!

This way, your callers are able to answer many of their own questions without bugging your team, but you can still have your agents on hand in case a tough question comes up.

One quick way to analyze the effectiveness of your self-service options is to see if simple questions are still coming through to your human agents. If they are, then you might want to rework your IVR’s options or choices. (Maybe the language is unclear, or the menu you’ve set up is too complex.)

How will you use your contact center analytics?

For omnichannel contact centers and growing teams specifically, having access to analytics is essential to improve the customer journey and make sure every customer interaction is as efficient as possible.

Whether you’re using text analytics. By using data-led insights, staying ahead of the curve and adapting to industry changes as they occur is made much easier, increasing your chances of maintaining high levels of customer satisfaction.