Invoca launched Signal AI in 2017, revolutionizing call analytics, call classification, and conversion attribution for marketing. It’s been a big win for our customers, but we haven’t taken that as an excuse to take a vacation. Signal Discovery is the latest enhancement to our conversational analytics suite and a first-of-its-kind tool that provides marketers with a complete view into their customer conversations, helping them access new first-party data and insights.

Why Marketers Need Conversational Analytics

How do you find something when you don’t know what to look for? Imagine that you were dropped into a giant, strange warehouse filled with millions of objects and you were directed to find “the thing.” Without knowing what the “the thing” is you’d have to pick up every single thing and ask “is this the thing?” until you found the right thing or lost your mind trying. 

This is a lot like the world that marketers live in when they are looking to understand what’s really going on when their potential buyers call their contact center. If you were tasked with analyzing tens or hundreds of thousands of calls to find “other” types of calls, you’d find yourself in that giant warehouse, buried by conversations and piles of words while you scream “A FREAKING ROBOT SHOULD BE DOING THIS!” And you would be right, and that’s why we created Signal Discovery — to help you discover valuable customer insights that were once buried in your call recordings and develop new first-party customer data for you to power your marketing with. 

Despite the fact that many B2C companies drive most of their sales over the phone, more than 56% of marketers have no idea what’s said during the calls that they drive or what the outcomes of those calls are. “Conversations are overflowing with insights that don’t always see the light of day outside the contact center. As a result, many companies are missing out on opportunities to create a more consistent and positive customer experience across human and digital touchpoints,” said Dan Miller, lead analyst and founder at Opus Research. It’s a big data gap that marketers shouldn’t have to live with.

How Signal Discovery Conversational Analytics Helps

Signal Discovery solves this issue by enabling marketers to quickly gain new insights from tens of thousands of conversations and take action on them in real time. Signal Discovery shows you real conversation topics with your customers in a full-color map, eliminating guesswork and assumptions about caller behavior by providing you with hard data. “Signal Discovery shines a whole new light on conversations happening in our contact center, giving us the data we need to enact changes across the organization”, said Noah Brooks, Manager of Digital Engagement and Analytics, University Hospitals

The beauty of Signal Discovery is that it not only validates what you may think you know about caller behavior, but it also uncovers caller behaviors that you may not know exist. Pause to let that sink in… Imagine uncovering buyer behaviors that you don’t even know are occurring. Things like:

  • “I was treating all call-driven appointments the same in my paid search optimization efforts but I discovered that there were actually 3 different types of appointments coming in through calls and over HALF were related to customer service and rescheduling. Now I know how to optimize only for new business appointments.”
  • “I discovered that 34% of my calls driven by paid search are from existing customers calling with technical or other support issues. By optimizing my bid strategy, I cut our cost-per-order and I freed up our call center sales agents’ time to focus on other high-intent callers.”.”
  • “I discovered an unknown IVR issue that was causing some customers to spend a lot of time on hold, get lost in the IVR, or get sent to voicemail. I also found out that almost 20% of my calls were being transferred multiple times and often to the wrong agent. Now I know how to route calls and deliver the best possible buying experience.”

How Signal Discovery Works

At heart, Signal Discovery creates valuable insight by using learnings extracted from past conversations to detect occurrences in future conversations. Signal Discovery uses a type of AI called “unsupervised learning,” which means the technology automatically groups conversations into distinct topics without any human guidance or training data needed. It listens to everything in every call and then creates clusters of conversation topics based on similarities in speech patterns. From there, you’re able to drill down into each topic to understand caller behavior, see topic-level insights (including the top marketing sources, channels, and campaigns associated with the topic) and then create a “Signal” to detect the topic on new inbound calls, so you can automate your marketing, such as pushing data in real-time into Google Ads or Adobe Analytics. No more guesswork, no more risky call assumptions.

It all starts with a call map that looks like this:

Topic cluster visualization in the Signal Discovery conversational analytics platform.

All those colored bubbles are actually unique conversational topics. Each one is numbered — in the example above Signal Discovery has identified 83 unique topics. Each topic is also color-coded to signify groups of similar topics. 

You can hover over each topic group to see the most common spoken words for the group.

Topic filtering in the Invoca Signal Discovery conversational analytics platform.

In order to focus your analysis, you can filter your conversational map in a number of different ways, including spoken keywords, marketing medium, marketing campaign, and more. In the above example, we’ve filtered the topics to only display those featuring “mortgage.” 

Zooming in on a specific topic category in Signal Discovery to see what is going on in a particular group of conversations.

From there, you can click on a topic bubble to see the most common words spoken on that cluster of calls. A few things to notice: The largest words in the bubble are those spoken most often. Also, take a look at the information to the right of the topic bubbles:

Digging into conversation topic details like medium, marketing channel, and UTM source in Signal Discovery conversational analytics.

The insights tab provides you with key information about the topic — such as topic frequency and average call duration — as well as analytics on the digital touchpoints that are driving these conversational topics.

How to play phone call recordings in Signal Discovery.

Signal Discovery also offers a player where you can listen to sample call recordings to hear how the topics came up in conversation. After you listen to a phone call you can also add notes that you can refer to later.

Now that you’ve listened to and identified this buyer behavior, all it takes is two clicks to set up an AI signal to detect the presence of this topic on all new inbound calls.

The insights you uncover with Signal Discovery are all brought to fruition when you use Invoca’s industry-leading integrations to push conversational data to your preferred marketing platforms, so you can optimize your media campaigns, segment audiences, and drive revenue.

(since we know you like GIFs, here’s one more)

Will Signal Discovery make you jump up on the table and dance? Maybe not, but if it does, send us a video. What it definitely will do is provide you with a new source of untapped first-party data and insight into caller behavior that can make all your marketing efforts more effective.

Stacey Watkins

Posted by Stacey Watkins

Stacey is a Senior Product Marketing Manager at Invoca. One of her favorite parts of that role is talking to other marketers about the power — and revenue potential — of call tracking and how it can affect a business’s bottom line. Prior to her position at Invoca, Stacey held product marketing and content roles at LogMeIn, Citrix and Commission Junction.