Nectar Executive Interview
Joe Fuccillo, CTO, Nectar
Sheri Greenhaus, Managing Partner, CrmXchange met with Joe Fuccillo, Nectar CTO during Enterprise Connect 2022.
Sheri: Please give us an overview of Nectar.
Joe: We're in the monitoring business and we released our first product in 2008.
Companies need to understand all pieces of their environment. It's not enough to say there's a problem. Is the customer, the user, the employee getting the experience that company wants. Is this a quality engagement? Whether it's a conversation, an email, a chat experience in the contact center, or employees collaborating from anywhere, are they able to do their job right? Are the applications performing correctly? How do I gauge an experience? That's where the monitoring has evolved. Our focus is on communications because the necessary tooling is very different than a website and other applications.
People are always amazed at our logos. We have approximately 200 of the fortune 500 in our customer base. That's partly because we have a good channel program. Certain partners were able to build solutions, embedding our software into their managed services. For many of these customers we are integral to their operations center. We have a strong financial services base. For example, Deutsche Bank has been with us for over 10 years, using more and more of our tools. There's hundreds of people that use our tools daily.
The priority of the contact center is how to drive a healthy, happy contact center. Part of it is the agents. The questions are: How do you onboard your agents? How do you keep them happy? How do you make sure they're working and productive? Agent attrition is expensive.
There's a lot of investments in technologies.
Successful operations are really involved in three areas. Is the technology working correctly? Does it deliver a great customer experience and a good agent experience? While it sounds simple, there is a lot of complexity. You need a broad portfolio of tools in order to uncover any issues that a company may have with their technologies. There are vendors that may focus on one area or the other. You can find vendors who will test that your dial in numbers or menus that are not working, or test your average speed of answer, but they don’t touch the technology underneath it.
This is where we're different. As the industry is shifting from premise to cloud, the amount of technology that is the customer's responsibility for monitoring is dropping, but in the other two quadrants there's also a hybrid world. A lot of our large customers, for example, want to maintain the carrier access. They don't want to take all their phone numbers and port them. So they're keeping a session border control layer. They also have contracts with the carriers and those contracts typically include network and voice. If they just take all the voice out and lose it, it may affect their spend, their drops. There's a lot of conflict, a lot of reasons why customers are keeping their carrier contracts themselves.
The other area we focus on is the agent experience. We've all heard the agents say ‘my system is slow today’. Sometimes agents repeat the same questions over and over again. Sometimes they did not get a screen pop, or what started on a chat escalated to a call. Was the call transferred correctly? All these things frustrate agents and customers. When you look at our portfolio, we have specific modules that deal with all of these issues.
Sheri: You're tracking when the system's not working?
Joe: Yes. For example, we have a robotic version of Nectar, that's acting like an agent and we're sending transactions. When measuring how many milliseconds it takes for a Salesforce screen to pop up, we determine if it comes with the right information. Another part of a tool can simulate the customer greeting. The ultimate function of it is when I call, did I get the right experience? Did I hear the right greetings? And when I spoke to a person, did their screens pop up correctly? Was the call transferred correctly? We monitor the underlying technology.
Sheri: Are you monitoring the voice, or the connection of the voice?
Joe: We're not doing voice biometrics where we're looking for expression. What we're looking at is the quality we've all experienced. That's what is difficult for consumers today.
For example, we have integrations with the headset vendors. It tells if the boom is in the wrong direction. I can tell how many decibels the mic is picking up. In early fall we will be able to send a message to the agent in the browser saying your microphone is in the wrong position, drop it down. The headset is intelligent. They know when you're speaking into mute because they hear the voice coming out and they know mute suppressed.
Some may think it’s because the mic is broken. For example, some of us have wireless headsets. Can you distinguish between walking too far from your desk versus your wireless while your internet connection is not doing well? They both manifest themselves as choppy audio, but the reason is radically different. We're collecting information from all sorts of things and information is easy to collect. What's difficult is to put it into a meaningful context and to understand and isolate where the bad experience is coming from.
People may not be a hundred percent sure what the issue is, but I can tell you what it isn't. When you try to find a needle in a haystack, when you have all these possible things that can cause an issue, how do you detect what it might be?
Dovetailing into these are the modules of the DXP experience platform that deal with all these areas.
When we started the company in 2008, we were monitoring at the application level specifically for voice applications. That's still very viable to do today. Our unique capability is what we've developed with the industry today with call AI automation. We auto discover an enormous amount.
To drive costs down, we developed Auto Discovery. For example, our software will go in the Genesys environment. It will go through to the Genesys configuration server that knows everything about that environment. We go to the configuration server and we take inventory. We see what's in the system and learn all the components. Then based on what we find, we monitor it. It learns the environment consistently, and then it adjusted the dynamics. The value of that is, the software finds all this stuff globally and quickly. We have those knowledge modules, Cisco, Avaya, and Genesys for Microsoft.
Because of our age, there's a depth of portfolio that allows us to do everything on premise. Then we develop things for cloud. There are different modules. We've talked about this a little bit.
CX assurance is our testing of a contact center.
Web Interaction Manager allows us to emulate a web browser and interact like a user would; either on a chat system or with an agent and an agent desktop. These are the two key pieces that deal with experience management.
The Endpoint client is a new product. It’s software on the call center agent’s device and it's monitoring their internet connection health. It measures bandwidth and speed. It's like a speed test. It's very light, very quick. The goal is to provide three or four major use cases. One is in predeployment, you have a new technology and new agent. You want to make sure the environment's good. So rather than hoping things works, the software can create voice calls. It can log in. We can simulate the same types of transactions they'd have in the real world before they do it.
It provides a comfort level of what is and is not working.
Take an example of a network issue. The closer we can monitor quality in real time, the closer to understanding the network. If an agent is working on a network that is not good, it’s optimal to stop calls coming into that agent and instead let them answer chat and emails. Take them off of voice.
We had an instance where we watched two hours of an agent with 50 bad calls in a row. We can see it happening, but there was no mechanism to stop it. There was no automated way to deskill the agent. We developed a real time API that computes the agent’s health index score. Customers are using the health index to determine if the agent should continue to take voice calls. If an agent’s score drops, there's an automated way to deskill the agent so they don’t take the call. That's one of the most exciting things we brought to market in the last year.
Sheri: What happens with that agent base? Are they taken offline?
Joe: The workflow would be to deskill the agent so I don't get on the call instantly. The next step is to use an API and Nectar to fire up synthetic transactions and test more aggressively. If they're not healthy, the health index is going to stay poor. The agent gets a message of what happened to them. When the index improves, they'll go back into a voice channel. While waiting to return to voice they can answer emails and chats.
The last thing you want to do is have a customer waits on hold for a few hours. This component is really exciting for the customer base, particularly in the COVID world. It's this combined with the quality monitoring. When you look at our tools, we have an operational advantage. We have an end-to-end capability. Nectar's one of the only platforms that truly can talk from the customer to the agent and all the big components in between.
These capabilities you see on the slide here analyze different components of our solution, analyzing different parts of the customer journey. For example, we have a SIP module that deals with that session border controller and there's approximately 60,000 concurrent calls coming into this session border control network. And then anything that happens there affects every conversation. We have this diagnostics model where the SIP analyzes all the SIP messages and all the media stream going in and out at high speed. It's creating a measurement saying its good here or at the border. Then from there it goes on its journey. Nectar’s analyzing that and supporting it all the way out to the agent desktop.
Each of our modules deals with a piece of either the media or the applications. Then the testing side allows us to look and take a step back and say, is it really working from a customer experience? If I dialed in and I go through this SBC and I hit this application, did I get the right manual? Was it audible? Can an agent answer? Did the screen pop? Can they work from home? And that's why we feel we have the most comprehensive platform for the contents. Customers can get some of these pieces from other vendors, but it won’t be complete.
Sheri: Does anybody else do the whole thing?
Joe: We are the most complete solution.
Every dot you see on this slide represents an experience where the employee was. Geospatial is really important; knowing where they working from. Some of the things we correlate is where you are in the world and who your internet service provider is, and what's the quality of that ability? Because one of the things customers will say is, “I have an agent that's having a problem on Comcast” and wherever they might be located or which internet provider they’re using, is it global to them? One of the things our AI engine does is to look for patterns as well. There's a whole lot of people having a problem at the same time, similar timeframe with a similar things. So that gets highlighted in the program saying, maybe moving towards is the ability to say this is a hot zone. You may want to let employees in this area know proactively, you might experience some issues based on what they're seeing in your area.
Sheri: That’s the health index?
Joe: Yes, this health index is a single measurable number. One number like a score. You're green you're good. And so there's a range and that helped them. This doesn't just apply to contact centers. We compute it for every employee, every user of our products.
Sheri: The employees themselves get to see what it is?
Joe: They can, but it's a customer choice to let them to log in. We're actually releasing a simplified view of this page in the spring that will make it a little less technical. We want to change some of the language to make it a little more insightful for an end user. Another thing they can see is a Teams meeting or Zoom meeting. One of the things we didn't mention is that our platforms help index: Teams, Zoom, WebEx, Microsoft, your premise, your buyer, or Genesys. For the amount of what Nectar is uniquely positioned for there is very few vendors in our space have the ability to support that many platforms as we can. We have so many global customers because they have a little bit of everything.
The old and newer systems, they want to have everything consolidated. They want a score card. They love the idea of a score card that's in all their platforms. They also can benchmark user acceptance. How's Zoom more successful than Teams? Many of our customers have both. Is there a different digital experience?
Sheri: Is this one of your reports?
Joe: Yes, that's a meeting screen. And the color coding is how their experience was. This was when they joined, they left, if they use the camera or voice or screen share, and there was 12 people in the meeting. If you clicked into one of those screens, it takes a deeper dive so the IT people can go in and see detail like who’s sitting there and what headset they used.
Sheri: So it really digs down?
Joe: It gets really granular. There's somewhere around 3 to 4 hundred individual metrics per person, per call.
Sheri: So, does it only show them issues?
Joe: We sift through all of it and then we categorize them. The idea is that health index is computed continuously on all your experiences. If you start to have bad ones, your score starts dropping. The more bad experiences an agent has the faster it drops.
Sheri: Then can you pinpoint where the issue is?
Joe: We can notify who gets alerts. Now we can notify the end user directly, IT operations and of course executives to C-suites. IT would like to know if their CEO is having a bad Zoom meeting in real time. Depending on the impact of the conversation, that's why contact centers are going back to that. It's so important because they're talking to the customer engagements and then what the product works for us are a lot of what most of our large customers will use us in their contact centers and all their global communications.
This is an Endpoint. This gets into the speed of your bandwidth your devices. We're getting unique metrics from each device, from your headset, from your wireless network, and the application of the cloud. That's why there's so much information coming in, but that's also why we've purposely built generic tools.
Sheri: Is there anything else you would like our audience to know?
Joe: A key point is the audience thinks going to the cloud will cause all these problems to magically go away. They think it's going to be simpler but it actually becomes a little more complex because they lose control of certain areas. They lose insights to some of the things they used to see. We need to have ways of measuring and monitoring experience and being able to deliver it and budget for it. As part of your migration, if you don't have the right tools today, you need to have them in place or you’re just guessing. You don't know where the problem might be. Monitoring for the health of your systems is paramount.