Verint Speech Analytics

One of the things that I find extremely beneficial about Verint Speech Analytics is the transcription. Sometimes you have an hour call, an hour and a half call, and you don't have time to listen to an hour and a half call. The beauty is that we can find the transcription and then we can look in it. It specifies where if you look in the interaction details in the back of the call, you can actually see the transcription. You can see it'll identify where in the call, what line at what timestamp did that call hit your category and with what string, with what term did it hit. And that is extremely beneficial. And we're trying to go through and find those little points of, sometimes they're points of abrasion, sometimes there are things that are going really well. I love that we have a transcription. I love that we can also see the screens that is vital because a lot of times, and again, not that we're trying to find something bad, but the reality is in any industry, there's in any call center environment, should I say, there's call avoidance. There's always going to be some agents, it's just in any call center, in any industry, it happens.

Or sometimes our agents are just struggling. And if you see in real life, for example, a very high silence time, when we notice a trend, we'll start going into those calls and actually looking at those screens and seeing, is silence time call avoidance or could it be the agents are having difficulty utilizing the resources? Are the resources not clear? Is that what the delay is? Are they having trouble trying to find the answer? And a lot of times we found that too. We found they're searching and clicking and searching and clicking and it's because they just didn't know where to go. So that is a coaching opportunity. And sometimes it's a coaching opportunity, not so much in a positive way because they're purchasing things from Amazon. But I think, like I said, that's inherent in any call center environment. I do find the screens, the transcription, the trend analysis that's available within speech, there are so many things I could literally sit here and talk to you for quite some time.

We love it. And I'm even more excited personally now because I learned so much, as I was saying, about being in the cloud and being in that open-source environment where all of these enhancements that Verint has. And they're so often that they enhance their products and their offering now these little bots and AI to help with all of these different aspects, but that's only available if you're in the cloud and the open source. So there's a lot of benefit in being in prem, in net platform, which is where we are. And we were able to provide a lot of meaningful insight and guidance into a lot of different pain points in the company. But I anticipate, and I have full confidence, that we'd be a much more effective team once we're able to move into the cloud. So I'm very hopeful for that. I'm anticipating what else we could discover.

When you're in design categories and you're adding a term or searching a term, there's this little area on the right-hand side, a skinny little area. I don't understand why that area can't be larger or more it's just a little box where you have the very important terms and a little box for important and a little box for avoid, but then the entire bottom of that column on the right-hand side is blank. So I always thought, can we somehow just make that little section a little longer to fill the page and perhaps allow us to have the feature to be able to alphabetize those terms in there. To be honest, in some categories that we're trying to build, such as I say Covid or ID cards, which is a huge one, there are so many different parameters we have to add in to exclude those conversations about let's say with covid conversations that were just educational or IVRs automated messages.

And there are so many different ways. There are so many different things that we have to do to avoid those conversations that we're not looking for. We're not looking for educational conversations that looking for internal conversations. We're not looking for training classes. Everything is recorded. But sometimes you have so many different terms and so much logic in that little area. You're scrolling and scrolling and scrolling and scrolling, I'm being honest, it would be so beneficial if we could please whoever's listening. If we could make that area on the right, extend it a little, and just give us the ability to alphabetize it. That's it. I don't know how difficult that is. I'm sorry I'm not in it, but if that's possible, I'd really appreciate it. That is the biggest thing that I could say that I deal with on a daily basis is that little freaky box on the right.

What we do is sometimes there are such large categories that we have to keep on stopping and then of course you save, but then we have to get out, export the terms, and export it into an Excel file. So then I can alphabetize it and then look at it and then go, oh man, I missed this one. And then go back in, do it all again, and then export again. That process of constantly having to export the terms and then alphabetize, not that it's difficult, it clearly is not, it's just time-consuming. It's repetitive and it's a little irritating when you're working on so many different categories at the same time. And the thing is, we're a very small team. Blue, like I said, we are somewhat new to speech analytics and I've taken everything. Speech analytics and I've absorbed it.

I love it. But being a small team, I utilize the time of the team as efficiently as possible because we're small. And if you really look at how many times we have to do that and stop and export and stop and export and stop, we'd probably save a lot of time. If we could somehow enhance that feature, it would be lovely. That would be a great addition and that would be fantastic. I tried to add it to the suggestion in the RIN Connect, but I didn't get enough votes. I don't think everyone looks at it because there are so many users of Verint. I didn't truly know how large Verint was and how many different industries were utilizing Verint technology. I really wasn't aware until I came here. Just give us the ability to alphabetize it and just make it a little bigger so we can actually see what we're doing. Once you put the terms in there, it's like in font five and it's a little bitty box. That's a great suggestion.

That's something that is still in the air because it's so old. Again, it's very manual. So I think for the most part we want something that's more technology-based that can identify those words in the transcription and give us a better readout versus just listening to each and every call. So I'm looking for this application that we have. Again, it's very old, but I'm hoping to find something that we can kind of innovate that will pick up the transcription for us.

I've heard a lot, so since I'm fairly new to the company, the organization, nothing but good recommendations. So I'm looking for this partnership with my organization in Verint to increase and grow over time.

With the proper phonetic boosting. It does very well as far as capturing and transcribing the conversations, so making sure that you have kind of your layers of the cake I feel is very important when it comes to something like the speech. So once you know that and you've got those layers in place, I think the transcription, we've had a lot of improvement. We just recently went through a phonetic boosting and we're already seeing upticks in some of that since we started getting the results. So I'm very impressed after that because we had some, you don't have those same English dictionary, even Allstate or mobile or mile wise things that we have that are unique to our company. We need to make sure that those are picked up and when the customers are saying those to be able to relate and pinging those categories to get the proper analytics and see those charts and see those call trends.

I feel like I could have answered that just last week, as good as you make it. So with that said, I'm kind of going to go back to getting with the NT team and fully understanding the way it works to know the best ways to go about it. So on that same flip side, I would say it was the transcription and then we've also had some call recording issues and some, because we've actually got a ticket open now for some duplicated calls and where it flipped the agent and the customer in the recording. So there's little bumps. So if I had to say any of those would probably be problematic, but just specifically for us, I can't speak to everybody, but the Verint team is very supportive and they're doing what they can to keep us in the loop and what they're doing to fix it. So we're doing, I think it's good considering

Transcription is excellent

easy to build categories in compared to competitors

Easy cross correlation

Better canned reporting

more widget space in dashboards

It helps with customer insights as to what they are calling us about

It was very resourceful during Covid

transcription data

accuracy of transcription

Import custom data

Easy to look up calls in speech

Easy to export data

Flexibility to link to other applications

Specific Call Monitoring

Topic Trending

Ability to target specific call types

Ability to link it with other applications more easily

Category Building

Category Comparison

Interaction Identification and Categorization

Reporting

Trending

Real Time Feedback

Put out fires

Identify Trends Quickly

Easier to create talk paths

Easier to train new users on

Verint Speech Analytics has been solidly reliable in enabling call transcription

Accuracy; currently our call accuracy is measured at 73% which creates some confusion when listening for select key words which we use to determine whether regulatory disclosures have been provided.

Evaluating agents at a rate manual audits could not achieve

Looking for specific phrases and providing comprehensive data on the use of those phrases

It has been difficult to exclude certain calls that should not be AMQ'ed, such as web-ex and internal calls

High topics analysis

Customer satisfaction

Agent analysis

First call resolution

Getting the customer voice

Knowing about the topics in call center

Finding calls quickly

Rating calls to match certain conditions

Ability to create detailed categories

Difficulty getting words updated in the dictionary

Only shows so many calls at once

Occasionally will pull calls that don't have relevancy to the category

PCI compliance

Trascription

AQM

Categories

In my opinion, integration within AQM could be better

Provide the data model so we can generate our own reports

Categorization

Identify opportunities

Transcription

Call intents

Customer Experience

Trending

Scorecards

Transcription

clearly shows dead air gaps in phone conversations

search for specific words / phrases is mostly good.

search / find / review seems easy to setup.

In my opinion, search is not always accurate - looking for words like "uh" or "um" often finds false positives.

predictive searching from Verint would be helpful

tagging areas of interest to surface recordings

recording reviews are excellent with the ability to listen and find problem areas

the drill down capabilities within the recording database is amazing

when tagging, it takes a bit of skill to learn "how" to get the system to retrieve the words that are being searched on.

In my opinion, the actual implementation of the DPA triggers for pause and resume is very tedious and cumbersome. we have yet to implement because it takes an IT person and an agent to work together to create

In my opinion, the amount of training can be difficult as it is difficult to get leaders into a class for as much time is needed for training

Trends

Searching for particular words in interactions.

Looking at calls as a whole with the interaction of the agent.

At this time, there are not enough medical terminology terms. Sometimes "Obamacare" comes across in speech as "bomb scare." Yikes!

Medical terminology is lacking.

The accents of people and those talking in a low voice that you can barely hear, Verint Speech Analytics doesn't pick up as well.

Locating calls by topic is quick and easy!

Building new categories and managing them is simple.

Trends and changes in volume of categories is easily detectable.

We are on an older version so the transcription isn't prefect, but it's good.