Espressive Barista Reviews

Our primary use case is for an internal help desk.

We implemented this solution because we were trying to drive some scale for the organization. It contributed to the strategy of trying to automate more of the business.

We saw a pretty healthy climb in deflection rates when we started using this product. This is really important. We were well over 30% to 40% deflection right out of the gate. Over time, it starts to level out a little bit, but we're continuing to see at least 30% deflection rates, meaning an engineer doesn't have to get involved and people can self serve and find the answer to their problems without having to engage with an agent. We're looking for opportunities to continue to drive that forward, but I think it's pretty effective if you can get that level of deflection.

On average, we see 30% of support requests resolved without human support personnel involved.

The results we were seeing started very quickly, within a few months. Getting ramped up on the out-of-the-box content is pretty quick.

Given the out of the box content for common industry applications, we got good value out of the solution right from day one. It's hard to put an exact dollar amount or metric on that. One of the main points is the commonly asked questions and common tech stack. A lot of that was pre-built into the solution, which means that we didn't have to create our own content. This is something that you really want to be able to shy away from.

For example, it's helpful that the Microsoft stack, Okta, and other pretty well-known and widely used technology stacks already have content in the system. I found that the content in their broader database was pretty robust, which helped us ramp up. You don't have to spend a lot of time creating content prior to the launch, which makes it a fairly straightforward launch. More than anything, a lot of getting set up is educating your end-users.

Over time, managing it operationally has gotten easier. This may be a reflection of its ability as it's collecting phrases and building out the broader database. It's allowed us to spend less time having to add information about what scenario somebody needs when they say something. Over time, I've seen the operational overhead continue to decline, pretty drastically. At this point, we don't have to spend a tremendous amount of time on the upkeep of the platform, which is good.

When it comes to how well it performs, it depends somewhat on how good your data is. For example, you have to make sure that your in-house customizations are going to continue to work with the standard integration, such as with Active Directory.

I'm not sure exactly how much this solution has reduced our mean time to resolution, although it's probably at least 20%. Tickers are getting answered by a virtual support agent in real-time with a caller, versus a ticket being created, sitting in a queue, someone picking it up, then calling them and resolving it. It's a much faster turnaround. 

Using this solution has helped increase the availability of support for our employees to 24/7. We don't have agents all over the world and Expressive is on all of the time.

Our employee satisfaction scores have gone up a little bit. We had high sat scores before, but it was because we had people killing themselves answering all of these tickets. Now with our rate of deflection, they're able to manage the workload in a more scalable way.

The satisfaction of the employees has definitely not decreased, which is good because it means that people are still happy with the level of support. We already had a high customer sat score. The good news is that we were hoping it wouldn't make the sat score go down, because now they're not talking to a live agent, which many people, in some cases, would prefer.

This product has a couple of really good features.

Its real strength is in natural language processing. When you ask it a question, it has a pretty good sense of how to answer it. This is one of the things that they have done the best job on.

This was a pretty important feature for us. It means that you could be discussing the subject with a virtual support agent and it would be able to, as you ask the questions, interpret the types of questions you were asking and understand what you were trying to ask. This is a difficult task because not all people ask questions in the same way. Then, it is able to bring back content that is fairly accurate in addressing the question. That was definitely a contributing factor in our decision to implement it. Without that capability, the solution was not going to provide a tremendous amount of value.

Out of the box, the system is very good at being able to recognize phrases that people commonly use in support questions. I think that they've built some intelligence into that, and it includes a pretty broad database of standard technology stack questions and answers. 

It is very good to see the direction that they are taking this product in. For instance, the work that they're doing regarding automation. The platform is able to interpret a question and respond with a knowledge article that contains information on how to fix the problem. Examples of this are resetting a password and unlocking an account. The ability to automate that process is where it shows continued value.

Our developers have used the Barista Control Center to extend the platform by adding content, and they've found it fairly easy to do and manage. There have been instances where there wasn't standard content available, so they've had to extend the platform so that when people ask a question about an application or system, it's mapped directly to the content.

The solution has provided a solid set of automation capabilities but would like to see the continued expansion of even more advanced automation capabilities. 

Having more prebuilt connectors for people is always preferred because it's something you can start from versus having to go and build something custom. They are starting to do that with some of the recent connectors that I've seen for Workday and some of the Microsoft stack. They're moving aggressively in that direction.

I have been working with Espressive for approximately a year and a half.

There have been hiccups here and there, but the biggest and best technologies out there still have hiccups every once in a while. Overall, it's been a stable platform.

We're pressure testing scalability now, as far as trying to build that automation and how we extend the product. So far, it's proven to be pretty scalable. We haven't had any issues with performance because of the existing load on the system, so I've not seen any scale issues.

Espressive was involved in the employee onboarding process. Their customer success team helped us the whole way along. They have been good.

When we started with them, they were still building out their customer success program and there were some growing pains there. However, I wouldn't say that anything was problematic. I think they were always very receptive to feedback and made sure that we were successful. That was the overriding goal the entire time with them.

Their attitude was "Hey, we want to get you up and running to be successful, so we'll do whatever it's going to take to do that."

We haven't had to deploy the solution in a year and a half but talking to other customers, since then, just trading stories, we've learned that the onboarding process has gotten much more streamlined.

The technical support has been good. When there have been issues, our team has been able to engage with the support team. I haven't heard of any major issues with not getting the support we needed.

I would rate the technical support and eight out of ten because they're still maturing as a company.

Positive

We did not use a similar solution prior to this one.

We went looking for this type of solution because we wanted to know how to better manage investments in a help desk and be able to drive the strategy on automation and scale.

Espressive is set up as a SaaS subscription service.

The initial setup was fairly straightforward. More of the process is on the change management within your own organization and making sure that people understand how it works. They need to learn how to engage with it because you can have them go to the solution via Slack.

There are lots of different entry points into Barista and making sure that you clearly map out what's going to work best for your organization seems to be more of the challenge. The tech is there and ready. A lot of it was preparation inside the business in terms of adopting this new way of supporting people and teaching them how they were going to access it.

The change management and the communication was probably the longest part of the process. The deployment was a few months, maybe as long as a quarter. We were able to get people on board and adopt it pretty quickly out of the gate because we had done so much work on comms and training and a lot of announcements around it. Our employee base is pretty tech-savvy, so they were happy to engage with the technology. It wasn't a huge effort to get people to engage with it. It was more around just making sure people were aware of it and how to use it.

Barista has helped us in terms of cost savings, from a cost avoidance perspective. Instead of having to hire more service desk agents as the company has grown and expanded, we've not had to make those investments.

It is difficult to estimate how much money we have saved but it would be at least a couple of staff, which is equivalent to a few hundred thousand dollars per year.

When I look at technology and purchase it, I consider what the return on that investment might be. This solution has yielded a good return on that investment.

The price for the licensing is fair. We were an earlier customer, and that's always advantageous. I don't know what their pricing looks like today and I can't remember exactly what we paid but I thought it was a very fair price.

This wasn't going to be done by adding more people. It needed to be done with technology, and so going out into the market, we considered a few vendors that play in this space. In our search, we found Espressive to be the most complete platform for all of the different areas that we wanted to be able to address.

ServiceNow has a solution but we're not a ServiceNow shop. There's a company called Moveworks that has a solution that's more geared just to IT, versus Espressive, which allows you to work not only in IT, but HR, legal, and other areas. It allows you a lot more of an enterprise-wide ability to operate this platform, which is a real plus.

There were some other smaller players, like Zendesk, which has a little something there. Many vendors are dabbling in it, but with Espressive, that's their core business. That's what they do. Some of the things I talked about as far as expansion beyond just IT were big pluses and the technology stack as a whole was just better.

We run the business in English, so we haven't had a requirement to have it serve content in different languages.

Barista has just released some capabilities where you can start building out some of your own custom integrations around driving automation. We are now starting to work on that. We have a handful of automations around account lock and password resets, and things like that, but we're now starting to look at tasks related to procurement.

There aren't very many things that this product needs improvement on. Automation has been the one that I've been the most bullish on, and I've pushed the company and the customer success team that works with us on this. I ask them how we can continue to make automation a bigger part of our platform because that's where the long-term value proposition's going to be.

Having a database of questions, and answers, and being able to provide people with the content is great, but I think we're moving beyond that and into this space where people are going to just want the problem to be solved right then and there.

They don't want to have to go reading an article and then figure it out. They would prefer just to have something be done for them. I think we're to that place now where you should be able to do that.

The top 10 most commonly asked questions in every organization have many questions in common. With being able to automate those top 10, all of a sudden you get deflection rates that are 50% to 60%. That's where that market and where the tech stack is going, and I think Espressive is moving in that direction.

I would rate this solution an eight out of ten.