IBM Watson Customer Experience Analytics Reviews

Our primary use case is to analyze the behavior of customers across digital channels - how people interact on websites or mobile applications, and what issues they may have. It also analyzes those behaviors on the business channels. For example, it can check how a customer is moving through the shopping experience on a retail site. Are they focusing only at the top or were they trying to zoom in on a product and were not successful? Were they trying to pinch out some text because it wasn't visible on the mobile application? All those things can be tracked and utilized to generate meaningful reports. This includes for marketers who can improve customer experience, for developers to improve the applications, and for security use cases to avoid fraud because it can detect if a site or application is being used fraudulently. We are partners of IBM and I'm an enterprise applications consultant.

There are three valuable areas: One of them is the granular DOM level where it captures behaviors on the application. Another is the analytics, because you don't need to provide it with a rule to identify the issues, it will analyze the usage of all the customers accessing that site and see if there are any abnormalities and it will identify unexpected changes in the pattern and that there is something wrong that needs to be checked out. That is extremely useful. Finally, the ability to create meaningful reports and dashboards to tell stories. The reports enable you to re-target customers if they've had any problems. 

An area that can be improved is on the technical side where you identify an event. Currently it requires more development skills during implementation. It's not very friendly for business users trying to set or configure the behaviors or events that are being tracked. It would be better if business users could easily configure events and start capturing them without needing to have a technical background. They wouldn't need to share use cases beforehand and to take on implementation partners to help roll out use cases. It opens up the door for further innovation by the whole company.

I've been using this solution for five years. 

It's a pretty stable solution. 

The solution is often used by small customers who then grow and scale it to their needs. Maintenance staff isn't required if it's hosted on cloud. If it's on-premise, it requires basic maintenance that any kind of data center administrator can do. 

The support is provided by Acoustics, because the product is by Watson. The IBM or Acoustics support for these products has been very good, they're available 24/7. They are always responsive communicating when there's any kind of down time. They're also available across multiple channels including phone and support portals. 

I previously used Crazy Egg, which covered a subset of the capabilities we have here but it wasn't built as an enterprise solution. It's suitable for a small business, whereas this solution allows you to actually be extremely detailed in the events that are captured on the dome level. Crazy Egg doesn't give you all the extra features of knowing what is actually happening on the screen, doing the full session replays and carrying out detailed reporting including dashboarding and integrating with third party tools. Watson has got a direct connectivity to many other applications in the ecosystem of the data. It's a strong feature compared to what is offered by other competitors like Crazy Egg. Watson is built for enterprise.

For the public cloud option the initial setup is straightforward, because it's only a matter of provisioning for deployment. On-premise is a complex setup depending on the servers, where you're running your websites or mobile applications. It requires setting up some agents that will intercept all the traffic so it can collect the behaviors and interactions happening on every page. This is where the complexity is and you need to accommodate for storage. The more our website is accessed the more scalable deployment infrastructure is required. An implementation of the on-premise option can take between one to three months.  Implementation can be carried out internally or by using the expertise of partners that can provide those services. There's no ongoing requirement in terms of maintenance. 

The cost is based on the number of traffic and events and it is tiered pricing. The higher the volume, the lower the price becomes per event. The number of users is irrelevant, it's all about how much traffic is being captured and analyzed. It's not a solution that is usually used or deployed across many employees in the company, only those who need to access it, and the prices are based on the interactions. It expands relative to the traffic.

I recommend this solution and would advise anyone implementing or adopting this product to have in mind multiple use cases, some that are related to maximizing revenue, some that are related to providing better customer experience, some that are related to avoiding fraud or capturing any security related usages on their site. Those are three main areas. I've learned that there is more than one way to use this platform. You can let your imagination think about a use case and if it's related to customer behavior on the district channels, it can most probably be done. It can also be expanded to offline channels using the right integration so for example if you want to be capturing customer journey through all of the interaction points that are online and offline, that is possible. 

I would rate this solution a 10 out of 10.