AIOps is one of the most important IT milestones in the next decade; however, due to high cost and limited functionality, AIOps software is still an immature market. 

AIOps, or artificial intelligence for IT operations, is a crossover software between AI engines and IT management solutions. It uses machine learning to analyze data collected from different monitoring solutions and databases. AIOps platforms can correlate data sets and proactively identify anomalies within them. This greatly reduces the IT operations time by eliminating the need to manually look for problems. Many IT and development teams are using AIOps today to increase productivity and maximize user satisfaction. 

AIOps may not be the right solution for your business

With all the benefits in mind, many companies still aren’t willing to buy this software due to a variety of reasons. In the G2 review data below, the top reasons why many buyers dislike the AIOps software are cost, functionality, UI, reporting, and integration.

I will discuss each of them so you can consider these factors before buying the software.

Expensive and a high ROI period

The most mentioned reason to dislike AIOps is the high cost of implementation. An article on eWeek shows that an AIOps can cost around $10,000 per system. For an enterprise that has thousands of systems, this is surprisingly expensive to many IT buyers. According to G2 review data, nearly a third of the buyers mentioned price is an issue. It is also not about how big the price tag is, but also the pricing model. Here is a review that sums up other similar reviews.

An administrator in logistics and supply chain reviewed: 

“The pricing model is confusing and hard to compare to comparable offerings apples-to-apples. That can make cost justification more difficult than it needs to be.”

A high price results in a high ROI period. According to the G2 AIOps category Grid® Report, the average ROI months for AIOps software is 16 months, while the average contract period is also 16 months. This means that if a company buys an AIOps software, they can expect to get their value only by the end of the contract period. While this doesn’t seem like a good deal, AIOps software is getting more commoditized with increasing competition. The licensing price is expected to decrease in the future. 

Limited functionality for its price

The second most mentioned issue with the software is that users want more functionality for the price. Many functionalities need to be expanded for different use cases. The most common use case is IT monitoring and anomalies detection. This is not surprising because most AIOps software grows out of application performance monitoring (APM) tools. However, when DevOps starts to use AIOps for production purposes, it requires more functionalities on log ingestion. 

In this case, I recommend software buyers list out specific criteria for their use cases before vendor selection. By selecting the vendors that fit the criteria, the buyer will be more likely to get what they need right out of the box. This will also reduce the need to customize additional functionality in the future. 

Needs more report customization and better UI

Since AIOps has different use cases, different personas need different reports. One trend that is catching up is AIOps reporting gradually becoming more flexible like business intelligence reports. However, there also should be more report templates so users do not need to build everything from scratch. 

The UI dashboards should also be configurable to the users’ needs. Complex AIOps products often come with a complex dashboard that is difficult to navigate. Some users can be disincentivized by the complex dashboard due to a lack of knowledge and having more training material will be more helpful. Below are some examples of the mentions. 

Steve S, a mid-market user, reviewed: 

“The initial configuration was rather complex, though this was mostly due to the high granularity and relatively deep integration supported.”

A user in enterprise retail reviewed: 

“Scheduling is quite complex for beginners, once we get to use to it, it will be easy work. To overcome this issue, if you have some video guide, that would be a real help.”

Reliance on legacy and cloud integration

Integration is a big topic in every IT system. With the rise of cloud adoption and increasing usage of legacy software, AIOps platforms need to integrate with both to survive today’s complex IT environment. While legacy software integration has always been an issue, cloud integration should be the buyer’s new consideration. Cloud-based AIOps software require monitoring microservices and Kubernetes. For AIOps software that isn’t cloud native, it becomes a problem. Here is a review below that showcases why non-cloud-native AIOps software might create a problem.

An administrator in IT and services reviewed:

“There is no easy way to get logs from containers and k8s.”

I strongly recommend buyers prioritize cloud-native AIOps solutions as the world is moving toward clouds. Non-cloud solutions can still be considered for compliance or security issues, but that depends on your business and industry (such as government, finance, or healthcare).

AIOps will mature in the next decade

While businesses have started to look at implementing AIOps for improving IT service management, it is not the most ideal solution given the high cost, limited functionality, and difficulty in deployment. 

However, with the growing demand for autonomous IT operations, AIOps software will eventually catch up to the market demand and become a must-have software for almost every organization within the next decade.