As organizations look to streamline their automation tools, they need to look for tools that help them move from reactive problem resolution to tools that provide end-to-end visibility and can predict events, downstream effects, and SLA impacts before they occur.
There are very few instances in which more isn’t better – more money, more time, more chocolate – but when it comes to enterprise software, having more tools rarely means better outcomes. Consider the results of a recent survey of IT leaders that found that 80 percent of organizations use three or more automation platforms, and 74 percent use three or more observability tools. Theoretically, having several automation solutions targeting the same function – say, batch processing and other automated business processes – ought to make it easier to address problems. Not so. In fact, survey respondents report having too many tools delays remediation, decreases efficiencies, drives up costs, and impacts service level agreements (SLAs).
When you consider over a quarter of respondents said they use six or more automation tools, it begs the question, why? Often, teams invest in automation tools to support a particular platform, application, or environment. In some cases, it’s a matter of team preferences predicated on quick turnaround time and skill availability on the team. Other teams may inherit solutions due to mergers and acquisitions. Regardless of how teams end up with multiple tools, this tool sprawl is certainly not by design, and most teams said they want fewer tools.
Tool sprawl obscures automation issues and delays remediation
With good reason, it turns out. The survey revealed numerous repercussions of using multiple disparate tools, the most obvious being alert storms. These “alert storms” or “cascading alerts” occur when different systems report automation issues at the individual job level that may or may not have a business impact, generating numerous alerts. According to 74 percent of survey participants, all those alerts make it more difficult to resolve the situation because there is so much uncorrelated information that obscures the root cause of the problem. With too much information and not enough clarity, respondents say too many tools reduce the MTTR (mean time to restore).
Using many tools also creates data silos and redundant information. Participants also point to increased costs and inefficiencies as the teaching tool requires maintenance, paid support, and user expertise. As there are few individuals who are skilled enough to use all the different tools, organizations end up hiring for specific tools.
See also: The Synergy of Intelligent Automation and Real-Time AI: Redefining IT Workflows
Existing tools lack visibility and prioritization capabilities
Digging deeper, the survey revealed another major challenge is the lack of visibility. In fact, three-quarters of participants reported their organization doesn’t have end-to-end visibility for their automation jobs. And two-thirds indicated they don’t have end-to-end visibility for automated processes across all workload automation environments. Many indicated they only know that an automation job has issues when they receive an alert that the job failed, already causing delays and loss of revenue.
The research shows this constrained visibility leads to additional downstream issues that can impact SLAs. Consider that many alerts are tied to a specific process or automation job, such as an inventory update or completing a transaction. If there’s an issue with one of those jobs, what is the downstream impact? For example, is the uncompleted order suspended? Maybe there’s a shipping delay or a delay in creating a compliance report. Each of these failed actions would likely breach an SLA.
Added to that, the survey found only 28 percent possess tools that can predict automation issues, and 61 percent lack the ability to simulate potential workload additions or changes before deployment, putting the business at risk if the deployment fails. Further, just shy of 70 percent cannot model or predict whether a new or updated automation job will actually meet SLAs.
Despite numerous monitoring and observability tools, when an alert occurs, they often fail to specify what system is or will be affected. With no downstream visibility to determine if an issue will result in a breach of an SLA, and no way to predict or prevent such issues, most organizations are crippled by the multitude of inadequate automation solutions.
What should teams be looking for from their automation stack?
Issue prevention is always preferred to remediation. So, what capabilities should IT leaders look for in their automation business process tools to help prevent SLA breaches in the future?
- End-to-end visibility – Look for tools that provide end-to-end visibility across all environments and all automated business processes.
- Event/data correlation – Tools for managing and monitoring automated business processes, batch jobs, data streams, etc., need to provide correlated event analysis to understand real business impact.
- Predictive alerting – To enable the prevention of automation problems, look for tools that offer trending and threshold alerting capabilities.
- Workload simulation – A newer area for automation and IT teams is the ability to simulate workloads and their performance to ensure its operation will meet IT and business requirements as well as applicable SLAs before it is deployed.
Automation tools that offer these more advanced capabilities move the discussion from just IT systems and cryptic job names only understandable by select team members to which business processes and SLAs are being affected, providing critical information that business stakeholders can understand.
For automation success, less is more
The survey findings show an ad-hoc approach to monitoring and observability doesn’t work and ultimately subjects the business to unnecessary risk. As organizations look to streamline their automation tools, they need to look for tools that help them move from reactive problem resolution to tools that provide end-to-end visibility and can predict events, downstream effects, and SLA impacts before they occur. When it comes to automated business processes, fewer advanced tools are better for the business.
