How a simple AIOps platform increased labor productivity of IT in a retail company by 250%
In the modern world, it is impossible to imagine retail without IT. Intense competition forces the largest retail chains to invest in technology — they can provide advantages that others do not have. The rapid development of technology requires business to be constantly on alert so they can quickly adapt to these changes. This ceaseless IT growth is often accompanied by an increase in the problems associated with it, which companies often cannot solve on their own. The fragmentation of systems, lack of a single control center, large amount of information noise, decrease in the speed of processing incidents — and now IT is no longer just not solving the tasks, they are creating new ones. Given these conditions, how can a company stay competitive, not lose money and employees and ensure IT specialists can be the superstars they are? And how does AIOps factor in? I’ll tell you with a real example of the implementation of my brainchild — the Acure platform.
My name is Nick and I am the CEO of Acure.io — AIOps data monitoring and observability tool with powerful low-code automation. And here is the story of one of our clients.
True story
One day, I was approached by a large retail company with a request to improve IT productivity. Despite the widespread automation of processes and modern software, IT was not producing the desired results. Due to the scale and large number of branches, the IT structure of the organization was fragmented and multi-layered, including networks, data centers, cloud services, application support, automation of stores, warehouses, and more. In each of the branches the technical component was developed differently, in stores they generally used networks provided by third parties. All these factors made it difficult to control the already complex multi-level IT.
For a long time, due to fragmentation, the main IT resources were spent on support, whether it was a logistics system, online cash register or video surveillance. At the same time, the IT complex could be divided into two components: CHANGE (creation of IT resources, implementation of new and modern support of systems, such as financial or logistics) and RUN or operations (maintenance of hardware, servers, legacy and others). OPS, in turn, had its own NOC, which monitored the status of the entire IT complex, including IT equipment.
From a technical point of view, the IT complex of the company could be briefly described as follows: many disparate IT systems controlled by different monitoring platforms and managed by different working groups but everything is serviced by one situational center (and it was true for the bosses unfortunately).
Now, my article will not surprise you with the unpredictability of the plot, because I assume you’ve already guessed the problems my client had.
- Firstly, the presence of more than 30 disparate monitoring systems (Zabbix, Oracle, SAP, Instana, Elasticsearch for analyzing logs, etc.) produced a huge amount of information. It was often duplicated or sometimes unavailable because product teams responsible for some systems didn’t want to share this. Only the investigation of each incident took more than half an hour — precious time that could have been spent on its elimination.
- Secondly, more than a hundred alerts per day and a large amount of manual routine work led to burnout of the employees responsible for monitoring, and as a result, a high staff turnover. The employees did not stay for more than a year. And again, precious time was wasted but now for the search and training of new employees.
Then the business sounded the alarm and asked for help. And we helped.
A cure for IT
Acure is a kind of add-on to existing systems that allows you to collect data in a single window and display the status of the entire IT complex on one screen. Even if it is complex like in our case. At the same time, information about new configuration items is added automatically by the autodiscovering function, and then an algorithm of certain actions is also configured using automation mechanisms. But first things first.
- Having deployed the enterprise version to the client, we received event data from all local monitoring systems, using auto-discovery and dependency mapping, built the connections and got the information from more than 30,000 devices. Based on this data, a single resource-service model was built, showing the health of all components of the entire complex in real time.
- the problem of fragmentation and duplication of data
+ transparency and accessibility of all IT components
- Links between components and business services reduced the time to find the root cause of the problem and the type of failure by 60%. So, this freed up additional time for specialists to solve the incident instead of spending it searching for the source of the problem. Alerts of the same type and corresponding tasks were classified, creating a single knowledge base for solving problems.
- time to find the source of the problem
+ time to fix it
- For similar tasks of the same type, Acure allows the setup of automatic actions — from notifications to scripts. In our case, more than 30% of these tasks were automated. Without recruiting new employees and keeping the same staff, this significantly reduced the manual processing of tasks and increased labor productivity by 250%.
- 30% manual work
x 2.5 labor productivity
Conclusion
In this article, I shared one of the cases from Acure’s history, which clearly shows how a simple AIOps platform can successfully handle complex business challenges. Over time, I will share more stories, but in the meantime, I look forward to your feedback and invite you to visit Acure website, where you can learn more about the platform and all the benefits mentioned.
