“60% of the time, it works every time…”

It is great to be a consumer in the Digital Age! With a few clicks of a button, or even a simple voice command, our favorite goods may be delivered to our door within a few hours, or even minutes. Businesses have built their infrastructures to be more customer-centric and provide the best user experience possible.

But what happens when that technology fails? All the information communicated between the ordering systems, the transportation companies responsible for shipping, the banks exchanging money, and the warehouses packaging the items all have to be working 100% of the time in order to provide the best customer experience.

Enter the world of IT operations. Thousands of failed integrations, batch jobs, and Electronic Data Interchange (EDI) connections fail on a daily basis (think EDI for logistics/supply chain [shipments managed by UPS, FedEx, etc.] and batch jobs for Banks/Financial institutions). Credit card swipes often show “pending” because they sit in a queue for a few hours, or even days, before banks can post the payment; this is an example of a batch job (although there are many other types of batch jobs run by organizations in many business segments, this is an easy one to relate to). Finance really holds the “mother lode” for number of transactions: imagine every time a credit card is swiped or a high-frequency trading terminal executes a stock transaction.

Organizations with great IT Operations are typically able to resolve the failures before a customer notices any degradation in service. Logically, people would assume that transactions running at a 98% success rating would be great. Let us do some quick math:

  • A major online retailer ships roughly 600+ million packages a year, which comes to about 1.6 million packages a day. Within those orders, anywhere from 1 to 10 items may be included. Each item is a separate transaction, coming from various vendors who need to be paid and from whom inventory needs to be decremented and/or reordered. One person buying 3 different items online may result in 10+ transactions/automated integrations.
  • Thus, at the risk of greatly understating the volume, roughly 16 million transactions take place daily (including online retailer’s vendors, payments to banks, emails for confirmation, warehouses receiving orders, shippers planning for pickup/delivery, etc.) between the retailer, the retailer’s partners, and the consumer. With a 98% success rate, it equates to 320,000 failures on a daily basis, somewhere in the process. That is over 100+ million failures per year spanning across the retailers, vendors, shippers, and banks.

Today, professionals in IT Operations are resolving these failures by scripting/coding to put automation in place. But those automations often only work within a monolithic and extremely predictive environment. If one thing changes in the environment it typically results in the automation failing within the system. Once this happens, an IT professional has to begin the troubleshooting process:

  1. Be aware that the failure occurred
  2. Understand the severity to prioritize how quickly it is addressed
  3. Identify the root cause
  4. Attempt to resolve using various strategies

As many can imagine, it may take anywhere from 5 to 30 minutes to complete these steps, which increases a business’ internal costs and bogs down IT resources.

In comes the machine-first delivery model (MFDM) and AIOps. A machine could complete that process in seconds, providing a more streamlined user experience and allowing IT to focus on the strategic areas of the business rather than trying to keep the systems performing at a high level. Robotic Process Automation (RPA) is great for a predictive environment, but is not necessarily designed for an ever-changing environment, and therefore lacks the agility to handle the above scenario with a high rate of success. The automation needs to be cognitive and intelligent. Digitate is providing these Intelligent Automations through ignio, its AIOps platform, and allowing our partners to be more intelligent, Agile, automated, and on the cloud.

by Joseph Alvarez

Leave a Reply