Operational intelligence. Do those two words ring a bell? If you answered no, this post is for you. Here, we’re going to define what operational intelligence is and explain what you should do to leverage its power to help your organization thrive. You’ll also learn why doing so is in your best interest.
What if your answer was “yes”? You should also keep reading. Even if you already have some understanding of operational intelligence, you’ll still learn some valuable insights in this post: insights that you can apply in your company and that will help you take your operational intelligence approach even further.
Operational intelligence: A definition
Let’s begin by trying to come up with a reasonable definition of operational intelligence.
We can start by seeing what Wikipedia offers us and work our way to a better definition from there. Here’s their definition:
Operational intelligence (OI) is a category of real-time dynamic, business analytics that delivers visibility and insight into data, streaming events and business operations. OI solutions run queries against streaming data feeds and event data to deliver analytic results as operational instructions. OI provides organizations the ability to make decisions and immediately act on these analytic insights, through manual or automated actions.
What do we get from the first sentence? Essentially, operational intelligence refers to a set of techniques and tools meant to extract knowledge from large amounts of data generated by business operations. We’ve also learned that this process is meant to be dynamic and real-time, which brings to mind automation. The third sentence attempts to dive a bit into the working of operational intelligence, explaining that queries are run against varied data sources in order to gather analytic results in the form of operational instructions.
And those two words—“operational instructions”—tie nicely with the final part of the definition, which I believe explains the core of what operational intelligence is about: decision-making.
So, here’s how I would define operational intelligence:
Operational intelligence refers to a set of technologies that can extract useful knowledge from a variety of data sources, in a dynamic and real-time fashion, to help organizations make decisions.
It’s definitely far from perfect, but I think it’s good enough for our purposes in this post. With the definition out of the way, it’s time for the next question: Why would you bother with it?
The case for operational intelligence
Now that you know what operational intelligence is, it’s time to decide whether it’s right for you. What about this practice would offer your company a competitive advantage?
There’s no such thing as a free lunch, and implementing a new practice in your organization isn’t going to be free, either. Be it in money, time, or—most likely—both, operational intelligence requires an investment. So, if you want to get your organization to adopt operational intelligence, you’re going to have to sell it to your coworkers and bosses. How would go about that? Keep reading to learn how to show why such an investment is worth it.
Logs: You probably have tons of them…
When developing software, your company—hopefully—uses some approach to ensure the application works as intended. But even with a very sound quality strategy, things sometimes go wrong. Once an app is deployed to the wild, what should you do? Cross your fingers and hope for the best? As they say, hope isn’t a strategy. And a strategy is what you need: a monitoring strategy.
Logging is, undoubtedly, a key component of a great monitoring approach. Which is why you probably have lots of data in the form of log entries. And that takes us to the next point.
…And they’re probably hiding a lot of value
Logs often have humble origins. They usually start as helpers for a postmortem debug, consisting of a time stamp plus an exception’s stack trace, but not a whole lot more.
As time goes on, however, our once-humble logs keep accruing more data. Be it for administrative or security purposes, or even regulatory demands, the fact is that log entries quickly become full of data that can be incredibly valuable.
The needles of insight in a haystack of data
From the previous two sections, we can draw two conclusions:
1. Since logging is a vital part of a modern application’s monitoring strategy, a typical application will have lots of log entries.
2. Those entries might contain a buried treasure in the form of valuable insights.
The key word in that last item is “buried.” What good is it to have valuable knowledge and not be able to access it? It’s even sadder than not having the knowledge in the first place.
And that’s why operational intelligence can have such an impact on your organization. It’s the missing piece of the puzzle that will let you make use of the information that’s already there, extract it, and transform it into knowledge that will help your organization make strategical decisions.
Enabling operational intelligence in your organization
If you’ve read this far, then you agree operational intelligence is worth it. Now, how do you go about actually implementing this strategy? That’s what we’ll talk about next.
Far from being a comprehensive tutorial, though, this section is going to be more of a brief, basic guide. We’ll cover some initial steps to get you up and running with operational intelligence. As you learn more, you can evolve this bootstrap approach into something more sophisticated.
Laying the groundwork
Let’s begin with an analogy. You know that not all code is equally testable, right? Before adding unit tests to a code base, you must first ensure that the code meets some criteria. It has to have low levels of coupling, make use of dependency injection, be modular…the list goes on.
Similarly, not all log collection is equally prepared for operational intelligence. Depending on how your logs currently look, you might be in line for quite a bit of prep.
First, you must ensure your logs follow well-established standards. For instance, here’s a common problem: You have logs from clients spread around many time zones. But the log entries feature time stamps expressed in each client’s local time. If you want to perform any kind of analysis based on time, you’re in for trouble. The recommended practice here would be to always express time stamps in UTC while logging and always using the ISO-8601 format. Alternatively, you could also log in each client’s local time, adding the offset from UTC.
It’s also important to use standard labels for the entries’ levels, such as INFO, WARNING, and so on. You want to be able to quickly distinguish between the different types of events being logged.
You could take all this one step further and get into structured logging, which consists of storing log entries in a structured language, such as XML, JSON, or similar. This allows for way easier and more powerful analytics operations to be performed on your logs.
Which tools can I use to gain operational intelligence?
Once your data is prepared, it’s time to start thinking about tooling. And here’s a tool that matches operational intelligence’s requirements perfectly: Amazon Elasticsearch Service.
Amazon Elasticsearch Service is a fully managed Elasticsearch service that you can use for search and log analytics. It centrally stores your data and lets you ask it questions, so you can access the insights that are buried deep within. And have I mentioned that it’s scalable and secure? Because it’s scalable and secure, it allows you to get your insights securely regardless of the volume of data. With Amazon Elasticsearch Service powering your operational intelligence approach, your organization will make decisions even more quickly, which can be the difference between beating your competitors to market or losing the competition entirely.
Harness the power of operational intelligence today
In this post, we’ve talked about operational intelligence: what the term means, why it’s worth the investment, and how to prep your organization before implementation. We also saw how Amazon Elasticsearch Service is the perfect tool for recovering the valuable insights lost under your mountain of log data.
Give Amazon Elasticsearch Service a try today. Your relationship with your log data will never be the same.