David Williams: There are a number of challenges, one of them the amount of data.
Today, there is more data than ever before, but also, there are many different types of data that organisations want to use as part of their decision making to ultimately perform better.
An example of this is the large amount of operational data that comes from various systems and records.
There is data that you can capture, called experiential data, where we are seeing more and more companies competing. They are using this data to factor in decision making and for improving performance.
There are challenges around how you access data and how do you apply meaning to all of the various types that you are capturing.
"There are many different types of data that organisations want to
use as part of their decision making to ultimately perform better."
Data quality is also an issue, particularly when it comes to trusting it for decision making.
There are organisations who are making information more pervasive and usable across the enterprise, in order to drive decisions.
You can look at this like a maths equation, where you have amount x, quality x, usage, equals a particular value. And if any of these are zero, then you end up with zero value out of your data.
All of these elements are very important when it comes down to data reporting and, ultimately, decision making.
David: We do see companies who try and build their own infrastructure, or at least have done so in the past.
What tends to happen, however, is that they hit a wall when challenges arise from attempting scalability of the platform.
The performance starts to bog down, security challenges start to present themselves, and data quality can be problematic, as often, there is no documentation.
Depending upon how it has been stove piped together, and who has done it, if that specific person or team leaves, the maintenance of this these systems can become a challenge.
There is also the issue of sensibility, which is the ability for it to connect with, or integrate with, other systems.
"We call this spreadsheet hell - where you are buried in spreadsheets, spending
countless hours manually aggregating them."
Lack of capabilities is another challenge, depending on how it has been constructed. It can be limited when it comes to scope of functionality.
Around this issue, we often see a heavy and pervasive use of stand-alone spreadsheets. We call this spreadsheet hell - where you are buried in spreadsheets, spending countless hours manually aggregating them.
There is usually no process around it, you have security issues, and it is common to see people emailing spreadsheets within and outside the company.
These are becoming tough challenges, particularly as an organisation starts to grow, which is what you hope to have happen.
"AI is about using its scalabilities to outperform the
competition and be more efficient"
You see, this now within the hottest area of technology, which is AI machinery. This is an area that is constantly on people’s minds.
It’s about using its scalabilities to outperform the competition and be more efficient. What people may not know is that a lot of these capabilities have been around for a very long time and are not all that new.
What is changing is the ability to make these capabilities more usable, and embedding them into an existing process that can be leveraged.
You want to be able to get maximum usage and utility out of data information within and outside of your organisation.
"The system can tell you whether there are salient points in your data,
automatically discovering the root cause of what is driving a KP."
It is about getting it more in the hands of information workers and business analysts, versus only being available for more technical staff or data scientists.
We wrap this up into what we call augmented analytics, which is about augmenting manual processes, enhancing
insights, with items, such as dynamically generated texts.
So rather than just looking at a chart, the system can tell you whether there are salient points in your data, automatically discovering the root cause of what is driving a KP, or being able to search for information like you
would in a search engine.
This is preferable to having to use some kind of lengthy statement or sequel script.
These are all capabilities that a modern analytics platform provides, as opposed to trying to do this on some legacy, cobbled together, stovepipe infrastructure.
David: There are certainly challenges when trying to integrate various systems, whether they are legacy or not. An organisation will have multiple systems of record or data repositories that they use to pull data from, in order to make decisions.
New software does help, whether it is in areas like providing live connections or pre-built connectors to certain systems which allow you to leverage meta and master data that is already developed, versus having to recreate it.
There are a number of different ways that you can get access to this different data, whether it resides in a repository or if it is sitting in a system of record.
This is certainly something that we are doing with analytics cloud, where we are providing live connections to other
systems, as well as bringing together data and analytics that allow for the ability to consume it through the analytic front end.
