MM034 – Real-Time Analytics: is what we “really” need?
A lot of articles across the Internet tell us how much Real-Time is important to succeed in your business. But in the majority of cases, Real-Time Analytics is not the solution we need.
Nowadays Real-Time is a well-known term in analytical fields. Briefly, it is the possibility to analyze data at the same time they have been updated. For example, with dashboards that show instantly product trends in your store, you can react immediately and take strategic business decisions.
In this article, we’ll be trying to make some considerations about Real-Time to help IT’ s project managers understand if they really need it.
Before you start thinking about a Real-Time Analytics solution, you have to catch the real needs of your company and make a very detailed cost-benefit analysis. When you think about a project of this kind, it is mandatory to understand if our analytics scenario needs to be updated in real-time. This feature has to be a power-up for your company.
The questions we have to answer are:
Do my users really ask for faster updates for their analysis?
What are business strategies that need to be supported that way?
Is the effort required to develop this process worth it?
Almost every company has one or more Data Warehouse (DWH). A place where company data are stored, organized and historicized. Business Intelligence tools (BI) read data from DWH and the users get the updated information they need. In this kind of environment, data are often updated daily.
A Real-Time environment have immediate updates but also specific components. When you think about Real-Time probably you have to start from zero and could not re-use the architecture you have, so the start-up cost could be expensive.
Technologies used for Real-Time are very recent. You have to consider that people working in your company couldn’t be prepared enough and may need some training (this is another cost to consider).
A successful Real-Time project
On the other hand, there are topics where a Real-Time updating process is the solution.
This is the case of an assurance company that started to get information about their customers and their prospect from a multitude of sources.
Those are for example insurance policy information, customer agency details, IoT data (i.e. GPS tracking), …
Every source of information has a specific system that tracks data that are collected and then will be processed into the DataWarehouse.
The need is to let agents have all the customer (or prospect) information updated at the time they are selling a new policy. In this case, the loss of information could be crucial.
Let think about a policy renewal, if a customer just has a car crash (tracked by the GPS system) the agent needs to know in order to make the correct policy.
In this context, having Real-Time Analitycs for agent’s reporting systems is a real value for the company.
Real-Time it’s a very precious ally but it’s not a solution for every business problem. You have to take a step backward and understand well what are the needs.
Traditional DWH and BI tools are useful to analyze historical data. You can make analysis and compare different strategies, departments, financial data, order information, etc. with regards to revenue, costs, and other KPIs. Shortly, a DWH only allows for analyzing events that already happened.
With a Real-Time Analytics environment, you can manage operations and actions while things are happening. So the key difference is that Real-Time allows being proactive while events occur.
If you are sure that your users can react proactively to new data, you know what is your solution.