Learning Library

← Back to Library

Four Pillars of Modern Analytics

Key Points

  • The four pillars of modern analytics—descriptive, diagnostic, predictive, and prescriptive—progressively transform raw data into actionable insights, moving from “what happened” to “what to do.”
  • Descriptive analytics provides historical views through dashboards and reports, answering questions like “What was my churn last quarter?”
  • Diagnostic analytics investigates why events occurred, using driver analysis and specialized visualizations to answer “Why did my churn spike last quarter?”
  • Predictive analytics employs AI and machine learning to forecast future outcomes, such as “What will my churn be next quarter?” based on trends and seasonality.
  • Prescriptive analytics recommends specific actions (e.g., offering a 20% discount to at‑risk users) and, together with workflow automation tools, bridges the gap between insight and implementation.

Full Transcript

# Four Pillars of Modern Analytics **Source:** [https://www.youtube.com/watch?v=m9emhDSDKcs](https://www.youtube.com/watch?v=m9emhDSDKcs) **Duration:** 00:04:27 ## Summary - The four pillars of modern analytics—descriptive, diagnostic, predictive, and prescriptive—progressively transform raw data into actionable insights, moving from “what happened” to “what to do.” - Descriptive analytics provides historical views through dashboards and reports, answering questions like “What was my churn last quarter?” - Diagnostic analytics investigates why events occurred, using driver analysis and specialized visualizations to answer “Why did my churn spike last quarter?” - Predictive analytics employs AI and machine learning to forecast future outcomes, such as “What will my churn be next quarter?” based on trends and seasonality. - Prescriptive analytics recommends specific actions (e.g., offering a 20% discount to at‑risk users) and, together with workflow automation tools, bridges the gap between insight and implementation. ## Sections - [00:00:00](https://www.youtube.com/watch?v=m9emhDSDKcs&t=0s) **Four Pillars of Modern Analytics** - The speaker outlines how descriptive, diagnostic, predictive (and the implied fourth) analytics sequentially transform raw data into actionable business insight. - [00:03:08](https://www.youtube.com/watch?v=m9emhDSDKcs&t=188s) **Bridging Analytics to Action** - The speaker explains how prescriptive analytics—such as issuing a targeted 20 % churn‑prevention discount—can be connected to workflow automation tools that automatically update CRM records and alert staff, illustrating the gap and maturity needed to turn insights into concrete actions. ## Full Transcript
0:00So let's talk about the four pillars of modern analytics 0:03and how each pillar helps the organization better understand its data. 0:07Now, the entire point of analytics is to take your raw data 0:10and to convert that into an actionable piece of insight. 0:14And between your data and action, we have what we call business analytics. 0:27And as mentioned previously, there's four pillars to it. 0:31And the first pillar is called "descriptive analytics". 0:41And the focus here is on understanding 0:48what had happened with your data or your organization. 0:52So this is essentially a historical view into your business data. 0:56And typically you represent these through 0:58dashboards, reports, visualizations and so on. 1:02An example of a descriptive analytics type of question could be, 1:05"what was my churn last quarter?". 1:09Now, once you understand what had happened, 1:12the logical next step is to understand why something happened. 1:16And this is called "diagnostic analytics". 1:24And the focus here is on understanding why something may have happened. 1:31So an example of a question here could be, "why did my churn spike last quarter?" 1:36Now typically at this point you would 1:38do some sort of driver analysis to understand the relationships between 1:41various targets and various drivers, 1:44and you would have specialized visualizations to represent that. 1:48And of course, now moving on to the third step, 1:50once you understand why something had happened, 1:54you would want to understand what will happen in the future. 1:58This is called "predictive analytics". 2:04And the focus here again is on understanding 2:09what will happen in the future with your data. 2:18So at this point you would bring in AI machine learning algorithms, 2:21look at your historical data, trends, seasonality patterns and so on 2:25and predict where the data would be at some point in the future. 2:28So essentially it's a forecast. 2:30And an example again here would be, "what would my churn be next quarter?". 2:35Now, at this point if the forecast is great, you may choose to do nothing. 2:41But if the forecast is not trending in the right direction, 2:44you may want to take some action. 2:46And that's where "prescriptive analytics" comes in. 2:53This is the fourth pillar of analytics. 2:56And typically at this point you would run some decision optimization algorithms 3:00and so on to come up with a recommendation or a piece of insight 3:05that you could action to reverse a certain trend. 3:09So, an example of a piece of prescriptive analytics would be, 3:14offer a 20% discount to users, 3:19showing high probability of churn at time of renewal. 3:24So that covers everything from descriptive, diagnostic, predictive to prescriptive analytics. 3:30However, you have this chasm now between analytics and action, 3:35and to cross the chasm between analytics and action, 3:38you have workflow automation capabilities, 3:40or workflow automation tools that come in. 3:42So it could take the input of a prescriptive piece of insight 3:46and then automate a bunch of downstream actions. 3:50And an example of one of these actions could be 3:56updating your CRM tool with the 20% discount, 4:02and sending an email out to your CSMs 4:06letting them know that that action was taken. 4:09Now, of course, organizations with high analytics and automation maturity 4:13may automate some or all of these steps 4:16for increased organizational efficiency. 4:19If you have any questions or want to share your thoughts about this topic, 4:22please leave a comment below. 4:24If you liked this video and want to see more like it, 4:26please like and subscribe!