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.
Sections
- 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.
- 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
# 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
So let's talk about the four pillars of modern analytics
and how each pillar helps the organization better understand its data.
Now, the entire point of analytics is to take your raw data
and to convert that into an actionable piece of insight.
And between your data and action, we have what we call business analytics.
And as mentioned previously, there's four pillars to it.
And the first pillar is called "descriptive analytics".
And the focus here is on understanding
what had happened with your data or your organization.
So this is essentially a historical view into your business data.
And typically you represent these through
dashboards, reports, visualizations and so on.
An example of a descriptive analytics type of question could be,
"what was my churn last quarter?".
Now, once you understand what had happened,
the logical next step is to understand why something happened.
And this is called "diagnostic analytics".
And the focus here is on understanding why something may have happened.
So an example of a question here could be, "why did my churn spike last quarter?"
Now typically at this point you would
do some sort of driver analysis to understand the relationships between
various targets and various drivers,
and you would have specialized visualizations to represent that.
And of course, now moving on to the third step,
once you understand why something had happened,
you would want to understand what will happen in the future.
This is called "predictive analytics".
And the focus here again is on understanding
what will happen in the future with your data.
So at this point you would bring in AI machine learning algorithms,
look at your historical data, trends, seasonality patterns and so on
and predict where the data would be at some point in the future.
So essentially it's a forecast.
And an example again here would be, "what would my churn be next quarter?".
Now, at this point if the forecast is great, you may choose to do nothing.
But if the forecast is not trending in the right direction,
you may want to take some action.
And that's where "prescriptive analytics" comes in.
This is the fourth pillar of analytics.
And typically at this point you would run some decision optimization algorithms
and so on to come up with a recommendation or a piece of insight
that you could action to reverse a certain trend.
So, an example of a piece of prescriptive analytics would be,
offer a 20% discount to users,
showing high probability of churn at time of renewal.
So that covers everything from descriptive, diagnostic, predictive to prescriptive analytics.
However, you have this chasm now between analytics and action,
and to cross the chasm between analytics and action,
you have workflow automation capabilities,
or workflow automation tools that come in.
So it could take the input of a prescriptive piece of insight
and then automate a bunch of downstream actions.
And an example of one of these actions could be
updating your CRM tool with the 20% discount,
and sending an email out to your CSMs
letting them know that that action was taken.
Now, of course, organizations with high analytics and automation maturity
may automate some or all of these steps
for increased organizational efficiency.
If you have any questions or want to share your thoughts about this topic,
please leave a comment below.
If you liked this video and want to see more like it,
please like and subscribe!