Step 1: Use the Pareto Chart to Identify Priorities
Collect Data: Gather information about the problems you want to solve. For example, if you want to filter customer complaints, categorize them by type (late delivery, defective product, poor service, etc.) and count how many times each type occurred.
Create a Pareto Chart: Order the types of complaints from most to least frequent. Calculate the cumulative percentage and plot the data.
Identify Vital Causes: The cumulative line on your Pareto chart will show which types of complaints account for the top 80% of the total. These are the "vital few" that you should prioritize.
Practical Example:
If your Pareto chart showed that 80% of complaints are about "late delivery" and "defective product," your filtering system should have options to focus on these two categories.
Step 2: Use the Ishikawa Diagram to Deepen the Analysis
Choose a Main Problem: Take the vital cause identified in Pareto (e.g., "delayed delivery").
Draw the Fishbone: Place the main problem at the head of the fish.
Brainstorming the Causes: For each fishbone (Labor, Machine, Material, etc.), list the possible subcauses.
Labor: "Lack of training," "insufficient staff."
Method: "Inefficient logistics process," "lack of tracking."
Machine: "Slow management system," "defective vehicles."
Identify the Root Causes: Analyze the subcauses and find the true origins of the problem.
Building the Filter System Based on the Findings
After using Pareto and Ishikawa, you will have a clear list of which problems and their root causes are most important. Now you can create a targeted and effective filter system.
Essential Filter System Features:
High-Level Filters (Pareto-Based): Create filter options for the root causes that generate the most problems. If Pareto indicated "delayed delivery," "defective product," and "poor service" as the main causes, your filters should have buttons or drop-down menus with these options.
Middle-Level Filters (Ishikawa-Based): Allow the user to drill down using the categories and subcauses of the Ishikawa diagram.
Example: If the user selects the "delayed delivery" filter, you can offer subfilters such as "Logistics Issues" (Method), "Tracking System Failure" (Machine), or "Shortage of Personnel" (Labor).
Detail Filters: Add filters for more granular data that help validate root causes.
Example:
By Period: "Last 7 days", "Current month".
By Location: "Southern Region", "State of São Paulo".
By Agent/Employee: "Complaints handled by John", "Orders processed by Mary".
Example Filter System Structure
Imagine you are creating a dashboard to analyze customer satisfaction.
Main Problem Filters (Pareto):
[ ] Shipping Problems
[ ] Product Problems
[ ] Service Problems
Cause Category Filters (Ishikawa):
When selecting "Shipping Problems," new filters appear:
Root Cause: [ ] Logistics, [ ] Team, [ ] System
Cause Detail:
If "Logistics" is selected: [ ] Delay, [ ] Wrong Address
If "System" is selected: [ ] Tracking Failure, [ ] Address Error
This way, your filter system is not a mere collection of options, but an analytical tool that reflects the reality of your problems and allows for targeted and effective action. It uses the intelligence of Pareto and Ishikawa analyses to get straight to the point.
But the 80/20 principle applies to lottery numbers, or the minimum vs. maximum configuration of each filter.
You make an excellent observation, as the application of the 80/20 principle goes beyond just numbers and can indeed be used to optimize filter configurations.
The 80/20 Principle and Filter Configuration
The 80/20 principle is not restricted to lotteries or any other type of data. It is a heuristic, a rule of thumb, that applies to a wide range of phenomena. It states that, in most cases, approximately 80% of the effects come from 20% of the causes.
Here's how you can apply this concept to your filter system:
1. To Prioritize the Most Used Filters
Identify the "vital few causes": Analyze which filters and filter combinations users apply most. You'll likely find that 20% of the filter options are used 80% of the time.
Organize the interface: Place the most frequently used filters prominently at the top of the interface or in an easily accessible side panel. Less frequently used filters can be moved to an "Advanced" or "More Options" menu.
Example:
If your e-commerce filter system shows that 80% of searches are for "size," "color," and "price," place these three filters prominently. The "material" or "brand" filter can be placed in a secondary section.