Every portfolio has concentration. A few states, a few segments or a few products account for most of the customers or the value. The Pareto chart makes that concentration visible in a way a table does not, and helps you decide where to start.
How the Pareto works
The chart has two parts. The bars show each category, sorted from largest to smallest. The line shows the cumulative total: the first category, then the sum of the first two, and so on up to the total.
The reading lies in the point where the line starts to flatten. The categories before it hold most of the total. The ones after it form the tail: many categories with little volume each.
The idea comes from the observation that a few causes usually account for most of the effect. It works as a starting point for investigation, and every portfolio has its own curve.
Choose the cut by the decision
The Pareto answers the question the cut asks. By state, it shows where your data is. By segment, it shows which profiles weigh most. By days overdue, in a collections portfolio, it shows where the volume is.
Before building the chart, define the decision it will support. If the decision is which region to start a campaign in, the cut is geographic. If it is which group to contact first in the collections sequence, the cut is the range or the segment.
Count or sum
The same portfolio produces two different Paretos. Counting records, you see where the people are. Summing a value, such as outstanding balance or average ticket, you see where the money is. A state can have many customers with low values and show up at the top of one chart and in the middle of the other.
For effort decisions, such as message volume, count is usually the right cut. For return decisions, the sum of the value.
Build the Pareto without a spreadsheet
In a spreadsheet, a Pareto requires grouping, sorting, calculating the cumulative total and building a combined chart. In DeepMiner, you ask the AI assistant: “make a Pareto of the concentration of records by state in the Collections Portfolio index”. The answer brings the chart with the bars and the cumulative line, plus a text with the highlights.
On the dashboard, the Pareto widget accepts metrics such as count, sum and average, with grouping by value or by range. With global filters, the same Pareto can be viewed for just one segment or just the records imported in the last month.
From concentration to action
The Pareto does not act on its own. The cut at the top needs to become an actionable list. In DeepMiner, the top categories go in as the filter of a flow strategy, with the is one of operator, and the strategy shows the expected records. The tail can get another strategy, with another channel or another recurrence.
After the send, the result of each message returns to the record and shows up in the index’s record list.
Want to see a Pareto of your portfolio? Book a demo.
Frequently asked questions
What is a Pareto chart?
It is a bar chart sorted from largest to smallest, with a line that adds up the cumulative total. It shows how many categories hold most of the total.
Are Pareto and the ABC curve the same thing?
They are close ideas. The ABC curve uses the Pareto cumulative to split items into classes, such as A, B and C, according to their share of the total.
How do I make a Pareto in DeepMiner?
Ask the AI assistant, for example: make a Pareto of the portfolio by state. The chart comes ready and can go to a dashboard.