Flow, process and networks

Conversion-rate trend chart

Shows how the proportion reaching an outcome changes over time.

  • Chart
  • Conversion
  • Trend over time

What it answers

Is conversion improving over time?

Shape
Rate lines across periods
What it shows
Conversion changes and volatility.

Example

The question

Is the share of visitors who buy rising or falling?

Conversion rates by month, 2025 (%)

Conversion rates by month, 2025 (%)Visit-to-order conversion rises from 3.8% in January to 5.6% in December. Jan 3.8%; Feb 3.9%; Mar 4.1%; Apr 4%; May 4.4%; Jun 4.6%; Jul 4.3%; Aug 4.5%; Sep 4.8%; Oct 5%; Nov 5.3%; Dec 5.6%.Visit to order (%)0%2%4%6%8%JanFebMarAprMayJunJulAugSepOctNovDec3.8%5.6%
Conversion rates by month, 2025 (%)Visit-to-order conversion rises from 3.8% in January to 5.6% in December. Jan 3.8%; Feb 3.9%; Mar 4.1%; Apr 4%; May 4.4%; Jun 4.6%; Jul 4.3%; Aug 4.5%; Sep 4.8%; Oct 5%; Nov 5.3%; Dec 5.6%.Visit to order (%)0%2%4%6%8%JanMarMayJulSepNov3.8%5.6%
Visit-to-order conversion rises from 3.8% in January to 5.6% in December. Jan 3.8%; Feb 3.9%; Mar 4.1%; Apr 4%; May 4.4%; Jun 4.6%; Jul 4.3%; Aug 4.5%; Sep 4.8%; Oct 5%; Nov 5.3%; Dec 5.6%.

Illustrative data. It belongs to no company.

When to use it

Use it when

Comparing consistent conversion definitions across periods.

Avoid it when

Changing eligibility, attribution or tiny denominators.

How to read it

Read the rate trend and check the counts behind each point.

Note

Use consistent converted and eligible populations.

The data you need

Layout
Long table
Fields
Period: date/text; EligibleCount: integer; ConvertedCount: integer
One row is
One period and the counts defining its conversion rate. What is the grain?
Rules
Require 0 <= ConvertedCount <= EligibleCount. Define eligibility, attribution and conversion window.
Optional
Alternative Rate with denominator and documented calculation.
Preparation
Calculate converted/eligible counts for each comparable period; preserve both numerator and denominator.
Units
Converted/Eligible entities per period. Distinguish percentage-point changes fromrelative percentage changes.
Example of the input layout: Long tableA long table: one row per observation, few columns. New data adds rows.PeriodEligibleCouConvertedCoAABBC
A long table: one row per observation, few columns. New data adds rows.

Worked example

Jan 50/1000=5%; Feb 72/1200=6%. Improvement is 1 percentage point or 20% relative, not 1%.

PeriodEligibleCountConvertedCount
2026-01100050
2026-02120072