Win-Loss Analysis Sankey Diagram: Editable Template
Break closed deals down by customer segment, outcome and the main reason deals were lost. Replace the counts with your own win-loss data to see where losses concentrate.
A win-loss Sankey connects segment to outcome to reason. You can see not only how many deals were lost but which segment lost them and why.
Five hundred closed deals are split across SMB, mid-market and enterprise accounts. Each is won or lost, and each lost deal is assigned one main reason.
Loss reasons are only as reliable as the way they are recorded. A single required field with a short fixed list works better than free text. The figures are invented.
How to read this diagram
- 500 deals closed: 147 won (29%) and 353 lost.
- SMB and mid-market both win 30%; enterprise wins 24% (12 of 50).
- Two reasons account for 65% of losses: choosing a competitor (120) and price (110).
- A quarter of lost deals (90) ended in no decision rather than a decision for someone else.
The data behind it
Invented figures for illustration. Not data about any real business.
| From | To | Amount (deals) |
|---|---|---|
| SMB | Won | 90 |
| SMB | Lost | 210 |
| Mid-market | Won | 45 |
| Mid-market | Lost | 105 |
| Enterprise | Won | 12 |
| Enterprise | Lost | 38 |
| Lost | Price | 110 |
| Lost | Chose a competitor | 120 |
| Lost | No decision | 90 |
| Lost | Missing feature | 33 |
The same data as CSV
source,target,value
SMB,Won,90
SMB,Lost,210
Mid-market,Won,45
Mid-market,Lost,105
Enterprise,Won,12
Enterprise,Lost,38
Lost,Price,110
Lost,Chose a competitor,120
Lost,No decision,90
Lost,Missing feature,33How it works
Fix a short list of loss reasons
Use four to six mutually exclusive reasons and require one for every lost deal.
Pick the period and the segments
Use deals closed in one period, grouped into a few segments.
Count outcomes and reasons
Count won and lost per segment, then lost deals per reason.
Enter the flows
Check that the reasons add up to the total lost.
Frequently asked questions
Why show only the reason for lost deals?
Reasons for winning are usually less well recorded. If you have them, add a second set of nodes after Won.
Can a deal have more than one reason?
Not in a Sankey, since each deal is counted once. Record the primary reason, or use a different chart for multi-reason data.
Edit this win-loss Sankey
Open the example as your own chart and enter your closed-deal counts.