How often claimants actually win in published tribunal judgments
How claimant outcomes break down in published employment tribunal judgments, including merits wins, partial wins, strike-outs, and withdrawals.
Claimant wins
29.8%
6,632 claimant or partial wins out of 22,235 visible cases
Respondent wins
29.2%
6,499 published judgments are tagged as respondent-successful
Non-merits exits
33.7%
7,500 cases end as strike-outs, withdrawals, or settlements
Unclear / other
7.2%
1,604 visible cases are outside the main outcome buckets shown below
What the current published data says
Tribunal Intel currently indexes 22,235 visible employment tribunal decisions in this dataset view. Of those, 6,632 are tagged as either fully claimant-successful or partially successful.
That works out to roughly 29.8%. In other words, published judgments in this dataset are not dominated by claimant wins. Respondent-successful outcomes on their own currently account for 6,499 cases, which is about 29.2% of the visible set.
A large block of cases also never reach a clean claimant-versus-respondent merits win at all. Strike-outs, withdrawals, and published settlements together account for 7,500 cases, or 33.7%.
That matters because users often ask “how often do claimants win?” as if every published case ends in a straight merits judgment. The published record is messier than that: many cases exit on procedural grounds, are withdrawn, or are recorded in forms that do not map neatly onto a simple claimant win rate.
Outcome mix in the current visible dataset
How to interpret the headline number
Published cases are selective
The published judgments skew toward cases that reached a tribunal decision point. That is not the same thing as all claims issued, negotiated, or resolved.
Procedure matters
Strike-outs and withdrawals are a large enough share of the visible record that a simple claimant-vs-respondent framing leaves out a meaningful part of the dataset.
Partial success still counts
A claimant can win something important without winning every issue. That is why partial-success cases are worth separating out rather than folding them into a crude pass/fail headline.