How often do tribunals reduce repairs and maintenance charges?
Across 2,268 individually challenged repairs and maintenance items in First-tier Tribunal service charge decisions from 2019–2026, 39% were reduced or disallowed; where the tribunal made a reduction, the median was 50% (n=343 reductions with amounts stated). (n=2,268, as of 15 August 2026)
The figures on this page cover every individually challenged repairs and maintenance item in our corpus of published First-tier Tribunal service charge decisions, and are recomputed nightly.
About these figures: Outcomes reflect disputes that reached the First-tier Tribunal, not portfolio-wide quality. Small samples are noisy; every figure links to the underlying decisions.
Outcomes for challenged repairs and maintenance items
| Outcome | Items | Share |
|---|---|---|
| Reduced | 466 | 20.5% |
| Disallowed entirely | 419 | 18.5% |
| Allowed in full | 1,302 | 57.4% |
| Withdrawn | 19 | 0.8% |
Most common grounds when the tribunal cut the charge
- Costs unreasonably incurred (s19(1)(a)) — 335 items
- Other grounds — 87 items
- Not payable under the lease — 105 items
- Landlord could not evidence the cost — 324 items
- Poor standard of work — 104 items
- Demand formally invalid — 37 items
- Works not necessary — 88 items
- No Section 20 consultation — 49 items
By tribunal region
| Region | Items (n) | Reduced or disallowed | Median reduction when reduced |
|---|---|---|---|
| London | 1,288 | 37.6% | 50% |
| Eastern | 310 | 34.2% | 52% |
| Southern | 365 | 41.6% | 46.8% |
| Midlands | 132 | 58.3% | 42.9% |
| Northern | 173 | 38.2% | 45% |
Methodology
These statistics are computed from the published decisions of the First-tier Tribunal (Property Chamber) in service charge cases (case types LSC, LIS and LDC). Each decision is parsed into a structured record — the sums challenged, the sums allowed, the outcome per cost head, and the orders made — and the aggregates on this page are recomputed nightly in plain arithmetic from those records. No figure on this page is estimated, modelled or hand-typed; each carries its sample size. Current corpus: 4,244 decisions covering 18,031 individually disputed items, last updated 15 August 2026.
Read this before quoting: Outcomes reflect disputes that reached the First-tier Tribunal, not portfolio-wide quality. Small samples are noisy; every figure links to the underlying decisions.