Background
1. Passenger assistance remains a key enabler of independent rail travel for disabled and older passengers. Through their Accessible Travel Policies (ATPs), train and station operators commit to providing assistance to disabled and older passengers. This assistance can be booked in advance through Passenger Assist or requested at the station through turn-up-and-go arrangements. Reliable delivery of passenger assistance is fundamental to ensuring disabled passengers can travel with confidence and enjoy equal access to the rail network.
2. Demand for passenger assistance continues to increase year-on-year. Record numbers of assistance requests are now being made across the network. This growth places greater importance on operators being able to deliver assistance consistently, understand the experiences of passengers who require that assistance, and respond effectively when things go wrong.
3. ORR’s latest survey of users of passenger assistance (Experiences of Passenger Assist survey) shows that satisfaction remains high: 87% of survey respondents are satisfied with the end-to-end experience from booking through to receiving assistance. However, reliability remains a significant challenge: a proportion of passengers continue to report receiving only part, or none, of the assistance they booked, with potentially serious consequences for their ability to travel independently and with confidence. The survey also demonstrates the importance of reliable assistance in supporting passengers' confidence to make future journeys.
4. In response to these challenges, ORR introduced a benchmarking framework in 2025 to provide greater transparency about operators' performance and organisational capability. Benchmarking helps us identify where performance is strongest, where regulatory attention may be required, and where good practice can be shared more widely across the industry. It also supports operators by highlighting opportunities for improvement and providing insight into approaches being adopted elsewhere in the sector. This year's report builds on that framework.
Benchmarking methodology
Evidence base and metrics
5. ORR uses passengers’ experience of assistance to make an assessment of each operator’s delivery. In our 2026 assessment, we use four metrics derived from two sources: three from ORR’s ongoing Experiences of Passenger Assist survey; and an additional metric new for 2026, ratings given by passengers through the Passenger Assist mobile app. For both sources, data is for the period April 2023 and March 2026.
6. Metrics based on data collected through ORR’s Passenger Assist survey are
- Reliability, measuring the percentage of passengers who reported receiving none of the assistance they had booked;
- Satisfaction with assistance received at station, capturing the proportion of passengers who were ‘satisfied’ or ‘very satisfied’ with the assistance they received, when met; and
- Satisfaction with staff knowledge and proficiency at station, reflecting passengers’ views on the competence of staff delivering that assistance, when met.
7. For each of these metrics, up to a total of 26,389 survey responses have been analysed across all operators (depending on the specific metric). See Annex A for individual operator sample sizes.
8. It should be noted that a different approach to asking passengers about the reliability of their assistance was undertaken in periods one to six of 2025 to 2026. Following a review in period seven, the question reverted to the original question text used in previous years for the rest of the year. For 2025 to 2026, only data from period seven onwards has been included in analysis of this metric. Data from periods one to six has not been included.
9. The new metric based on ratings submitted by passengers through the Passenger Assist mobile app is:
- Satisfaction with assistance experience at stations, as submitted through the Passenger Assist mobile app. This measures the percentage of respondents who rated their assistance experience at a station as four or five stars out of five.
10. This Passenger Assist data offers value to the benchmarking process by providing us with a more rounded view of passenger experience. The dataset’s sample size is also substantial: across the three-year period, 123,234 ratings through the Passenger Assist mobile app have been analysed.
Passenger Assist rating via mobile app – source of data
11. Assisted journeys are recorded in the Passenger Assist system, whether they are booked by phone, online, through the Passenger Assist mobile app or at a station. Passengers who have set up an account on the Passenger Assist mobile app can view their journeys either where they have booked in advance, or where they are making a turn-up-and-go (TUAG) journey and staff have both entered the journey details into Passenger Assist and connected the journey to the individual’s Passenger Assist account.
12. After completing their journey, passengers whose journeys are linked to their account on the Passenger Assist mobile app are invited to submit a satisfaction rating, with a range of one to five stars. Ratings are provided for the experience of assistance at a station – for a single-leg journey, a passenger therefore can rate assistance at the boarding and alighting stations separately.
13. Passenger ratings are stored by Transreport, the company who operates the Passenger Assist system under contract with the Rail Delivery Group (RDG). RDG download anonymised ratings data from Transreport and pass it to ORR. ORR carries out all data analysis.
14. Over the last three years TUAG assists account for 9% of ratings analysed. This figure also includes short notice bookings accepted within the 2-hour notice period normally required by operators for ‘booked assistance’.
Confidence to travel
15. Since April 2025 we have included a new question in ORR’s Passenger Assist survey to understand how passengers’ experience of assistance has impacted their confidence to travel in the future.
16. As we only have data for a single year, we have not used responses to this new question in our overall assessment of operator performance. Once we have sufficient trend data, we intend to include a new metric in our assessment, capturing the proportion of passengers who were ‘less confident’ or ‘much less confident’ about making a similar journey in the future.
Benchmarking
17. Performance has been attributed to the operator with overall responsibility for assistance at a station. In most cases this is the station operator. However, at some Network Rail stations one or more train operators are responsible for delivering assistance instead and our analysis reflects this.
18. To ensure a robust and representative analysis, as in 2025 we applied a weighted scoring system across all three-year metrics: weighted 1:2:3 towards the most recent year.
19. As outlined above, only data from period seven onwards has been included for 2025 to 2026 when deriving the reliability metric. Other years and metrics are unaffected. This variation in data coverage is reflected in our weighting calculation, as set out below – as the most recent year, 2025 to 2026 is subject to triple weighting but this is halved to one-and-a-half times weighting because just half the year’s data has been included in the analysis.
Illustration of how an operator’s weighted score for each metric is calculated
Reliability metric: (Y1 + (𝑌2x2) + (𝑌3x1.5)) ÷ 4.5 = Result
All other metrics: (Y1 + (𝑌2x2) + (𝑌3x3)) ÷ 6 = Result
Table 1: How an operator’s weighted score for each metric is calculated
| Metric | 2023-24 | 2024-25 | 2025-26 | Three-year weighted score for each metric |
|---|---|---|---|---|
| Reliability (no assistance received) | 12% | 11% | 8% | (12+20+12)/4.5 =10% |
| Satisfaction with assistance | 92% | 93% | 94% | (92+186+282)/6 =93% |
| Satisfaction with staff | 94% | 94% | 96% | (94+188+288)/6 =95% |
| Satisfaction via app | 82% | 84% | 86% | (82+168+256)/6 =84% |
20. The overall Delivery score was calculated using the three-year weighted score for each metric and following weightings: 50% for reliability and 50% across the three satisfaction metrics (split evenly).
Illustration of how an operator’s overall score is calculated
Overall: (((100 – Reliability) x3) + Satisfaction with assistance + Satisfaction with staff + Satisfaction via app) ÷ 6 = Result
e.g. ((91 x 3) + 93 + 95 + 84) ÷ 6 = 91
21. Note that in this calculation the Reliability score is subtracted from 100 to provide a metric that can be weighted versus the Satisfaction metric scores. So, for example, a 9% ‘no assistance received’ reliability score becomes a 91% ‘some or all assistance received’ score for this purpose.
22. This year we ranked operators based on an overall Delivery score; last year operators were ranked on reliability alone. We are now able to rank on overall delivery because of the addition of the satisfaction data via the app, which is both reliable and provides greater performance variability between operators.
Operators in scope
23. To ensure statistically robust and meaningful comparison between operators, operators must have had at least 50 survey responses per year for all metrics to be included in the overall Delivery assessment. The following operators were therefore included in the full Delivery assessment (full sample sizes for each metric are provided in Annex A):
- Avanti West Coast
- Chiltern Railways
- East Midlands Railway (EMR)
- Greater Thameslink Railway (GTR)
- Great Western Railway (GWR)
- Greater Anglia
- London North Eastern Railway (LNER)
- Network Rail
- Northern Trains
- ScotRail
- South Western Railway (SWR)
- Southeastern
- TransPennine Express (TPE)
- Transport for Wales (TfW)
- West Midlands Trains (WMT)
24. Operators excluded from the full Delivery assessment due to low response rates or because they do not manage stations include First Rail London (London Overground), c2c, CrossCountry, Grand Central, GTS Rail (Elizabeth line), Hull Trains, London Underground and Lumo. We continue to consider opportunities to expand future coverage by increasing survey sample sizes and exploring alternative data sources.
25. The addition of ratings data submitted by passengers via the Passenger Assist mobile app has given us improved visibility of these operators’ performance in delivering assistance, due to the larger data sample – therefore, while not included in the full Delivery assessment, we are able to engage individual operators where we have had cause for concern.
26. Note that ownership of both GTR and First Rail London changed in May 2026, resulting in name changes. For the duration of the three-year benchmarking period, these operators were known as Govia Thameslink Railway (GTR) and Arriva Rail London (ARL – referred to here as the more familiar London Overground).
Capability to Improve assessment
27. In implementing the benchmarking framework last year, our stated intention was to undertake detailed Capability to Improve assessments of the weakest performers. In 2025, we undertook such assessments of SWR and WMT, alongside our ongoing work with Northern to improve its performance; these were the weakest performers in terms of reliability of assistance. For more information, see the 2025 Benchmarking Report.
28. In 2026, these three operators remain amongst the weakest performers. We have therefore not carried out further detailed Capability to Improve assessments but have continued our engagement with them as they developed and delivered their improvement plans. We are also undertaking targeted engagement with TfW and Avanti, where the latest benchmarking results or passenger feedback have identified new and emerging concerns.
29. The result of this work is set out in more detail in Chapter 3.