Delay repay compensation volume jumps 55% in a year

[AlterEgo]

The political pressure was needed to make operators actually operate the service they said they would.

No, it wasn't. There are vastly larger financial incentives that are automatic, immediate, and certain. Delay Repay is a drop in the ocean, especially when revenue risk (or even the operation itself) is wholly nationalised. DR15 came about as a political sop following the evil and naughty privately owned GTR having some awful performance. But who was actually responsible for most of that?

UK Government Web Archive

This Page is [ARCHIVED CONTENT] and shows what the site page https://www.orr.gov.uk/search-news/orr-investigation-finds-network-rail-breach-licence-2014-15 looked like on 2 Jun 2026 at 05:44:53

webarchive.nationalarchives.gov.uk

Oh look, it was actually publicly owned Network Rail who were found to be in breach of their licence.

(Note how the ORR suggested NR could pay the fine, or offer reparations to customers it delayed. What do you think it did?)

Back in the days when I was commuting into London from Hemel it was literally one trip in ten which arrived on time. A good half were heading towards 10-15 minutes late; LM had no financial incentive to fix it

If they were responsible for the delays, yes they did, via delay attribution. This is mostly an industry assurance scheme but it is this which is front and centre of the TOC's mind alongside CaSL and PPM.

Delay Repay is not a mechanism to improve train performance or provide better service. It is almost never a consideration in operational decisions. DR15 has been around for 10 years now, and yet here we are in a thread talking about how the railway has paid out a fat ton of money because its performance was poor. Is it really an incentive?

We all love getting money back, but let's not kid ourselves that this is necessarily good policy.

== Doublepost prevention - post automatically merged: Friday at 13:29 ==

And in many jobs you'd be on your way to being dismissed. But that's fine, because it's the railway and you should leave the house earlier and wait around for an hour before work.

£400 a year should make up for it then.

== Doublepost prevention - post automatically merged: Friday at 13:31 ==

Others have compared to the UK to other rail systems in Europe and used it to argue the UK is too generous. But I’ll tell you what: Deutsche Bahn wouldn’t offer such a terrible service if it cost them money when they delayed a train.

DB is an operational basket case, a better passenger compensation regime there would not move the dial.

== Doublepost prevention - post automatically merged: Friday at 13:34 ==

As I’ve edited my last post to say, London Midland always had a fondness for cancelling the Tring terminators which, at the time, were the only trains to serve the smaller inner-WCML stations. Under the old system they had every incentive to do so: they saved the cost of running the train and the passengers left waiting had absolutely no recourse against them.

That's right, there were no financial implications at all for cancelling the Tring terminators (!)

 

[Tetchytyke]

If they were responsible for the delays, yes they did, via delay attribution.

Delay attribution is a money-go-round, always has been, always will be. The TOCs have historically always done very well out of Schedule 8 payments.

It is almost never a consideration in operational decisions. DR15 has been around for 10 years now, and yet here we are in a thread talking about how the railway has paid out a fat ton of money because its performance was poor. Is it really an incentive?

It’s more of an incentive than it was before.

But if it isn’t an incentive, perhaps the payments back to passengers need to be increased until it is an incentive. Airlines only stopped playing silly buggers when EU261 came in and started hitting them in the wallet.

DR15 came about as a political sop following the evil and naughty privately owned GTR having some awful performance.

It wasn’t just GTR who, it turns out, were ripping off the taxpayer as well as their passengers.

And if Network Rail really are to blame then it doesn’t matter if Delay Repay goes up because Network Rail have already paid compensation to the TOC under Schedule 8. So why shouldn’t some of that Schedule 8 money make its way to the people actually affected: the passengers?

== Doublepost prevention - post automatically merged: Friday at 13:37 ==

That's right, there were no financial implications at all for cancelling the Tring terminators (!)

Nope, there weren’t. No financial consequences at all.

 

[mmh]

So what is £2.72 going to do there then?

A trivial amount of money, so yes, it'll do approximately nothing. Perhaps it would be better to refund 100% at a far shorter delay than now to make it less insignificant.

Delay Repay is a drop in the ocean

I agree, and have often said the same here. It'll be an insignificant amount of money and any fiddling with the current system on the grounds of cost would just be pointless tinkering.

the railway has paid out a fat ton of money because its performance was poor

So which is it? Is it an insignificant amount of money which should be scrapped as irrelevant, or a significant amount of money which should be scrapped as too costly? You seem to be suggesting it's somehow both.

 

[AlterEgo]

Delay attribution is a money-go-round, always has been, always will be. The TOCs have historically always done very well out of Schedule 8 payments.

This begs the question about whether you feel this Schedule is geared too strongly in favour of TOCs and whether the formula is correct (and how could you possibly know?).

Because obviously as it's compensation for a failure or acceptance of liability, surely it's not good faith to say they "do very well out of it" while arguing that delay repay is "not free money", it's compensation you the passenger are rightly due for bad service.

It’s more of an incentive than it was before.

It's not. Never ever heard of it being the basis or even the consideration for any operational decision.

And if Network Rail really are to blame then it doesn’t matter if Delay Repay goes up because Network Rail have already paid compensation to the TOC under Schedule 8. So why shouldn’t some of that Schedule 8 money make its way to the people actually affected: the passengers?

Train companies have to pay Delay Repay from their own revenues. Schedule 8 NR to TOC payments provide assurance only against the harm to future revenues and reputational risk caused by delays which were the responsibility of NR. If you want to argue for a different performance regime where Network Rail also pays delay repay to passengers (either directly or indirectly) then feel free to make an argument for it, but that is very different from what Schedule 8 actually is.

== Doublepost prevention - post automatically merged: Friday at 13:37 ==


Nope, there weren’t. No financial consequences at all.

Makes you wonder why they ever ran the Tring trains at all. Just cancel half of them all the time then. No consequences.

So which is it? Is it an insignificant amount of money which should be scrapped as irrelevant, or a significant amount of money which should be scrapped as too costly? You seem to be suggesting it's somehow both.

There are two aspects to DR15. Firstly the compensation due is generally quite small, but all Delay Repay has a processing and staffing cost. DR15 does generate a large volume of claims because this band of delay is the most frequently encountered as it is the least severe. Yes, much of this can be automated, but any such system will always need humans to be employed to look at appeals, and as a point of escalation before the Ombudsman gets involved. There's a fair amount of bureaucracy to support a scheme which does not really generate future business or keep existing business. Good customer relations should involve proper relationship building, not "press the button for two pounds you might notice in a week". Delay Repay almost never factors into any consumer decision to take the train.

 

[Tetchytyke]

It's not. Never ever heard of it being the basis or even the consideration for any operational decision.

Schrödinger’s Compensation: too small to be a financial incentive to improve performance, but too much of a financial burden to be allowed to continue.

 

[Baxenden Bank]

There are two aspects to DR15. Firstly the compensation due is generally quite small, but all Delay Repay has a processing and staffing cost. DR15 does generate a large volume of claims because this band of delay is the most frequently encountered as it is the least severe. Yes, much of this can be automated, but any such system will always need humans to be employed to look at appeals, and as a point of escalation before the Ombudsman gets involved. There's a fair amount of bureaucracy to support a scheme which does not really generate future business or keep existing business. Good customer relations should involve proper relationship building, not "press the button for two pounds you might notice in a week". Delay Repay almost never factors into any consumer decision to take the train.

My decision to take the train will, generally, not be affected by past, present or potential future Delay Repay compensation. But, when I have been substantially messed about and I can claim some money back (automatically, as of right) it does make me feel slightly less annoyed. My overall perception or 'feel-goodness' of the railway is raised. Prior to Delay Repay (and its predecessor) you were entirely reliant on a railway person making an ex-gratia payment of whatever amount. Receiving nothing after being substantially messed around did not give rise to any 'feel-goodness'.

 

[AlterEgo]

Schrödinger’s Compensation: too small to be a financial incentive to improve performance, but too much of a financial burden to be allowed to continue.

You wrongly frame the two outcomes as mutually exclusive for a soundbite. The real bite is in the administration costs it imposes.

It is perfectly possible in straitened times for sums of money to be too small to generate a desired and measurable effect, and also justifiable for that spending to be cut.

This is before we get to, once again, the crux of the matter anyway: Delay Repay is not a performance incentive scheme for the train company, was never designed to be, and as a result there are no measurements taken against this metric.

 

[Snow1964]

There are two aspects to DR15. Firstly the compensation due is generally quite small, but all Delay Repay has a processing and staffing cost. DR15 does generate a large volume of claims because this band of delay is the most frequently encountered as it is the least severe. Yes, much of this can be automated, but any such system will always need humans to be employed to look at appeals,

The processing time ought to be the same if the delay is 5 minutes or an hour.

The railway chooses to make the system labour intensive, whilst manual processing of strange ticket types will always be needed, the vast majority of tickets have a bar code or ticket booking ref.

In theory if a customer puts in the reference or code number and their departure time, then system ought to be able to look up the first train, add minimum connection time, then take booked or next train, then repeat for subsequent connections against the real time database, to check arrival time vs planned arrival time.

Oddly the operators who don't have 15-30 minutes category are usually those with itineraries which are easiest to automate.

I suspect some appeals are due to not providing sensible cut offs on automation, system rejects because train log shows arrived 29.5 minutes late at platform, passenger claims 31 minutes because that includes stopping and door opening time

 

[Haywain]

the operators who have got rid of the 15-30 minutes category

Which operators have got rid of that category? There are a number that have never had Delay Repay for sub-30 minute delays, but I don't believe any have had it and withdrawn it.

 

[AlterEgo]

The processing time ought to be the same if the delay is 5 minutes or an hour.

Nobody is saying it isn't. But DR15 introduces, by volume, a much larger number of claims. If 1 in 100 claims need human intervention and you process 10000 claims a month, and suddenly you now need to process 20000 claims a month, that doubles the person hours required.

Oddly the operators who have got rid of the 15-30 minutes category are usually those with itineraries which are easiest to automate.

No operators have "got rid of" DR15.

I think we may be in for an unpleasant learning experience now we have an increasingly nationalised railway and the government cannot pretend to be hurting private companies in response to poor service standards any more.

 

[Mark J]

Not surprised in the slightest.

Train delays and cancellations have been woeful this year. Particularly around Summer.

It seems every week trains are delayed, or cancelled for some reason, or another.

 

[meld3]

So what is £2.72 going to do there then?

It will put 50 kwh into my EV for another 225 miles of peak hour driving....

 

[Sonic1234]

Yes, much of this can be automated, but any such system will always need humans to be employed to look at appeals, and as a point of escalation before the Ombudsman gets involved.

The railway has a large labour force doing nothing for most of the day in the form of ticket office clerks, which politics says must be kept. Could they get involved in helping with assessing claims and customer relations work, rather than staring into space or at their phone?

== Doublepost prevention - post automatically merged: Sunday at 11:01 ==

the government cannot pretend to be hurting private companies in response to poor service standards any more.

This was the motivation for DR15. The Government pushed it as your chance to clip the wings of private fat cat operators.

Although I can think of people who if they knew Delay Repay was a rebate from the taxpayer it would motivate them to claim. Not everyone has the collectivist approach to public funds shown on here.

 

[duffield]

...

This is the thing to bear in mind: in isolation a 15-minute delay is no big deal. But when it is happening 3-4 times a week, as it did with me when I was commuting with LM, it is a big deal. On a bad week I’d be losing two hours a week to their incompetence.
...

Then maybe a compromise would be to only allow claims between 15-30 minutes when two or more are being claimed together during a week or other suitable period, so only claims for a total of 30+ minutes are allowed.

 

[mmh]

Then maybe a compromise would be to only allow claims between 15-30 minutes when two or more are being claimed together during a week or other suitable period, so only claims for a total of 30+ minutes are allowed.

Complicating processes rarely makes them cheaper.

I really doubt that the staffing cost for processing delay repay will be significant. If it is, then the TOCs should have proposed something like reducing duplication between them by centralising it.

 

[duffield]

Complicating processes rarely makes them cheaper.

I really doubt that the staffing cost for processing delay repay will be significant. If it is, then the TOCs should have proposed something like reducing duplication between them by centralising it.

I would think that the admin+payout savings from eliminating the processing of all the single "one-off" claims for 15-30 mins would more than cover the cost of the additional complications of only allowing claims totally 30 minutes or more. Most of those claims subject to "complication" would probably be regular commuters with a simple journey getting repeatedly delayed and would likely go down the automation route. It doesn't seem like much of a complication to introduce in order to eliminate a whole bunch of "trivial" claims completely.

 

[GRALISTAIR]

All I can say is - “thank you Avanti! “

 

[styles]

I once arrived at Emeryville on the California Zephyr which was 8 hours 37 minutes late. There was no apology and certainly no delay repay. So perhaps we should be grateful for small mercies here in the UK!

The GB Delay Repay regime is one of the most generous I've seen, both in Europe and around the world. I'm sure there's some examples of more generous arrangements, possibly operator-specific on a discretionary basis, but GB does have it pretty good in terms of what you're in theory entitled to. The difficulty sometimes is the practicality of it, as the disputes forum has plenty of examples of being passed between operators when they can't agree who should pay the delay repay, or only giving partial compensation because a customer used the word 'refund' instead of 'compensation' in an email, or whatever else.

 

[Sonic1234]

I really doubt that the staffing cost for processing delay repay will be significant.

Depends how much TOCs are willing to trust their automation. On GTR a simple no-changes journey gets processed really quickly. Journeys involving connections, particularly non-GTR connections, seems to get sent for manual review.

 

[Krokodil]

I've claimed with LNER, TPE and TFW in the last 2 weeks and have got all the funds already. I'm still waiting for an Avanti from August...

Avanti can be very quick if it's a straightforward claim or very slow if it's not.

Which bit of the scheme do you think is generous? The delay thresholds at which payment is made? The percentages of the fare paid used to calculate the payment?

It's considerably more generous than anywhere else. In Germany a two hour delay will only get you 50% of the fare. That's fairly typical.

It has the effect of discounting unreliable operators such as Great Northern, which perhaps has a benefit in terms of reducing the risk of modal shift back to private transport

I doubt that anyone ever takes delay repay into consideration when making decisions about their commute to work.

The processing time ought to be the same if the delay is 5 minutes or an hour.

I imagine that shorter delays are simpler to handle - complex journeys involving connections (and the potential to miss them) are more likely to result in a big delay than a straightforward a-b journey. On a simple journey a 5 minute delay is inconsequential, but on a complex one it might mean missing a connection and setting you back an hour or more.

The railway has a large labour force doing nothing for most of the day in the form of ticket office clerks, which politics says must be kept.

I must be incredibly unlucky if they're really doing "nothing" all day as you claim, because I needed a bike reservation from Euston on Friday and waited ages in the queue. Then going from my local station on Sunday there was a queue for the booking office too.

 

[D1537]

Avanti can be very quick if it's a straightforward claim or very slow if it's not.

Pretty straightforward, this one. Included a connecting journey with another TOC which was also late but made the connection, the Avanti arrived 20 late and ended up 47 late. Not their fault, but of course that's irrelevant.

 

[mmh]

I must be incredibly unlucky if they're really doing "nothing" all day as you claim, because I needed a bike reservation from Euston on Friday and waited ages in the queue. Then going from my local station on Sunday there was a queue for the booking office too

This isn't just the case at huge stations like Euston. A station I use all the time on the North Wales coast which has fantastic staff, many of which recognise me and say hello before I have, has staff doing all of selling tickets, dispatching trains, providing passenger assistance, manning the ticket barrier and providing onward travel advice.

One went out of his way to suggest I looked into buying a season ticket when he recognised that I'd used a "bog roll" onboard ticket a few days in a row.

But there are people on here who'd say that's the wrong way to run a station, and they're "sat doing nothing."

 

[Dr Hoo]

This isn't just the case at huge stations like Euston. A station I use all the time on the North Wales coast which has fantastic staff, many of which recognise me and say hello before I have, has staff doing all of selling tickets, dispatching trains, providing passenger assistance, manning the ticket barrier and providing onward travel advice.

One went out of his way to suggest I looked into buying a season ticket when he recognised that I'd used a "bog roll" onboard ticket a few days in a row.

But there are people on here who'd say that's the wrong way to run a station, and they're "sat doing nothing."

I think that the concept of staff with ‘little to do’ is perhaps more aimed at wayside suburban stations with no ticket barriers and an infrequent service, e.g. around Manchester. Some of these stations get very few passengers needing assistance as at least one platform has no ramp or lift. In any event, all trains have guards and carry ramps suitable for the unit type.
There may well be a few passengers at the window for a short while when a train is due but a majority seem to have e-tickets of some sort anyway.
Conceptually there would be a ‘pool’ of Delay Repay cases needing human review. At a quiet moment a clerk could ‘dip in’ and start working on a case. Even if a customer comes to the window they could hit a ‘hold’ button and return to the task at the next opportunity.

 

[andy33gmail]

I doubt that anyone ever takes delay repay into consideration when making decisions about their commute to work.

In fairness, I do

 

[redreni]

Was wondering about DR, and whether there would be any sense in removing delays where the railway is not at fault, [...]. LNER DR is tax payers' money at the end of the day.

No it isn't, it's fare payers' money.

People pay their fares on the basis that when they are delayed above certain thresholds (and if you're a regular traveller it is when, rather than if), they get some of the money back. If you buy an annual season ticket you do so in the sure and certain knowledge you're going to get some of that money back. Ditto if you travel regularly on single or return tickets. These aren't discretionary gestures or goodwill payments; they're contractual.

As for the hostility in this thread to DR15; when the Thameslink Luton to Rainham service first launched I was commuting on it - or would have been if anybody had anticipated that drivers with route knowledge would be required to deliver it. It was invariably cancelled and I was delayed by about 18 minutes each way every day, 5 days a week, for months. The idea that I should have got zero compensation for that is absurd - we are talking about 150 plus delay minutes a week. People who make occasional leisure journeys, encounter occasional delays and regard any DR entitlement as a bonus really ought to think about the impact of their proposals on people who use the railway in a different way and are impacted severely by the cumulative effect of constant short-ish delays and have a contractual right to modest compensation.

 

[Watershed]

No it isn't, it's fare payers' money.

People pay their fares on the basis that when they are delayed above certain thresholds (and if you're a regular traveller it is when, rather than if), they get some of the money back. If you buy an annual season ticket you do so in the sure and certain knowledge you're going to get some of that money back. Ditto if you travel regularly on single or return tickets. These aren't discretionary gestures or goodwill payments; they're contractual.

I think the fact that the payments wouldn't arise unless someone bought a ticket is the clearly indication that it's funded by fare payers rather than taxpayers. It's not like the cost of running the network or employing staff, which is incurred regardless of the actual number of passengers.

 

[WillC]

No it isn't, it's fare payers' money.

People pay their fares on the basis that when they are delayed above certain thresholds (and if you're a regular traveller it is when, rather than if), they get some of the money back. If you buy an annual season ticket you do so in the sure and certain knowledge you're going to get some of that money back. Ditto if you travel regularly on single or return tickets. These aren't discretionary gestures or goodwill payments; they're contractual.

As for the hostility in this thread to DR15; when the Thameslink Luton to Rainham service first launched I was commuting on it - or would have been if anybody had anticipated that drivers with route knowledge would be required to deliver it. It was invariably cancelled and I was delayed by about 18 minutes each way every day, 5 days a week, for months. The idea that I should have got zero compensation for that is absurd - we are talking about 150 plus delay minutes a week. People who make occasional leisure journeys, encounter occasional delays and regard any DR entitlement as a bonus really ought to think about the impact of their proposals on people who use the railway in a different way and are impacted severely by the cumulative effect of constant short-ish delays and have a contractual right to modest compensation.

I was under the impression the railways were heavily subsidised, vis. it's more than just the fare. In any case, to your point, in the event the railway is clearly at fault, then compensation seems reasonable; my original question related to whether the same is true when the fault lies elsewhere.

 

[JamesT]

I was under the impression the railways were heavily subsidised, vis. it's more than just the fare. In any case, to your point, in the event the railway is clearly at fault, then compensation seems reasonable; my original question related to whether the same is true when the fault lies elsewhere.

How much extra resource would be required to work out whether it's the railway at fault? Is that going to cost more than just giving compensation even when it's not the railway that was at fault?

Even if the railway isn't the root cause of the problem, the Consumer Rights Act places obligations on providers to use 'reasonable care and skill', so their response might still leave them liable. Blanket Delay Repay doesn't necessarily excuse them, but will satisfy most customers and saves needing a more complicated scheme.

 

[AlterEgo]

How much extra resource would be required to work out whether it's the railway at fault?

None, really. It used to be the case that compensation was only paid out on this basis and even more than a decade ago it was trivial to do this owing to delay attribution codes.

 

[Krokodil]

Just had Avanti approve a claim from Sunday. As far as I was concerned, the delay was two hours, yet they've only approved 50% compo. Looking at RTT, there's half a minute in it. I'll see what the response to my appeal is.

This will be quite common where a delay bounces you back onto the following service, which makes your delay 30/60/120 minutes but because that service ran on time the TOC says that the delay was only 29½/59½/119½ minutes. So should the thresholds be changed to something like 25/55/115 minutes? In this case that half minute (according to TRUST) is worth £93.