Wellington has been making some noises about moving towards integrated ticketing and fares for some time and doing so is formally listed Greater Wellington Regional Council’s Regional Public Transport Plan (RPTP). In the RPTP they say “Improving the fares and ticketing system is the next significant element in the modernisation of Wellington’s public transport system”.

As Auckland learnt changing a ticketing system can be very difficult, especi byally when you have some operators wanting to force their own system on the public. Now it seems that Wellington could be set to have a repeat of some of Auckland’s ticketing issues.

Hop Card

The days of Wellingtonians swiping their Snapper cards on the capital’s buses could be numbered.

The Government is pushing to have Auckland’s Hop card system in the capital instead and Wellington’s political leaders are worried about the potential impact of that on ratepayers, as well as the future of Snapper.

In 2018, the entire Wellington region will shift to having one electronic smartcard that people can use to pay for all bus, train and ferry travel, in much the same way Aucklanders use their Hop card.

Snapper, which is already on more than 300 Wellington and Hutt Valley buses, is expected to be among those bidding to provide the smartcard technology when Greater Wellington Regional Council opens the contract up to tender.

But their chances of success look slim after the New Zealand Transport Agency wrote to the regional council stating its preference that Auckland’s system be extended to the capital.

The agency also proposed that its wholly-owned subsidiary company, New Zealand Transport Ticketing Limited, be directly appointed to provide the $50 million Wellington network, as it does in Auckland.

The door was left open for the regional council to chose a different technology provider, but the agency said that company would have to represent better value and less risk than simply bringing the Auckland system south.

It is expected Wellingtonians will be able to use their new cards in Auckland, and vice versa.

Paul Swain, the regional council’s transport portfolio leader, said while the agency was leaving the decision in council’s hands, the message was that it should be buying into the Government’s system rather than Snapper.

But for a project as costly and as complicated as this, he believed a tender process was needed to make sure Wellington was getting the best technology on offer.

“We want the best deal possible for the ratepayer, for the taxpayer and for the passenger,” Swain said.

“But our concern here is that the bar is going to be set so high that it will be difficult for us to get funding for any system we get a tender for.”

The Transport Agency and regional council are planning to split the $50m cost of the new system between them. But the agency can opt to contribute less if it is unhappy with the technology provider.

I’ve seen some comments concerned by the stance by the NZTA over this but it is actually nothing new and harks back to decisions made many years ago.

  1. At the time Auckland had held a tender for an integrated ticketing solution which Thales won. Infratil – the owner of Snapper – didn’t like the outcome and pushed the then mister to review the decision.
  2. The minister asked the NZTA to review the tender process and when they did so they found the system to have been the best on offer. However they also recognised that integrated ticketing was something that was going to be pushed by many councils over the coming years for which the NZTA would be asked to help cover the costs of installing them all.
  3. To manage what could be a lot of expensive system requests the NZTA took over buying of the system with the intention of it being used nationally. They said at the time other regions would need a good business case for not using it.

It will be interesting to see what happens. Snapper is much more entrenched in Wellington than it was in Auckland and so I imagine there will be a lot of strong views down there.

What do you think will out should happen?

Share this


  1. I think any other city should avoid it like the plague.
    The system is flawed in so many idiotic ways. Just one example: if the system messes up my tag on/tag off and puts my card into negative balance, I can’t ride the bus even though I have a current and active $190 monthly pass. That’s patently stupid and unnecessarily punitive.
    Or like how it takes up to 3 days for the system to recognise a payment made online. Mental.

    It smacks of old world IT and any self-respecting city should expect better.

      1. So AT policy is to blame for the infrastructural stupidity that dictates the terminals on buses are dumb slaves rather than smart connected terminals, making online transactions take “up to 72 hours” to register?
        And for not allowing monthly passes to operate completely independently of a cash balance?

        Obviously AT can override the decision to make negative balances block the card but can they do that selectively for active monthly pass users or would they expose the network to wholesale fraud?

        1. Online payments happen overnight, the 3 days is an arse covering statement by AT in case a bus doesn’t get back to a depot overnight for a WiFi update. If they rolled out connections to buses they can easily make updates more regularly.

          1. I know how it works. My point was that the bus returning to the depot to download a fresh batch of transactions is antiquated, stupidly limiting, and dictated by the system (as opposed to AT’s strategy).

          2. It’s an AT decision not to either have more points around the city where buses can be updated or have mobile connections to enable it to update more frequently. That’s an AT decision not a fault if the system

          3. Barry, couple of comments:

            1. Despite its imperfections HOP has contributed to rapid patronage growth since being implemented; and
            2. HOP cost Auckland not much by world standards. MyKi in Melbourne cost $1 billion, for example, whereas Auckland paid (which I understand was about $100 million)

            Now my understanding is that the issues you mention are being progressively fixed, but would also suggest that it wasn’t worth delaying HOP until after this was complete.

            In terms of HOP being antiquated, my answer is I don’t think so. I think that the issues you mention are caused by wider technological dependencies.

            For example: You ask why the buses don’t have real-time wifi onboard. Well, not all buses have wifi and it’s reasonable to suggest that installing wifi would cost mmore money. And let’s say that AT and the operators wanted their new onboard wifi system to not only deal with ticketing related information but also vehicle positioning and even vehicle performance. Moreover, let’s say that over the next few years many of the old buses are about to be retired from service and replaced with new buses. In this context, AT and the operators may have quite reasonable decided it wasn’t worth installing wifi tech on old buses. Instead, they may have decided it was more efficient to simply ensure it was installed on new buses as they enter the fleet, once the whole fleet has wifi tech, then move to the new system overnight.

            This is just an example of how you may be underestimating the complexity involved in efficiently deploying new technology. Sure, with unlimited money we could throw out all the old buses and technology and move to a whizz-bang system overnight. But we don’t have unlimited funds, hence the need for these kinds of transitional arrangements. I don’t think it’s perfect. but I don’t think it’s worth getting too enraged about. And it may ultimately be a more efficient way to spend tax-payer money than trying to do everything at once.

            P.s. I suspect AT are working on implementing exactly the system you describe as we speak. But my bet is they’ll be doing this via new PTOM contracts and their associated vehicle procurement guidelines.

    1. Surely the 3 day delay is about the banking system. If we all used the same bank, there would likely be no delay…..but we don’t.

      I have a Canadian phone account. It can take an Internet payment (from a Canadian bank to a Canadian telco) to my phone account TEN DAYS to hit the phone account. NZ’s banking system is paradise to what Canadians must endure.

      But it does mean we have to wait the weekend for payments to clear. I top up my HOP Monday 8am to Friday Noon…and it’s normally (very close to 100%) there later in the day or the next morning. This is Kiwibank debit VISA to AT HOP. Maybe your bank is slower.

      1. It’s not the banking system. Payments are processed in real time, like any other online credit/debit card transaction.

        It’s that the buses don’t have an always-on internet connection, so there’s no way of pushing top-up details to the card readers on the fly. The system on each bus can only update its pending transactions when the bus returns to the depot and connects to the network.

        1. Thanks. OK….I’m seeing what I see because most of my public transport usage involves a train ride at one or the other of almost every journey. So the issue is the connectivity to the buses. I can’t see how any payment system with a chip in it can avoid that issue. (???)

    2. Use Auto Top – it’s the card’s best feature. I haven’t had to do anything with mine since the day ATHOP hit town. Top ups are instantaneous, and you can set it so when it hits say $10, it throws on 20.

    3. AT’s brand new HOP system is still well behind London’s 10+ year old Oyster system in terms of features and usability. You would only consider HOP good if you had never used a similar system elsewhere.
      It sounds like Snapper is already far more advanced than HOP.
      Why is integrated ticketing so hard to do on HOP? Makes you wonder what software is sitting behind it…

      1. Myth: Snapper is more sophisticated than HOP.

        Reality: HOP has faster read times than Snapper, i.e. 300ms for the latter versus 600ms for the latter. This is a noticeable difference in read times that does make a difference, e.g. at rail station gates. And it adds up to a lot of extra running time when you multiply it across every passenger every bus trip, every day, every year …


        1. Which is slower again than Oyster…. Oyster taps are instantaneous for the most part… not lag. They also work through things like wallets or jacket pockets etc.

          1. Perhaps. I haven’t seen any read time specs for oyster. It may be that tfl has implemented account based (rather than card based) tech, which should have a read time of say 100ms versus 200-300ms for hop (and 600ms for snapper). When i used oyster circa 5 years agi i don’t remember it being that much faster … but things may have have changed.

          2. My hop is instantaneous and I always use it through my wallet. In fact it’s so fast some times it tags on then tags a second time and gives me an already tagged on error before I’ve taken the wallet back again. Can’t say I’ve noticed oyster being noticeably faster than instant.

            What situations give you slow read times?

          3. I find tagging on a bus often you have to hold it/wave it etc for what seems a long time before it registers… and no it’s not my card other users have the same.

            Oyster doesn’t have to be registered and can be topped up both online and by machine. I’m not sure where the credit is held specifically but they can transfer the credit from one card/account to another easily if a card is lost/stolen/damaged. It is noticeably faster in that you don’t actually have to stop to do it.

          4. Bruce, I think most people struggle to distinguish between 100ms (account based systems), 300ms (HOP), and 600ms (Snapper). Where Oyster falls in this range I don’t know, although I wouldn’t be surprised if it’s in the 200-300ms range like HOP. If anyone can point me to some technical documentation/specification showing Oyster is faster, then please do!

            Otherwise I’m going to assume as a starting point that they’re much the same.

  2. We are a country smaller than Melbourne. A single ticketing system, one computer environment, one set of devices for stations / buses etc makes perfect sense from a support point of view.

    Given Hop is already in production in our only metro region I can’t see why we wouldn’t roll it out to the provincial centers like Wellington, Christchurch and Hamilton

    1. Because local governments are autonomous and can actually select what’s best for their own local solutions.

      Back in the ’80s, the SSC (and predecessors) determined where different departments would be located, how their staff would be appointed and promoted, and so on. That centralised control of resources and tools absolutely destroys autonomy, flexibility, and creativity.

      Competition – and that means each agent choosing the solution that best fits its own solution, works best.

      I’ve been part of a very large organisation that went from allowing each sub-group to negotiate its own cleaning, catering, and facilities support contracts, to one that imposed a single blanket contract. It cost our group (which had very sharp negotiators) MORE, and delivered LESS service. I’m sure it was better for the groups that sucked at negotiations, but it’s a terrible solution

  3. It is also worth mentioning that due to the intervention of Minister Joyce in the tender process, Snapper did get the opportunity to integrate their payment card into the Auckland Hop system, which has an open standard. They failed to achieve this in Auckland, but I’m not sure if there is an opportunity to repeat the exercise in Wellington?

    1. Snapper failed to achieve this because they didn’t have the Hop system to test with, and so pulled out (IMO, probably in frustration).

  4. I think that Wellington should evaluate it on a why not basis.

    But they need to look VERY carefully at whether HOP can do all the requirements it has and if not, whether they need to simplify their PT offerings first.

    Because things that HOP should have able to do easily, like manage train, bus and ferry fares and their associated products have proved to be anything but easy for HOP to handle.

    Supposedly AT have said because Thales told them the mix of routes, operators, products/fares used in Auckland is complex – which is why zonal fares have taken so long to roll out.

    I can’t imagine that Wellingtons requirements are going to much simpler or wildly different than Aucklands.
    Except perhaps with less operators, but fares, trains ,ferries and buses are all in the mix there like they are in Auckland and each is wanting the largest slice of the pie for any integrated fares.

    But the key point is that the larger one – being NZ bus, will be the reluctant party to HOP, so will fight tooth and nail to hang in there with Snapper, especially in their “home town”.

    However, it must be recognised, that AT have barely touched the capabilities of HOP in terms wider use, e.g. for parking, and that is where the power of a single system could really become beneficial not just for Aucklanders but for all.

    If you could have one card, whether branded as an AT HOP, Wellington “Whatever card”, Christchurch Metro card etc, and use it for travel/payments on all these systems it would be much better for all.

    And of course, NZTA should look to use the HOP system for payment for tolls via Northern Gateway and Tauranga tolled roads – either by linking your HOP card to your car registration, or via a transponder in your car, linked to your HOP “account” – that way you don’t even need to have multiple toll accounts.

    And leads naturally into time of use charging system – which is a likely future requirement that will become necessary in Auckland at least.

    1. I’ll agree that Auckland’s fare mix has been a mess. That’s a product of several private operators doing their own thing for well over a decade.

      When I shifted to Auckland in 2007 I could scarcely believe the incomprehensible mess public transport routes and ticketing were in. I’ve lived in cities that avoided the waste and cost of complexity through simplicity. Toronto is a great example of dead simple, cheap ticketing.

      I don’t think the answer to wasteful – mainly ideologically driven – complexity is to spend a vast sum of money on complex, fragile software that tries to do the same thing and – because it is complex – resists customising. It just highlights in real terms the costs of such wasteful complexity.

      If Auckland had a single, public transport operator and a handful of zones…..any system of ticketing would be cheap and straightforward.

      I don’t think it’s right to blame service providers like Thales or Snapper for “problems” that are actually consequences rooted in – and predictably flowing from – the many, MANY terrible decisions made about how to operate public transport in NZ generally.

    2. “And of course, NZTA should look to use the HOP system for payment for tolls via Northern Gateway and Tauranga tolled roads – either by linking your HOP card to your car registration, or via a transponder in your car, linked to your HOP “account” – that way you don’t even need to have multiple toll accounts”
      They want to link tolls to HOP.

      1. While I love the concept, it’s not technically possible without stopping the car. With AT Hop, the money is actually stored on the card, not on a backend system like with a bank. The card must be present at POS terminal to get the money off it. So it won’t work as an offline system (linked to rego), or by waving it outside the window while passing the toll gantries 😉

        I however really like what London has done – they’re moving away from Oyster to a credit (card) based system https://tfl.gov.uk/fares-and-payments/contactless – which allows so much more flexibility than what we have today with Thales.

        1. Incorrect, your money lives in the back end system, just like a bank , not on the card – Snapper does this however, but HOP doesn’t.

          That means along with protection against loss of the card itself, you could have multiple “devices” linked to the same account (like you can have joint EFTPOS cards or joint Credit Cards linked to the same account).

          One could be your HOP card, another could be on your mobile phone as contactless card and lastly a transponder in your car dashboard.

          Whether AT or NZTA want to allow this functionality is a seperate issue from whether you can.

          Also note HOP is NOT contact based – the card does NOT have to touch the reader to be read ok, it merely has to be proximate to the reader thats why it works even when its kept in your wallet.

          1. No Greg. The card is the master record and that was told to me directly by one of the project managers from Thales. There’s a reconciliation that occurs from the tags to ensure the back end systems are accurate.

          2. Yuck.

            But I suppose it means that the system can work even if totally offline. Provided you tag on or off at a connected reader somewhere sometimes.

          3. Yes that’s the point. If something goes wrong and no connection to the back end the system can still work perfectly fine and store up to three days of activity. If it was like bank payments there would be no way of knowing what the remaining balance of a card was if the bus didn’t have a permanent connection. Contactless credit cards work because the CC company can just bill you later but the need to check the exact balance is one reason why there is no contactless payments for likes of eftpos.

          4. Matt “but the need to check the exact balance is one reason why there is no contactless payments for likes of eftpos.”

            Except there is on debit; I have one.

          5. Thats because the Credit Card company manages the transaction and carries the risk if you don’t have the funds to cover the amount and the terminal is offline – just like they do for CC transactions.
            The retailer you use your debit card at pays a larger fee to the credit card company than they pay when you use EFTPOS, for that convenience though.
            So its more like a signature-less credit card than a true online debit card like EFTPOS is.

  5. They need a bit more thinking around the way the software works. I’ve had issues of having my card stop working after topping it up. It normally takes 24 hours but sometimes takes 2-3 days before the topup is applied to the card yet they take my money right away.

    1. The money goes onto my card within about 4 hours….on a weekday. I always top up on a weekday (via my phone / web browser). To get the money onto the card, you have to tag it…..but the tag that picks up the funds would be after the 3-4 hours has passed. Sooner may be too soon.

      Or….top up at a machine on a train platform and the money is on there instantly. But try to avoid doing it Britomart, where the huge queues most often are (too many people not thinking ahead) – anywhere else will do – a little awareness and planning is useful.

      But then…I always keep a minimum balance of $100 on the card and top it up to $150-$160 after it goes below $100. If it takes a few days…..no worries. But that’s generally just Saturday / Sunday factor.

    2. The balance is stored on the HOP card. But a ticket machine needs to learn of the top-up so that it can update the card. Normally this is done at the bus depo over WiFi, or by fibre for fixed units at stations. If you top up the bus units won’t know until they’re taken to the depo that evening.

      It could be better. For example dispatching Top-Ups over 3G would cost almost nothing if bargained properly (it’s a tiny amount of data).

  6. This is disgusting. I have absolutely no idea which is the superior solution, but we need an open tender with very clear requirements set.

    This sort of recommended solution stinks to high heaven of poor analysis.

  7. From a cold, rational perspective, I’d recommend dropping Snapper and adopting HOP in a heartbeat. Snapper is system that is inherently partisan with respect to the bus company who own it and has prevented integration of ticketing.

    The HOP system is now bedded in in Auckland and – for me – works perfectly. It just gets better all the time and using it on all modes is a major plus. Also….the public body gets to hold the millions in funds waiting to be spent instead of one of the operators.

    Being able to use these cards anywhere would be a huge plus….and this government will love it because the government gets to track the movements of everyone everywhere on public transport. As a surveillance tool, avoiding fragmentation of urban transport payment systems would be a priority for the SIS and police, surely.

    Paranoia (it isn’t really. I’m sure that’s part of why we use these services at all) aside…..HOP is great. Grab it, Wellington.

      1. It would have been cool to develop our own system, but that would take leadership NZ just doesn’t have. We had a chance 30 years ago to get in on the ground floor with developing wind power technology, but no one did it. Denmark did it instead and is reaping the huge rewards. It’s always cheaper in NZ to buy something someone invented elsewhere. We almost never back ourselves unless it’s about milk. There is a deep insecurity around “investing”. For good reason: there have been many failures.

        As for apps, those things come with time. “Agile development”. But in the meantime, the mobile web interface works really well and I can top up online with my saved payment method in a few seconds (assuming good connectivity).

        Tangent: I really like the “Track My Bus” app from AT, which shows how far away the bus I’m waiting for it. I also like the 3rd-party “Auckland Transit” app that lets me easily choose the bus stop or train station and quickly get up a real time board for it….and I can easily set and modify favourites. Plus, Google Maps’ public transport function is pretty good for generally finding out how to get from point A to Point B on public transport. I’ve now used it all over the planet. A killer app for any traveller.

        1. Melbourne developed their own system. It’s slower and less functional that hop, cost over a billion dollars and was delivered with limited functionality six years behind schedule.

          Thank the good lord above we didn’t go down that route!

          1. Projects can be done well or not well. To be done well, the goals needs to be clear, simple and achievable with existing methods. The projects I’ve seen go badly wrong were (any or all) ; poorly spec’d ; suffered from political interference and / or mission creep ; some key player in the project actually wanted it to fail to serve a wider agenda being at play….or something else. 🙂

          2. Isn’t it one of the rules of NZ government departments that they never buy available off the shelf software, they always get something tailor made for them at a huge additional cost with massive delays and that often never works.

          3. > Isn’t it one of the rules of NZ government departments that they never buy available off the shelf software, they always get something tailor made for them at a huge additional cost with massive delays and that often never works.

            Yes, pretty much. You might call it “Novopay Syndrome”.

            That said, NZ government departments often do try to buy something that’s “off-the-shelf” (but requires “just a little customisation”) and it turns out to actually be just the same as having to get something tailor-made, if not worse. When I worked at MSD some years ago they’d just spent $50m on a system called Cúram, which it turned out required more work to configure for MSD than it would to just write the software from scratch. So we wrote it from scratch, and that $50m went down the drain.

    1. Yes I agree. While there are frustrations with HOP, such as the length of time an online top up can take, the lack of an app etc, these feel like things that can be sorted, whereas Snapper works in a fundamentally different way that means it will be impossible to implement some basic features.

      With Snapper, your balance is stored on the card, but with HOP it is stored on “the system” and the card just links to that. So the only way a Snapper card can be topped up is to physically put the card next to a machine that updates the balance on the card. While you can pay for a device that you can use at home to allow you do top up via your computer, it will never allow the ability to top up from anything with a browser like HOP does. To me that is a fundamental weakness.

      Going to dairies to buy top ups does work, but it’s so 2002.

      1. This is not entirely correct.
        With HOP your card does have a balance, the trouble is telling your card that you topped it up online….

      2. You’re right there’s a fundamental difference in the two.

        Snapper was actually about trying to create a new contactless payment method before the big banks got on to it i.e. they wanted to be the new eftpos. In some ways PT was a loss leader to do get the cards on people’s hands.

        By comparison HOP has so far only cared about being a PT ticket. They do want to expand it to other things such as parking, tolls etc. but that is often constrained by being it hard to get priority within a large organisation and the issues with dealing with Thales. As an example of the former, I’m aware that even though it has been in the plans for ages some senior and middle management continued to argue that integrated fares weren’t needed – probably because they don’t use PT themselves. If AT put some sustained effort in to improving the quirks of the system there probably wouldn’t be much if a debate about whether Wellington skills take it.

        1. You know, I actually like the fact that HOP is only for PT. It’s simple and makes sense. Why go and compete in other payment areas with the multitude of other service providers out there? There’s no benefit to customers because we already need to carry means to pay for “other stuff”. We don’t need an EFTPOS alternative. It would only be of benefit if HOP could be used for everything, but that’s never going to happen.

          I suppose maybe street parking could be an exception. It’s another AT service and it’s a PITA to pay for otherwise, either having to have change or paying a fee to use a credit / debit card which probably won’t work.

          1. At the end of the day, when your phone is nearly flat as mine invariably is, a separate card is a nice standby.

  8. Name one thing Hop does that Snapper doesn’t already do better? The only barrier to Snapper being used as a full integrated ticketing system for the Wellington metro is barriers for railway stations, which the regional council seems to have just ignored in their update to Upper Hutt train station. Snapper is used successfully day in and day out, and has for many years. Topping up online actually makes your money available within minutes, there are fewer quirks with Snapper (a pass works even if your credit is otherwise negative), it works with many taxis and parking already, and—this is my biggest gripe when visiting Auckland suburbs—-almost every dairy, and hundreds of other retailers, have Snapper machines for topping up and checking balances. Combined with the heavily entrenched nature of Snapper with passengers, the fact that it is privately owned and Wellington-based, and you have a superior system that the NZTA is simply attempting to crowd-out with an argument about how it should be one system for many cities. People don’t change cities that often, so this is not a good argument. Is it instead a better use of resources to explore how the different card readers can be made interoperable so it doesn’t matter if you use Snapper or Hop?

    1. I can’t use Snapper in Auckland. But if HOP was in Wellington….then it would be awesome.

      But seriously….why have infratil / NZ Bus making money in interest on the millions in un-used fare money…..and everyone else not?

      This was one of the reasons why other bus operators in Auckland didn’t want Snapper…..it put cash into the pocket of a competitor and they had to pay for it. A neutral, publicly-operated payment system makes MUCH more sense….not least because the public then benefits from the interest on funds held.

        1. I don’t get religious about what solution is best. But in this case, the public solution is arguably (you haven’t responded to my points) the best one. I wouldn’t rule it out just because I spit when I say “government”.

      1. It was more than cash that was able to put into NZ Buses hands with Snapper – the operators patronage data came with that as well.

        Which to a bus company in a competitive route tendering situation is almost as valuable.

        So it was untenable for the other operators to contemplate that there would be 100% privacy of their patronage data from their competitor, hence why it never got a look in.
        And if the boot was on the other foot and NZ Bus had to use a competitors product on their buses they’d behave exactly the same…
        … except in NZ Buses case, past behaviour shows they’d probably behave a lot worse.

        Now if Snapper had handed their system and control of it 100% to NZTA (or NZTTL) from the get go, Snapper might have stood a chance.
        That opportunity has passed, and in the words of Joyce when he was Transport Minister “They’re [now] off the run”

    2. Things that Hop does better than Snapper:
      *auto top up (not available on Snapper)
      *top up online (Snapper needs a $25 feeder)
      *SuperGold Card can’t be loaded on to Snapper.

      And a lot of people travel between our cities!

      1. I’d also add one of the most important. It tags faster allowing faster boarding.

        Not sure if you’d call it a feature or not (depends on view) but HOP not being a cashless payment method means you don’t need multi trip ticket options so kids don’t go spend all of their bus money at the money dairy

        1. Matt’s second point above was important for me when my kids were younger; keen to fund their travel but not their sugar cravings. There’s safety and freedom for them always having a topped up HOP card, not so if it can all be blown on junk food, and for all their mates!

        2. Snapper has withdrawn the ability to use it for purchases other than fares and parking, so sugar (or other) cravings are no longer indulgeable.

          1. Soo, given that restriction, Snapper fanboys – how is Snapper now superior to HOP? Given that HOP can do that and more.

            All it needs is for AT to get regulatory approval from the Reserve Bank to allow such non-PT usage e.g. for Parking. And once thats done, Snapper is really no better for all its pretentions otherwise.

          2. The one area in which Snapper is infinitely superior to Hop is that it is free: no public money has been used in its development or operation. That’s something that Hop, for all its virtues, will never ever be able to match. With its benefits (however they compare with Hop’s) and no costs, the BCR of the Snapper system is infinity – absolutely unbeatable!

          3. Myth: Snapper was “free” to taxpayer.

            Reality: Snapper had considerably slower read-times than HOP, and would therefore add costs to bus running times and operating costs, which would ultimately be passed onto taxpayer. The difference in read times was the primary reason why the benefit cost analysis favoured HOP over Snappper. Also, there were other hidden costs to Snapper, Most notably the data was not publicly owned, which created the potential for issues around analysis and planning.


          4. The introduction of Snapper in Wellington had no effect on fares (demonstrated by the fact that the same fare structure applied to all operators both before and after Snapper introduction, whether using Snapper or not), or on contract prices (the contracts did not change when Snapper was introduced). I think that the same applied to AT Hop, but with the addition of the $m of publicly funded development costs.

            So the public purse and the travelling public have paid nothing for Snapper’s reductions in Wellington running times and operating costs (but perhaps not reduced as much as they would have been with Hop, but that was purely hypothetical at the time Snapper was introduced since Hop was unthought of), while the public purse paid $m for Hop.

            And there was no change with data provided to non-Snapper operators etc: none before, none after, so no cost (but no benefit, either)

            Myth certainly *not* busted!

          5. Also as far as I can tell from reading between the lines if that Snapper want free. When Joyce got the NZTA to review the ARTA decision – a move that ultimately resulted in the NZTA deciding to use the system nationally – they note that they assessed the system based on what what in the tender and indicated that Snapper made their offer free after loosing the tender. They then went public and to the minister saying ARTA were turning turning down a free system for a costly one

          6. There is more to a product than “price” and “free” doesn’t make it infinitely (or finitely) better than a not-free product just because its free.

            In Snappers case, the rate-payers of Auckland can attest to the delays, arguing, tantrums, special pleading, legal threats and everything else that Snapper pulled in Auckland to delay HOP once they’d lost the tender as evidence that just because Snapper was [supposedly] “free” that it wasn’t worth the bus ticket paper the price was written on.

            It may well be that NZTA paid a lot of money for HOP, and that AT paid a lot more to make it integrate and work, but all that money pales into insignificance given the amount of money NZTA routinely pours into investigating and building roads of Dubious significance. And/.or the sheer amount of subsidy that NZTA pay to NZ Bus and other PT operators each year to run what is by international standards a very mediocre bus system in many of NZs metropolitan centres.

            If you want to argue about value to the tax payer, go pick a different topic than Snapper, cos you won’t win that argument.

            And in any case if it was *that* good, and so cheap [as in “free”] – how come no other council in NZ has actually used Snapper for their system to date?
            Could it be that its not as good a value to anyone else but Infratil as you suggest Mike?
            Or is it a conspiracy?

          7. Agreed that free is not necessarily good, but it is a fact that no public money has been put into Snapper, and that it works pretty well from a customer prespective in Wellington (better than Hop in some ways, eg I have to take my Hop card out of my wallet, but not my Snapper), but not being accepted by all operators is a major drawback.

            As to why it hasn’t been adopted more widely, I would expect that its ownership, ownership of the data and its different technical spec, coupled with a national system lurking in the wings sponsored by a body that has already rejected it, would make it pretty hard to sell. No conspiracy theories required! And despite that, one other council does use it – Wellington City Council for the cable car.

            To my mind the fundamental point is that all NZ electronic ticketing systems should be interoperable, and given NZTA’s investment so far and its preferences the answer is highly likely to be Hop. It would be better with mobile phone and contactless capability, and let’s hope that these are being worked on.

          8. Mike I think Snapper is used by buses in Whangarei too. As for the wallet comment, mine works from within my wallet – although since I got a contactless credit card I have to have HOP in one half and credit card in the other

          9. Your definition of “public money” is so narrow as to be useless. Snapper slows down the buses, and hence costs money to run. That’s why the benefit cost analysis favoured HOP. You can try and spin it as much as you like but it’s simply incorrect to say that Snapper was free.

            It’d be equivalent to arguing that decisions about which vehicle to purchase would only consider capital costs and ignore operating costs when the bus is on the road. And that sounds to me like a load of financial bollocks. You’re splitting hairs between what is a cost and what is not and it’s simply hocus pocus stuff.

            Snapper was not free: It’s slow speed would have been very expensive to taxpayers. Myth-busted.

          10. Stu, you appear to be confusing ticketing costs (eg Snapper) with overall operating costs (eg NZBus, a separate subsidiary of Infratil). Whatever changes there may have been for the costs of the operating companies is of little relevence when talking about public financial support for Snapper (which, as we know, is zero).

            But your thesis also fails with respect to bus operating costs: Snapper was faster than anything before (and, until recently, since), and fares paid fell because of the increased availability of discounted fares (zone-specific 10-trips no longer required) – a win-win.

            I don’t know what your definition of public money is, but mine is what the government and councils spend, and that did not increase by a single cent as the result of the introduction of Snapper (unlike Hop), neither to Snapper nor NZ Bus. I’d be very happy to be proved wrong on this, but as yet you have provided no evidence.

            And if, in your view, Snapper represents such a cost to the public purse because of its slowness, what about Tranzit’s and Mana/Newlands’ cards, and Tranz Metro – Snapper is close to perfection in comparison!

          11. Let’s re-wind a little:
            – Your initial comment was that Snapper must have a high (“infinite”) benefit cost ratio because it was “free”. Regardless of whether or not this was true (as Matt suggested in another comment we don’t actually know what Snapper tendered to AT) I pointed out that your definition of “free” was narrow to the point of being useless.
            – Benefit cost analysis attempts to include all relevant financial/economic costs. When considering ticketing systems, the speed at which it operates is a highly relevant, if not the single most important, consideration. Why? Because the speed of the ticketing system determines other costs, e.g. dwell times –> patronage/fare revenue –> operating costs –> infrastructure (stops, gated stations etc.
            – These costs are *directly affected* by the choice of ticketing system, and they are also readily quantified. According to NZTA’s analysis, Snapper would impose greater costs on the system compared to HOP. Put another way, the difference in read times appears to have tipped the benefit-cost balance in HOP’s favour, even though the latter had higher capital costs.

            In a nutshell: You’re over-simplifying and/or mis-representing the issue when you suggest that Snapper was “free”, and you’re definitely wrong to conclude that these low capital costs would result in a higher benefit cost ratio. Snapper wasn’t “free” under any commonly-accepted financial/economic definition. And that is what is relevant to BCA.

            Still think that I’m confused?

          12. We’re talking different things. I’m referring to the public money that’s actually been spent in introducing Snapper ($0), with a light-hearted aside about an infinite BCR, which would have been the case if that measure had been used to assess its introduction in Wellington and Auckland initially since there was no external cost and negligible (if any) public disbenefits – no such analysis took place because no public money was involved; you appear to be talking about what would hypothetically have happened if Snapper had been awarded the Auckland contract.

            I’ve no idea what the BCRs for the different Auckland proposals were, but I do know that Snapper is entirely unsupported by public money, and I imagine that that would not have continued to be the case if it had won the Auckland contract.


          13. Stu Donovan, can you be clearer about this speed issue? And how do you control for that?

            A 20 year old uni student who uses tech and a smartcard everyday is obviously an optimal user right?
            A 50+ weekend warrior who mostly drives during the week and is perhaps less confident with tech is also surely going to slow service delivery down no?

            If tagging on is a before / after or even during service thing (all very logical) then tag speed would be significantly redundant wouldn’t it?
            Train systems generally speaking have gates or platform posts or even machines to tag on inside trains, is point of contact speed really that important? Gating systems are usually designed for peak periods too, so often there is large amounts of redundancy / under use.

            Can you quantify the time savings and then control for the real world scenario that users tag at differing speeds regardless of how fast they ‘might’ be able to do the tagging? What is the real difference in those numbers 600 whatever vs 300 whatever?

          14. “Read time” is the time the reader takes to recognise a card is tagging on/off.

            I understand Snapper averages circa 600ms, HOP averages circa 300ms. These are averages, across all different types of passengers.

            Read times are definitely not redundant, there the single most important aspect of a ticketing system. Slower read times basically equates to slower passenger throughput. On buses (onboard fare collection), this means it takes longer for passengers to board and disembark; at rail stations (offboard fare collection) this means that queues at gated stations will be longer.

            Is that clear enough?

        3. Snapper makes a quirky squelchy, fishy sound when you tag it doesn’t it? Not a boring beep. Not important but a difference.

      2. Stu not really, I wanted more real world concrete examples actually but it is ok.

        So can I run a theoretical, at a train station 2000 people arrive at peak over 1 hour and there are 2 readers / gates.

        1000 people per reader, 60mins… your reader/gate on average would need to read at 17 scans per min… or a scan every 3.5 seconds…

        With Snapper the theory would be that you can get 3 people through in 2 seconds (600msec) vs HOP 6 people through in 2 seconds (300msec), is that right?
        c.100 people per minute per reader with Snapper, c.200 people per minute per reader with HOP.

        So that is 6000 people per reader per hour using Snapper, 12000 people per reader per hour using HOP.

        Now those rates seem WAY too fast and quite irrelevant for almost the entire network wouldn’t you agree?
        So the argument of xx msec speed difference is nowhere near as important in the real world IMO.

        “Is THAT clear enough?” You should try being less defensive…

        Auckland doesn’t need 12000 (Snapper) – 24000 (HOP) reads per hour using 2 readers / gates per platform but you make it sound crucial.

        Also buses could easily have readers inside them so that tagging on / off is done during travel (with random inspections / CCTV), not at stops, again making reader speeds redundant and dwell times far more efficient.

        1. “Is THAT clear enough?” You should try being less defensive…”

          I wasn’t being defensive – I didn’t know what information you were looking for, i.e. it was a genuine question. I hope you find the following useful.

          “Stu not really, I wanted more real world concrete examples actually but it is ok.”

          If you want concrete then simply go look at a functioning major (gated) rail station at peak times. Britomart for example has approximately 15 gates. Most major stations I’ve worked at experience queues which form during peak times and sometimes take quite a while to disperse. In Auckland’s case we need to be designing our system to carry 3 times more people than it is currently … I believe Britomart is currently accommodating between 5-10k passenger movements (in and out) per hour in the near future and it has ~15 gates. Post-CRL Aotea will probably handle twice what Britomart does i.e. 10-20k per peak hour. Don’t forget that we need to allow for people entering and exiting the stations, further reducing the number of gates that can be dedicated to handling the peak flow (whichever direction it happens to be). And finally, passenger arrivals are not smooth – but bunched in line with timetables. Hence the peak 15 minute periods can be much higher than the average peak hourly flows. Auckland’s 6 car EMUs could, for example, dump the best part of 500 passengers at one time. If you have 4 arriving in a 5 minute period, then you’re talking about 2,000 passengers in that same time.

          “So that is 6000 people per reader per hour using Snapper, 12000 people per reader per hour using HOP.” Now those rates seem WAY too fast and quite irrelevant for almost the entire network wouldn’t you agree? So the argument of xx msec speed difference is nowhere near as important in the real world IMO.” Yes those rates are way too fast. Actual gate throughput will be much less, probably 1 person per 3 seconds, or 20 per minute = 1,200 per hour. 300ms additional read time will therefore reduce the capacity per gate by approximately 10% to 1,090 passengers per hour.

          For buses, consider your average peak hour bus trip on Mt Eden Rd. Say that along the entire bus trip circa 100 passengers board and alight with HOP, i.e. 200 “reads”. At 300ms versus 600ms you’re talking about an extra minute onto the running time of every 30 minute bus trip, i.e. ~2% of total running time. To put that in context, Auckland’s bus operating costs are ~$400 million per year (approximately half of which is covered by fares). If we assume that the difference in read times causes a 1% increase in total costs then you’re looking at an extra $4 million per annum in operating costs. But slower buses don’t just cost more to operate: They’re less attractive to passengers. The elasticity of demand wrt travel-time is circa 0.6, so we’d also lose 2% x 0.6 = 1.2% of total patronage, or $2.4 million in fare revenue.

          With regards to HOP versus Snapper, don’t forget that AT also get to hold onto the card balance, which generates interest to cover system costs. Say AT get 1 million cards in circulation (NB: This is not an unreasonable target) with an average balance of $20 = $20 million. Lets’ assume annual interest earned on is 5% = $1 million p.a.

          So all up on the HOP balance of the ledger we have:
          – Bus opex costs = $6.4 million
          – Annual interest = $1 million
          – Infrastructure impacts = hard to quantify, but likely to be significant. Note that the extra bus travel-time calculated above is all incurred at bus stops. In Auckland many stops are already capacity constrained at peak times, hence AT’s desire to adopt capacity vehicles like double-deckers. In this context, even a 5% reduction in average dwell time per passenger can help avoid the need for expensive infrastructure, and at the very least help to avoid bus congestion.
          – Data ownership and risk = hard to quantify, but very important to AT and NZTA for planning purposes. Personally, I don’t think many businesses which would out-source their ticketing system to the degree Snapper was proposing. If I governed a transport business, then I certainly wouldn’t have felt comfortable with the amount of risk inherent in the Snapper model. Much more straightforward for AT to own the data from the outset.

          Now we know value of HOP tender was approximately $100 million. We don’t know what Snapper tendered, but let’s say 50% less, or $50 million. We previously quantified annual savings of circa $7.4 million for HOP. So the read-time issue is really important in the context of the sorts of numbers we’re talking about. However, NZTA’s decision to award the contract to Thales was heavily scrutinized, so I think it’s fair to say that on the balance of factors it offered the better deal. At least on paper :).

          Does that help? Genuine question :).

          1. If is more helpful and reasoned.

            But a couple of things, I don’t believe the msec read matters as much as you say, i.e. I believe that the real world use of a smartcard at a reader / gate, the physical problem, means we as humans use it sub optimally and in disjointed file.
            For example I showed your 300msec HOP reader would mean 6 people EVERY 2 seconds, that just isn’t possible – technologically, yes… physically, no.
            Can you see where I am coming from and how I think it might negate your quantified benefits?

            I use a mid range Metro system by world standards daily, far larger than your Australian examples, it has 2m riders per day – 700m trips per year (not counting, bus and train) and I see ubiquitous smartcard use, the msec speeds you promote are irrelevant under a certain number.

            As for Britomart… sure if you want to triple the patronage and do nothing about the physical infrastructure then don’t blame the smartcard?? That’s crazy… design a better station for flow and add some gates / readers… but I think you know that.

            Again like I said for buses, look around the world – things can be improved. i.e. you enter from the middle door of the bus, you scan readers before you sit down / stand while the bus has left the stop and you scan out the front by the driver before and while the bus comes to a stop.
            I think problems of efficiency are getting conflated here.

          2. John I’ve edited my answer to clarify a few issues, such as impact of read times on gate throughput.

            Note that this is not about “blaming” the smart card, but instead analysing the impacts of read-times on the system we have, or will have in the near future. If we choose a slower smart card, then we *will* have to spend more on operating services and gating stations – and that has a cost. If AT chose Snapper, then they would lose $1 million in annual interest on the float etc, so that needs to be considered too.

            Note that in the above costings, I have only included $$$ for the impact of read times on operating costs and fare revenue, plus interest earned on card balances. I *haven’t* included any costs for infrastructure costs. Gating a station can cost about $5 million on average, so a ~10% saving = $0.5 million. Let’s say AT want to gate 10-20 stations over the next 20 years, then you’re talking about $5-10 million in savings from that – and that’s not counting savings for buses.

            General message: Small % efficiency gains can equate to large $$$ in larger PT systems.

          3. Something else that busy Metro stations do is use their double door / wide door accessibility gate as a peak time express lane.

            Generally at peak the accessibility gate will be kept open and have two poles either side with 2-4 card readers on them at different heights between 1200-1600mm, staff members usually oversee the flow of people.
            This way you have your normal x amount of gates with 1 or 2 for counter flow plus another 4-8 peak time card readers. and the poles can be turned around to face the other way in the counter peak… a simple, cheap and effective way of giving people more access to readers, the really important part, not the individual read speed.
            I’m assuming this could also be done / trialled at Britomart if gate speed really is an issue.

            Remember, 10000 people (Britomart), 10 gates / readers, 1000 people per gate per hour, 17 transactions per min, 3.5 seconds per transaction… the 300 v 600msec speed is irrelevant – access to readers is the issue / station design flow.

          4. I don’t think $7 million p.a. is irrelevant. And that doesn’t include infrastructure costs …

            10% difference in gate throughput is significant. Imagine 10% more gates, +10% more queuing time, +10% more queuing space.

            All very well to say “we need more readers” but each gate can cost $200,000 and require 3 sqm. And also remember that the CRL has a couple of underground stations –> space is constrained and/or expensive to get more of.

          5. I would assume Auckland’s future LRT will have onboard scanning, won’t they? At the very least it should be an option for a 450pax tram so that you can tag at your station stop but also on the moving LRT.
            Scanners inside and during bus movements could be an educational precursor..?

          6. Reposting:

            Something else that busy Metro stations do is use their double door / wide door accessibility gate as a peak time express lane.

            Generally at peak the accessibility gate will be kept open and have two poles either side with 2-4 card readers on them at different heights between 1200-1600mm, staff members usually oversee the flow of people.
            This way you have your normal x amount of gates with 1 or 2 for counter flow plus another 4-8 peak time card readers. and the poles can be turned around to face the other way in the counter peak… a simple, cheap and effective way of giving people more access to readers

    3. From time as a consultant I worked regularly in three cities (Auckland, Wellington & Sydney) and also carry the card for Melbourne which I visit occasionally, so the thought that people don’t move cities regularly is rubbish, you may not, but that’s a different deal.

      I don’t really care which system is picked for the country, but expect that during the functional design stage, all of the functions required designed for and then implemented.

      The debacle of electronic ticketing in Auckland and the inability to simplify structures and processes in line with contracting renewals doesn’t seem to make much sense, but fits with my impression of amateur solutions that PT gives off, although is growing out of as usage increases.

  9. I would highlight a very good related post by Steve Riddle over on Geekzone called “Snapper+GWRC+NZTA. Why integrated ticketing isn’t as simple as it should be.” that rightly concludes:

    With all of the talent we have in New Zealand, how did we end up with the NZTA spending $100 million to build a solution based on French technology that doesn’t fully meet the needs of public transport users? … Is it time to for Wellington to simply tell the NZTA were to stick HOP and build our own open standards solution for public transport ticketing?

    1. What’s “fully meet the needs” part? I can get on any mode of transport I use and the card handles passes just fine.

    2. “With all of the talent we have in New Zealand, how did we end up with the NZTA spending $100 million to build a solution based on French technology”

      Easy because the local incumbent (Snapper) spent nearly as much, if not more money, over more than a decade, and still couldn’t satisfy the requirements NZTA (note NZTA, not AT or ARC) had for a integrated ticketing system. And its a proprietary system to one bus operator who had no real incentive to make it open and easy to integrate with for all other operators. If it was, it would be everywhere in NZ by now. It ain’t and that tells you something.

      The answer is obvious, we are too parochial to build a NZ wide system from scratch – so we ended up buying one from overseas.

      In any case, the technology risk is huge with such products, they’re not like roads which will still be usable in 25 years time with some maintenance.
      Software products like HOP (and they are basically software underneath) undergo formidable technology rot and need constant and expensive updates as the underlying technology they are built on goes out of fashion/support.

      Bank ATMs have the same problem – in a lot of cases they still run Windows XP – an operating system thats over 14 years old, and created many years before the iPhone existed.

      And whatever the current payment tech (Cards, software etc) we have now (Semble, Apple Pay, Samsung Pay, Google Pay, EFTPOS etc) will all be totally obsolete in another 10 years.
      The rate of change is huge, the market not so much, so the risk is shared over a small pool of users/operators.

      Sound expensive and risky? it is.

    3. It’s a decent piece but there are a few inaccuracies in it that affect the narrative.

      Of course Wellington can build their own system and keen standards, I just don’t see the NZTA helping with the cost

  10. Now just a minute, HOP is not “Auckland’s system”! It was a design forced on Auckland by NZTA (which added considersbly to the project’s cost) so that all the regions would have (in theory) an interoperable transport card with common shared infrastructure.

    1. Yep. NZTA showed the sort of national leadership parochial Kiwis love to hate. (Only half joking).

      No wonder we have so little of it.

      But at least they managed to avoid the howls of horror that would have arisen had we (NZ) actually tried to do it ourselves. It would also have been even more expensive. Then we’d have to sell it worldwide….and it would have to be good.

      1. I wish we had had a go at building our own. We’ve got companies able to develop this kind of work to international standards and at a fraction of the cost. Mainfreight for example, competes very successfully all over the world running entirely on locally designed and maintained logistics software. Very fancy it is too.

        1. Building their own is actually what Snapper have done, based on the Korean model. Home-grown smartcards have often been financial disasters, eg Myki and Opal, and people forget that Snapper’s cost to the public purse was $0.

          But I think we need very good reasons for not going for a national system, and I haven’t seen too many of them about.

          1. Snapper’s cost to the public would have been significant had it been adopted because it was so slow.

          2. Snapper tendered their solution, as did Thales. NZTA put the tenders side by side, weighed up benefits and costs (including capital costs and impacts of read times), and decided that the Thales tender was better.

            Fair and square.

            I don’t see what you’re struggling to understand: Thales won the tender on the grounds that benefit cost analysis showed their solution delivered larger net benefits. If they hadn’t they wouldn’t have won the tender :).

            In the case of Wellington the answer may be different. All NZTA seem to have said is that the Thales system has some back-office efficiencies, given it was used in Auckland. No different from Snapper arguing they have an advantage as the incumbent … onwards!

          3. Stu – don’t disagree, but the fact the Thales won the tender means that the costs of developing and implementing the Snapper proposal were never incurred, so that issue is purely hypothetical.

            What is not hypothetical is that public bodies have paid out a lot of hard cash for the development and implementation of Hop, but not one cent for the development and implementation of Snapper (in Wellington, at least – not sure about the rather bizarre AT Snapper interlude).

          4. I’m still struggling to understand your general line of reasoning. The only reason Snapper’s costs can be described as “hypothetical” is because they didn’t win the AT integrated ticketing tender. From where I’m sitting, that’s not particularly insightful or useful to debates about which technology to roll out in Wellington …

            Anyway, time to move on methinks; I look forward to the tender process getting underway so that Wellington’s PT users have some certainty about which system will be used in the future.

        2. Well Patrick,
          You’ve obviously never tried to integrate software system to Mainfreight IT systems and I can tell you from experience its not a walk in the park to do so.

          So you have to ignore the superficial gloss and stand back and take a good hard look.

          I’d say that Mainfreight does well on the international stage not because they write/use world class IT products.
          But because the people they have, have a “can-do” attitude and go out of their way to deliver (literally and figuratively), despite the limitations.

          Something we don’t see with AT as far as the HOP rollout or leadership round that. So the technology is a small part of the product, the people make the system, not the other way around.

  11. True – just as The “Super City” was forced on Aucklanders by Central Government without allowing the citizens of the separate old boroughs a referendum on the matter.

    The owners of Snapper (NZ listed company Infratil) are also the owners of NZ Bus with which I think the public service unions and some Councillors had had a “fractitious” relationship. SO it was always going to be an uphill battle for Snapper in Auckland to overcome those squabbles and odds. So is it now Wellington’s turn to bow to Central government’s agenda?

    1. Actually the Super city was the result of the previous councils not being able to get their respective acts together. I believe it was a stadium on port land that was the catalyst for change.

      The result, whilst not perfect, is progressing better than many naysayers expected. AT whilst frustrating lots of us are generally doing some great work.

      1. What, central government got the pip because they couldn’t milk ratepayers for a white elephant on prime land? They certainly got their revenge!

    2. Having the ‘super city’ ‘forced’ on Auckland is the best thing gov has done for us this century: it is a vast improvement over those squabbling fractious little empires upsetting our ability to deal with gov (the most important issue for the city).

      1. I agree. The amalgamation of Auckland was a very good thing as an idea.

        But they gutted democracy when they did it. Having only 20 on Council – elected via First Past the Post! – was terrible. It’s meant that the two local body elections so far, more than 50% of the votes cast elect no one at all. Worse, the majority on Council were elected with less than 40% of the vote….and most of those with less than 30% of the vote.

        Even then…once elected….they can’t actually do anything with respect to the CCOs – full of people appointed by Rodney Hide and National.

        But understand that National don’t like democracy (Chch dictatorship, ECAN dumped andcommissioners appointed – still even today, etc…). But they did Auckland a HUGE disservice.

        At some point I want to start a campaign for elections via STV in Auckland.


        1. > full of people appointed by Rodney Hide and National.

          The CCOs were full of people appointed by Rodney Hide when the Super City was founded in 2010, but by now they’ve all been either re-appointed or replaced by council, which makes ongoing appointments to CCO boards. NZTA has a representative on AT’s board, but he’s non-voting.

  12. Cue the appearance of Snapper shills/astroturfers. Was it Mike Alexander and “Bob”? They were obviously (mis)representing Snapper whilst pretending to be ‘concerned citizens”.
    These two guys knew too much to not have inside knowledge of Snapper. It left a sour taste that a company could be so duplicitous.
    And as someone stated above: it is not “Auckland’s system”, it is the NZTA’s.

  13. My gripes with snapper are that I cant use it on newlands buses and that there is a financial penalty for transferring.

    Perhaps being able to buy a snapper at Wellington Airport (and claim a refund upom departure) would help foreigners from Auckland with AT hop cards. I also have a hop card and a go card and a melbourne one too, different card for different city – no biggie.

    1. The financial penalty for transferring is a bus company/regional council decision. It’s nothing to do with Snapper, which handles those few free transfers that do exist, eg at Queensgate, perfectly well.

  14. The one thing integrated ticketing did do was give AT, the agent on behalf of ratepayers subsidies, the opportunity to see exactly how much is being spent on each service, accurately.

    If I recall correctly, prior to AT HOP the ARC and AT had no access to such data from private bus companies owing to idiotic central government regulations preventing such information sharing. They had to accept the information handed over by private operators without objective vetting !

  15. I have a snapper card and a Hop and use both. Being based in Wellington I use Snapper a lot more. I find the AT hop card annoying in that I need a card – I cant just use my cell phone like snapper. I also find the topup on hop annoying in that it takes too long, I cant just touch the card to the phone to topup or check the balance like a can with my old snapper card. I also find it good how the snapper app has instant feedback – you can see your tag on and off and topups instantly. You can also top up someone else’s card from your phone if you need to (handy if friends are caught short) or you need to top-up on your journey using a physical card.

    In short hop is a bureaucratic product- it conforms with documented outputs- not customers needs. I understand it is more robust/larger system than snapper- it just doesn’t make my life easier like my snapper does. Perhaps the answer is for Infratil to set snapper as a standalone company?

    1. I think a lot of the problem is that AT haven’t had people focused on improving the system. If they had most of the things you mention may have been done. They’ve had a hard time just getting the work to enable integrated fares over the line

      1. And isn’t a large chunk of that problem been related to the need to combine this change to integrated zonal fares, with the rollout of the new bus network, which has needed extensive public consultation, retendering of affected bus routes and no doubt new contracts with the operators to boot as well as long lead in times.

        So while the delays may seem to be all HOPs fault, is that the true situation, is the Thales system that hard to modify or is that AT have by their decisions, actions and inactions made it harder than it ought to be?
        So Wellington should not take Aucklands problems with it as all the HOP backends fault?

        1. One of the major challenges to implementing HOP was Auckland’s incredibly complicated fare structure, which took a lot of effort to programme.

          In an ideal work it would have been better to implement integrated fares and then integrated ticketing. This would have greatly reduced the effort required to get HOP up and running.

          However, there were two barriers to implementing integrated fares: 1) the net cost contracts with operators which meant that they kept the fare revenue created a lot of complexity and 2) prior to integrated ticketing AT didn’t have high quality data and hence it was not possible to estimate the costs of integrated fares (either to individual operators and/or government).

          So it was very difficult to plan/budget for the former, because no one knew what the costs would be!

          At least having now moved to integrated ticketing AT can now see how passsengers connect through the system and subsequently design/implement integrated fare system in a way that doesn’t break the bank for either the operators or the government. Also, with PTOM I think the old net cost contracting model is kind of going out the window, such that the revenue goes to AT and is then distributed to operators from there. I think this is a much cleaner set-up for all concerned!

          While it’s not necessarily ideal, issues like technological dependencies, data quality, and budgeting are realities of the world we live in and really do shape what happens on the ground!

          1. “the old net cost contracting model is kind of going out the window, such that the revenue goes to AT and is then distributed to operators from there. I think this is a much cleaner set-up for all concerned!”

            Don’t disagree, but I think the bus operators are having trouble adjusting to the fact that the drivers are now cash collectors for AT not the company that employs them.
            The more people who use HOP cards the better really, as there is less cash to account for by the driver and less of a slow down to other users on the same bus.

          2. I have no idea whether they’re adjusting well or not. I must say that the new metro livery is actually pretty good and great to see it being picked up for mew dds. In terms of hop take-up i completely agree: now the system has bedded in they should aim to eliminaye cash within say 5 years by profressively making hop more attractive. Integrated fares is the next step!

    2. Living in Wellington and using buses regularly, I see very few people using their phones and plenty of people with red cards. That may change in future, but is how it is now.

  16. Why wouldn’t Wellington pick up a system where the “backroom” is already built and working. Most people’s criticisms are about local decisions made by AT on the fare structure, and the poor job they do educating people on how to get the best out of it. Snapper got shown the door in Auckland, and never took the much talked about legal action.

    1. “and never took the much talked about legal action”

      I believe that they talked about legal action to cut the Auditor General’s investigation into the whole sorry saga off. She was unable to carry out an investigation while the threat of a court case was in the air, and by the time it wasn’t her office seems to have moved on.

  17. Very interesting that the only new argument coming out of the Snapper astroturfers now is: “government monopoly”. I.e., stick with Snapper because yay capitalism? I think the ACT Party meeting is in another blog, fellows.

    (We really need a “like” function for comments. BTW.)

  18. I think a lot of he issue with HOP are AT related rather than inherent issues with the Thales system, after all it is very similar if not the same as the national smartcard they rolled out in the Netherlands (OV chipkaart).

    The GWRC tender is a good chance for a really good pitch on what is possible but not yet implemented for HOP e.g Semble/smartphone apps, parking meters etc. It is also a chance for Snapper to show that they can resolve their integration issues with the national clearing house, but they start at a big disadvantage, even though they are existing on NZBus as if they can’t connect NZTA funding is unlikely.

  19. I think that it would be preferable to have a single national system. For one thing visitors, whether based elsewhere in NZ or overseas, would not need to bother doing much more than obtaining whatever it is needed to utilise such a system… this would remove barriers to public transport use, For another a national system would surely at least leave the door open to the possibility of high-speed train connections between cities (e.g. Hamilton and Auckland). Furthermore, a lot of the comments have considered how long it has taken for AT (or whomever) to extend the use of HOP to tolls or parking (which I agree are useful and appropriate goals, still being transport… libraries and swimming pools? not so much). Some others have proposed that this is because AT has had to focus on other, more pressing, issues (particularly of co-ordination). A corollary of that is if you had somewhere not faced by those same issues they may well be able to dedicate time to those problems… at the very least providing a template.

    Now, I think the question becomes should HOP be that system? Well, personally (being a train user), I have never really had too much cause to be bothered by HOP (even when accidentally topping up a card some fiend had forgotten to take out of the machine rather than my own due to poor awareness on my part: AT transferred the top up to my card… and I had used cash) and, as a bus user, I have to say the speed factor is very important (also, relevant to barely catching trains/just missing them). Additionally, Auckland does have a substantially larger population so to miss out on the benefits of a national system, something would have to be truly awesome (and Snapper does not appear to be this*) to bother with the costs of replacing everything already done in Auckland. I think HOP needs to be given the benefit of the doubt here… the onus is to prove that something else would be better.

    tl;dr — it seems to me that it is perfectly reasonable for HOP to be introduced in Wellington (start with the trains and then gain ground slowly).

    *Let’s see, we’ve had data concerns, the fact it is in the control of private hands, it’s slower… but maybe it can be topped up more conveniently, auto-top up aside… so that would be a definitely not truly awesome, not even just awesome really.

    1. “so that would be a definitely not truly awesome, not even just awesome really.”

      Snapper is only one step “just usable”.

      Not exactly awe inspring is it?

      All up its a “me-too” system from a “me first” company.

    2. I find it bizarre that you can justify that because a system is in Auckland, it should therefore be introduced to Wellington – but that you disagree that a system working fine in Wellington should not be introduced to Auckland. RJ has it right when he/she says: “make it interchangeable where if a person from Wellington visited Auckland they can use their card here and same for the other way around.” That’s really quite simple.

      1. As Stu said earlier: about the open tender process for HOP in Auckland:

        Snapper tendered their solution, as did Thales. NZTA put the tenders side by side, weighed up benefits and costs (including capital costs and impacts of read times), and decided that the Thales tender was better.

        Fair and square.

        I don’t see what you’re struggling to understand: Thales won the tender on the grounds that benefit cost analysis showed their solution delivered larger net benefits. If they hadn’t they wouldn’t have won the tender :).

        In the case of Wellington the answer may be different. All NZTA seem to have said is that the Thales system has some back-office efficiencies, given it was used in Auckland. No different from Snapper arguing they have an advantage as the incumbent …

        So, GWRC has to effectively start from a position as far as NZTA is concerned (who after all pay a fair subsidy to GWRC every year to run those PT services, so have a right and legitimate say in what integrated ticketing system GWRC uses/mandates).
        That HOP should be the starting position as it *is* the NZ wide standard system for integrated ticketing. Whether Snapper or GWRC like it or not.

        The general arguments against Snapper is that it has three fundamental issues:

        1. Its owned by the same parent company that owns NZ Bus, therefore the perception of the other operators is that NZ bus will have a competitive advantage over them if they use snapper because NZ Bus will somehow get their hands on patronage data.
        This same concern ruined their chances in Auckland as well. and the pool of funds on the cards remains with Snapper, not with an independant 3rd party organisation, and that money earns interest for Snappers benefit and no one elses.
        [Yes you can argue that that offsets the cost of the system, but its the end users who subsidise that with their floats, and if it were my money I’d prefer it be handled like the money it is, by a responsible government/local government agency].

        2. NZTA already did a bake off contest and selected HOP not once – but twice! And both times Snapper was compared on a like for like basis with Thales and found wanting.

        3. If it was so much better than the opposition including HOP how come its not NZ wide already? Given its had the thick end of 10 years to do so, it hasn’t.got further than some of Wellington and Whangarei?
        Yet every major city in NZ has a bus system that has multiple operators that a integrated ticketing solution is needed for. The market in NZ alone for snapper (outside Auckland) should be reasonably huge.
        But basically the market has voted, with their ticketing systems, to go elsewhere.

        And there is the 4th [usually unstated] reason, that Snapper was so hell bent on forcing their system on Auckland, that they resorted to special pleading to the Minister of Transport, not once, at least twice, and threatened legal action several times as well when things didn’t go their way.
        Relying on the influence of the Minister of Transport rather than the merits of the system to try and get people to buy it is not a recipe for success as history has shown.

        And unless GWRC pick HOP – the chance of using the same card for PT use in Auckland and Wellington in the near future is absolutely zero. Because HOP is in Auckland and is here to stay.

      2. RJ does have it right but it’s not simple, because Hop cards won’t work with the Snapper system nor vice versa – so something’s got to give. Given the level of development etc of Hop, Snapper will have just a little bit of a struggle to become the system of choice.

      3. That is an extraordinarily disingenuous remark. Firstly, Snapper is not so widely and rigidly in place that a visitor to Wellington will find its presence immediately noticeable if they, say, use the trains. This is not the case in Auckland. In other words, that is why I mentioned “start with trains”.

        Secondly, I very clearly linked the difference to populations. If Wellington had a system, even one belonging to one of its operators, and was more than twice Auckland’s size it’d be ridiculous to suggest that what the smaller of the two has should be given a leg up. As it is, Auckland has a substantially larger population and a more ubiquitous system.

        Thirdly, there are several things about Snapper, mentioned in that post, which mean that one should be cautious about suggesting that it is a superior system. The two most important relate to data (not sure we can trust private companies, esp. one in competition with other providers, to have all the passenger data) and to tag on speed (which is particularly important for buses: which for Auckland is a very big thing).So, in other words, it’s pretty clearly laid out why it’s a case of one can reasonably adopt Auckland’s in Wellington but not Wellington’s in Auckland. If that statement seems bizarre to you that has nothing to do with the reasoning underlying that statement.

  20. A key part of a PT smartcard moving into the future (well NZ’s future, other places now) is the ability to use it in conjunction with ‘bike share.’

    Having all train stations, key bus stations and all major destinations accessible with a bike share system is where PT and PT transactions need to be.

    As an example, it you have train station X and it has a bike share system accessible only using HOP / Snapper you have done two things.
    1. with 3 to 4 smaller ‘satellite’ bike stations around station X you have increased the catchment from 400m – 5min walk to 1200m+ 5-10min walk / bike. 2. PT users are incentivised further to have a PT smartcard as first and last leg journeys using bike share are far faster and more convenient.

    This also puts biking into the mainstream, i.e. you don’t need a bike to be a city cyclist. The more people biking around their local neighbourhoods / central city easier to progress further cycleway developments.

    Things that may be needed to make it work (as it does elsewhere in the world already), helmet law repealed, smartcard integration and perhaps a higher alcohol limit for riders (optional).

    1. There is no specific alcohol limit for bicycle riders already. At some point, I’m sure the police and courts would decide that you were riding recklessly, but unlike for motor vehicles there isn’t an amount of breath or blood alcohol that’s defined as a limit.

      1. Isn’t the drink driving law about being (drunk) in charge of a motor vehicle. Last timeI checked bicycles don’t have motors in them.

        And even your bike is an e-bike [and therefore does have a motor in it], if its under 300W power then its still legally a bicycle.

        And yes its probably easier to ping errant cyclists for riding recklessly than trying to ping them for drunk driving.

        1. Yes, the specific limits for blood and breath alcohol apply only to motor vehicles (Land Transport Act 1998 s56). That is, 250mcg/L for an adult. This also applies to drugs, which don’t have a specific defined limit (the prosecution just has to prove you were “impaired” by your drug use). And as you say, e-bikes aren’t counted as motor vehicles, either, although there’s a power limit (300W, IIRC).

          However, in reality alcohol does genuinely impair your ability to operate even fairly simple machinery like a bike, albeit you’d probably have to be way, way drunker than 250mcg/L to have much effect. If you’re so drunk you can’t stand up, it’s doubtful you’d be able to competently ride a bicycle. You can’t be convicted of reckless driving for bicycling, but you can be convicted of, for example, careless driving, or careless driving causing injury. (Yes, legally you “drive” a bicycle). I’d say the odds of you riding carelessly would be a lot closer to 100% than 0% after drinking a whole bottle of vokda.

          And what’s more, if you do get convicted of careless driving on a bicycle, you’d have a mandatory suspension of your driver licence, if you have one (and if you didn’t, you’d be prohibited from getting one) 😉

    2. I’m not living in NZ, I just assumed if you were a cyclist and were on the road you were subject to those conditions.

      Where I am living has a zero tolerance alcohol limit for scooter driving, no alcohol limit and no helmet law for cycling and a large 24hr bike share system accessed through a PT smartcard… oh and cheap taxi fares too to be competitive with these end of journey trips.

      1. New Zealand’s helmet law is, while embarrassing, not really enforced much. There are occasional “blitzes”, but in everyday life cops won’t stop you if they just see you riding without a helmet, otherwise normally.

        What does happen is self-appointed yahoos yelling at you from cars.

  21. Discussion of HOP vs. Snapper misses the point of why GWRC are probably wanting to test the market for a new ticketing system. This is a rapidly developing area of technology and there may be other providers with newer smarter more cost effective solutions which weren’t of offer when Auckland/NZTA last went to tender.

    It would be sad to see NZTA force an older bit of clunk onto Wellington if there was a newer and more cost effective system on offer.

    As for Snapper being free… This gets repeated so often without adequate answer. There was no direct cost to the local authority so free in that regard. However Snapper take a fee for every transaction made (just like eftpos or Visa). Bus companies (excluding NZBus) were not keen on losing a few percent of their fare revenue. This is cost that is ultimately passed on in higher fares and/or contract costs. In looking at Snapper NZTA will have likely concluded that the cost of them funding the up front cost of the Auckland ticketing system was more cost effective than Snapper taking a cut of every ticket transaction for the life of the system. Has it panned out that way? We’ll probably never know.

    1. You seem to be making the asuumption that the introduction of electronic ticketing is a cost that has to be recovered, but I don’t think that’s the way Snapper see it. It seems to me that their view is that it is primarily a benefit in terms of operating efficiency and data, and that it will pay for itself. That’s certainly appears to be the case in Wellington, where fares have not risen (in fact effectively they’ve reduced) and contract costs have not risen as a result of the introduction of Snapper. They were also creating a product that could be marketed internationally, generating its own revenue stream (which has happened, to a limited extent).

      But let’s not forget that Hop is a much more complicated beast than Snapper, with the addition of the rail and ferry networks, so electronic ticketing would not pay for itself in cost reductions, hence the Hop and future Wellington contracts.

      1. Snapper are not a charity. So of course they look to recover the cost of their system and return a profit to their owner. Some percent of every fare collected by NZBus goes to Snapper as a fee for service. Just as every time you use an Eftpos terminal a fee goes to the bank. It would likely be true that contract costs haven’t risen in Wellington as no services have been retenderred since Snapper was introduced. And regardless NZBus needed to replace their ticketing system anyway so if not Snapper then they would have borne the cost of another system regardless.

        Saying fares in Wellington haven’t increased is not strictly correct. NZBus scrapped the popular and cost effective gold monthly pass tickets and got rid of the $5 day tickets and Star Pass around the time Snapper came in and that was a fare increase for specific groups of bus users. Other fares they couldn’t increase in price as they are set by the Council.

      2. Mike also curious what ‘operating efficiency’ you are referring to? The only recent timetable changes made in Wellington that I know of have added time to the schedule. The old plastic gold monthly pass tickets appeared faster than the current tag on tag off system to me. The drivers were surprisingly quick at clipping through the old ten trips as well.

        1. MarkB: some points –
          *those tickets that you refer to were withdrawn or had their price increased some years after the introduction of Snapper, so it’s hard to see a connection;
          *the introduction of Snapper reduced fares in two ways: the multi-trip ticket discount was increased to 25% at the time (it later went back to 20%), and that discount became available for all trips paid for by Snapper, not just those trips for which a 10-trip was held;
          *the most time-consuming transactions are cash fares, and Snapper/Hop have reduced these significantly, reducing dwell time;
          *as you say contracts have not changed, so neither Snapper’s costs (except for initial card purchase and some topup costs) nor its benefits (except for some cheaper fares and faster boarding) have been passed on;
          *as you would expect, many timetables have been adjusted since 2008, and Snapper has been just one of the many changes in that time. Some train services are operating to slower schedules too, but Snapper clearly can’t be a factor there!

  22. One thing that should happend is make it interchangeable where if a person from Wellington visited Auckland they can use their card here and same for the other way around. Look at Japan the card comes in so many different names yet no matter which city you are able to use the card. You just cannot use it for bullet train tickets or long journey buses but apart from those 2 you can use the transport card anywhere including convenience stores (similar to snapper) and vending machines

  23. I really don’t give a damn about the hardware, or the software: its all about the implementation. I’m in Melbourne at the mo, and their Myki system works fine on the trams – whether it is 100 milliseconds or 300 I couldn’t tell and I couldn’t care less – but their system for actually getting hold of a card, and then putting money on a card, is totally retarded. It may make sense for locals, but makes no sense for tourists. They have big machines that say “buy your Myki money here” but you actually have to purchase a card elsewhere. There is also a choice between “buying Myki money” and “topping up your weekly account” or something – completely incomprehensible to an outsider. A system needs to be simple and well thought through, to make sense to the end users. Mike certainly is not tourist-friendly.

    Similar thing to the situation with Hop – it seems to me to be less useful than Snapper, as Snapper is available everywhere and usable just about everywhere. I haven’t got a Hop card because A: I’m not in Auckland often, and B: not actually sure where I should / can buy one. RJ’s comment above is right: “make it interchangeable where if a person from Wellington visited Auckland they can use their card here and same for the other way around”. Now that is the most sensible comment that has been made here all day.

    1. You might care if you were travelling at peak times on a bus and/or train and slow read times held you up. And/or you were NZTA and had to pay higher operating costs for slower buses due to the slower read times.

      1. I honestly notice little difference in a crowd of AT HOP users tagging-on and a crowd of Snapper users tagging-on back in the day, both are still a lot faster than cash.

  24. Hop still has a long way to go. The website and online top up management are flakey and the delay in updates between the bus readers and home base is appalling. I still wouldn’t rely on the card, I always take cash as a backup and find I have to use it in about one in five PT trips. And storing the balance on the card presumably means an NFC phone app is still in the too hard basket.

    1. Snapper both holds the balance on the card and has an NFC phone app, so the two aren’t mutually exclusive (at least with their version of the technology).

  25. Prefered snappers system here, only negative was the fact it was only on NZ Bus buses. But other than that had much more top-up locations, nfc topups, less disabled card dramas and could be used for more than just transit. AT HOP have been promising us NFC tag-ons and instant top-ups via NFC for over 3 years now, still haven’t delivered, snapper have had it for 5-6 years or so and recently have it as part of semble wallet too. I do hope if Wellington chooses to move away from snapper that they have at least a similar or better offering in the tech area, and hopefully that isn’t an empty promise like AT HOP’s supposed NFC capability demoed at launch.

  26. If NZTE and the Jaffas want to impose HOP in Wellington then they should fund the entire changeover (including replacement cards for users), but only when HOP has been upgraded to cater for all the other things that SNAPPER can pay for as well as bus fares.

    1. All that Snapper can be used to pay for nowadays is fares and parking.

      But take the best of Snapper – compatibility with (some) moblles, instant topups, no blocking of cards, (some) free transfers – with the best of Hop – free online and autotopups, SuperGold compatability, no annoying messages at tag on, speed, region-wide intermodal acceptance – and add features like contactless card capability, daily/weekly/monthly capping and nationwide intermodal acceptance, and there’s a real winner.

      “How much?” and “who pays?” will be the killers…

Leave a Reply