There was something in last Thursday’s NZ Herald article about the punctuality and reliability of public transport in Auckland that really stood out like a sore thumb:

But buses stuck to their timetables – or were at least no more than five minutes late – in 99.24 per cent of cases last month and ferries were even more punctual, hitting the mark in 99.73 per cent of cases.

Brian Rudman picked up on these rather ludicrous sounding numbers in his column today:

Auckland Transport’s latest “good news” bus-service statistics read like the electoral results of some tin-pot dictator.

Indeed, they’re so fantastical any self-respecting dictator would have had them scaled down.

The transport overlords claim last month, Auckland’s public bus fleet scored 99.88 per cent for “reliability” and 99.24 per cent for punctuality.

In bus talk, “reliability” means a scheduled bus actually reaching its destination. To score on the punctuality scale, a bus also has to “commence the journey within five minutes of the timetabled start time”.

So AT’s transport number-crunchers are asking us to believe that of the 171,610 scheduled bus trips last month, just 206 failed to reach their destinations – and presumably didn’t start as well – and that only 1304 – 43 a day – failed to start within five minutes of their scheduled start time.

Later in his column, Rudman highlights why such ludicrous numbers occur:

That’s a key word: the statistics are “self-reported”.

Does seem overly trusting of Auckland Council, who hires these various firms at vast public expense to provide a timetabled bus service, to monitor outcomes using some school-yard, “cross your heart and hope to die” method of self-reporting.

The other issue is that, for some bizarre reason, the punctuality of bus services is counted at the beginning of its trip, while the punctuality of rail services are measured when they reach their destination – surely a far more logical measurement. I really don’t care when my bus or train begins its run, I care whether it’s late when it gets to my stop and I care whether it will get me where I’m going on time.

There’s a bit of hope that this laughable way of measuring bus punctuality and reliability is going to change – from the latest Auckland Transport Patronage Report:

 Auckland Transport is in the process of developing an automated tracking and monitoring system to report bus reliability and punctuality and provide enhanced data to improve service delivery. A review of the reliability and punctuality of all bus timetables has also commenced to ensure timetables continuously reflect operating conditions.

What we really should be measuring is how the buses compare to their timetable at all points along the route. This information would actually be really useful in highlighting where we need bus lanes or other bus priority measures. With the HOP card I’m sure all the data is available – let’s start using it to improve the system.

Share this


  1. One thing that’s always annoyed me about reliability and punctuality statistics is that they generally don’t account for varying passenger loads throughout the day. For example, if the 10.30pm train runs 10 minutes late then not too many people are affected. If the 8am train runs late then a huge number of people are affected. It’d be good to get the stats to account for this somehow.

    1. My understanding is that it is kind of reported for the rail network in the form of train delay minutes which also breaks down the cause of the delay. In Feb it appears there were just over 18k delay minutes which if you work it out is equivalent to 300 hours worth of delays which pretty high. Perhaps they need to do weighted average of it by comparing it to patronage which would be more useful to compare it over time (I think I just gave myself a new task for tonight).

  2. Brian Rudman highlights what happens as the “delayed” bus journey ticks over on the “real time board” to become a service that is effectively cancelled. In my experience of Auckland buses, this is particularly prevalent with off-peak services and during the weekend. It’s only when these services begin to run well that Aucklanders will feel better about catching the bus. At present it’s chronic problem and it only seems to be getting worse.

  3. But “self-reporting” has proven to be so effective.

    The banking, real estate and insurance businesses have all proven that.

    Oh wait

  4. Actually all of the time/GPS data for every trip should be published as open data along with the meta data about what route it was on. This way the stats could be fact checked against published timetables which are also published as open data. It should be a requirement of any company wanting to win contracts for public transport. If the data is being collected anyway then publishing this online shouldn’t cost any extra.

    1. That’s an excellent idea. All data that is collected using the ratepayer or taxpayer dollar should be freely avaliable on the internet so that interested citizens can use it to draw their own conclusions instead of being fed COCO puffery.

  5. You can imagine if Brian Rudman criticised Air New Zealand or another passenger service provider then there would be a spirited response in the media the next day defending their reliability. But when it comes to PT I can almost guarantee you’ll hear zip from Auckland Transport or the bus operators. The fact that they won’t respond confirms that Rudman’s “bullshit” call is accurate.

    The “self-reporting” of stats has been that way ever since I can remember, from the ARC to ARTA to Auckland Transport. Why do the directors and management of AT tolerate such utter frigging nonsense!? Most businesses rely on extensive KPI reporting, but AT are just running blind with these sorts of stats.

    They could also report the number of complaints as well, but as far as I can tell complaints go down the “10 day response time” black hole, never to be seen again.

    And again I say the average punter should be incentivised to complain about poor service – at a minimum you should be able to get a credit on your Hop card if your bus verifiably does not arrive. All too hard apparently.

  6. Surely the critical first step is that we actually correctly measure reliability and punctuality. Rudman is quite right to call the current system an absolute joke. Let’s see it properly measured, then apply some SOI targets to improving things. That’ll make Auckland Transport pull finger.

      1. Only had one stop between Avondale and the city, so pretty much express. I had intended to transfer to the 5:30pm 769, but I was actually in time to catch a late running 5:00pm 767. Well sort of, I missed it by seconds at Britomart, so jogged over to the first bus stop in Quay St and flagged him down. He told me off, saying “this isn’t a bus stop, you have to use the one over there (referring to Britomart). Funny, the signs said “Bus Stop” and it was on his route!

  7. What about buses that arrive on time but because they are full just drive straight past? This has happened to me a few times. It is really bad when it happens off peak as you know you will be waiting at least half an hour for the next one

    1. This happens all the time on New North Road. By the time they hit Kingsland at least half are full of people from New Lynn, Mt Albert etc…

  8. Oh yes, the bus services in Auckland are so reliable.
    The 891 to Takapuna turned up 10 minutes earlier today, I know as I was waiting to cross the road to the bus stop when it sped past.
    What is the point in driving a route so quickly that the times on the timetable become irrelevant??
    I had to wait for the 891X 20 mins later so I could get to Smales Farm, this is an Express service to Newmarket and a very popular one at that.
    So of course they chose to use a small bus on this route, by the time it got to my stop it was jam packed, my stop would be less then a 3rd of the route along.
    Bus services in Auckland are a joke, I am lucky that if I get to work a bit late it is no big deal, but it must be hell for those who actually need to get somewhere on time.

  9. My few recent attempts to catch an off-peak bus have been as hopeless as ever. Southbound on New North Road (the 6:46am 210 bus – Midtown To Avondale) is completely random, it may or may not show up anytime between 6.30am and 7.15am. The last time I mentioned it here the consensus seemed to be that since most people were going INTO the city along New North Road this service had a very low priority on the totem pole so would be amongst the first to be cancelled for any operational reason. One assumes a “cancelled” service isn’t included in the reliability statistics. The second was – of all things – the airbus, which gave me the same run around as described by Rudman of counting down, being “due” then simply vanishing from the board with the next bus taking its place. As mentioned above, it seems that PT in Auckland is adequate – just – to moiving rush hours crowds in expected directions. Everything else is a lottery.

  10. I don’t think the 005 reaches even 99.24% reliablity at the start of its route on Customs Street. When I boarded the bus at 6.20 pm last week, the driver did confirm that is was supposed to be the 6.05 pm service.

    Does it make any logical sense that buses competing with the vagaries of Auckland traffic should be 99 point something % punctual, while grade seperated rail operates at much more belevable 65-85% punctuality rate?

  11. The information from the Realtime Info System should be being used to record on-time performance (for train and bus), but it should also be being used to assess the suitability of the timetable. As previously mentioned, this could identify any choke points on a route, and help AT to focus resources for best value-for-money.

    This should guide the planners as to which timetables need adjusting, or where bus priority measures are needed, or which train services need more or less carriages.

    Ultimately, it should also give AT the knowledge to negotiate penalties for slack providers, or to require route or timetable changes.

  12. As a teacher I also see a lot of my students late due to buses. The school I teach at has a large number of students arrive by bus. On a rainy day at least half of the students that arrive via bus are late/skip stops due to being full. If I caught the bus to work I could not guarantee to arrive at work on time every day (or even 80% of the time) without catching a bus that left home at 6.30am (I cycle 90% of the time drive 10%).

  13. Can’t agree more about unreliability. I’m a frustrated ‘almost bus commuter’. Live in Sandringham and work in Onehunga – don’t want to buy another car for commuting but no problem I thought, there are two bus routes running down Mt. Albert Road I can take. (Don’t mind the 15-minute walk at each end). After three days of trying to bus commute, despite arriving ten minutes early each day at the bus stop for one service earlier than I needed (according to the timetable)…

    forget it. Late for work once, only JUST on time twice despite (according to the timetable) having 20 minutes to play with (on top of walk time) at the other end. The buses Just. Never. Came. And FWIW while I timed my boarding stop arrival to get a Metrolink bus (have HOP) despite the long wait it was the other company’s bus that came. So how late were the scheduled services?!?!?

    Now I’m biking to work, which is fine… except that there are no bike lanes on Mt. Albert Road and the four lane sections have no shoulder room at all 🙁 Time to find a back-streets route… but I digress…

Leave a Reply

Your email address will not be published. Required fields are marked *