Efficiency and Progress

Like many people I often wonder whether new ways are more efficient and I sometimes doubt they are. 
I'm also aware that automated systems allow much higher volume day to day so that we don't actually notice most of the efficiency gains, only the glaring non-gains.

Take these examples:

• Melbourne public transport: Myki vs paper tickets; automation vs tram conductors
• Petrol stations: self-serve vs full service
• Supermarket check-outs: self-serve vs cashier
• Stockbroking operations: "straight through processing" from trade date to settlement, vs the old days

Tram conductors vs ticket inspectors

When tram conductors were phased out in Melbourne some years ago in favour of automated ticketing, it did not seem to increase efficiency. Tram conducting was a good entry-level job for young people, and welcomed those who had been unemployed, were inexperienced or were covered in piercings. With almost no exceptions that I remember, they appeared to do a good job and some seemed to really like it. It fostered community on the trams, and made the system pretty easy for travellers.

When the tram conductors went, fare evasion seemed to immediately go up, and within a short time it seemed that whatever money was being saved on tram conductors was being spent on ticket inspectors. I would be very surprised if that cost has gone down.

On the other hand, could they really bring back tram conductors today? Would there still be people willing to take those jobs? Working nights and weekends as well? Could we really go back to leather satchels, uniforms, hole punches and paper tickets?

Long time since Melbourne streets have been this free of traffic
(photo: nicksarebi/Flickr Creative Commons)

Myki vs paper tickets

Myki is a pretty good system, from my experience. It really does seem to charge correctly, and charge you the cheapest available fare (though there have been issues with that). It's easy. It beats paper tickets that get bent or soggy or lost or forgotten (or maybe that was just me).

But there's a problem with Myki. You need to top up your ticket before you get to the station, as the lines there are long. There seem to be fewer places to top up a Myki than used to sell paper tickets, though maybe that will change.

Also, 'touching off' is very inefficient. Getting off a tram and touching off while you're being rushed down the stairs is hard to accomplish, and coming off a train it takes a LONG time to get through the turnstiles as everyone touches off in turn. It takes way longer than it used to for people to feed their paper tickets through. That will no doubt improve.

Supermarket check-out: self-serve vs cashier

I've mentioned before that I'm not an early adopter. I stuck to full-service petrol stations for as long as the last one existed in the southeast of this city, and I initially completely ignored supermarket self-service checkouts. 

I started using them when I had a small basket of stuff and the lines at the cashiers were long. They took a little while to get used to but after a very short time (say about, oooh, a year and a half) I began using them regularly. 

I have to say I quite like the short friendly chat with the cashier who scans your stuff, and it will be a shame when that goes (not to mention the lost jobs of course). But as former CBA CEO Ralph Norris said a decade ago when banks were closing branches, "We can't keep maintaining costly branches just because people want a social experience when they go to the bank." (It was something like that - I can't find the exact quote online because it was a long time ago, so I'm paraphrasing and perhaps I'm being unfair. But this remains one of my favourite PR mis-steps and quotable Evil Bank moments, so I hope I have it right).

It took awhile for us to accept pumping our own petrol, but we eventually did. We're accepting self-serve checkouts a lot faster from what I see.

Are they more efficient?

It's an interesting question. The answer must be yes, in terms of cost. Even factoring in people to show you how to use them and people to come over and correct your mistakes or unlock your DVDs, it's still more efficient, because those things are temporary.  Employing three people to help users, who will also perform other duties while they're there, is cheaper than employing six people as cashiers plus the people performing other duties.

On the radio today people were discussing self service, and the consensus was it was all very wasteful and inefficient, because of customers being dishonest, time taken to fix errors, etc.  And of course it would be quite easy to lie and steal when scanning and weighing things. No doubt plenty do.  Interestingly though, the system (at least at this stage) makes no attempt to guard against that.  

To me that says they've factored it in. The system has to be easy to use to keep customer goodwill, at least until they have a critical mass of acceptance. So they might be luring us in making it easy to use until they make it more theft-proof later.  Or, the cost efficiencies are so great they are prepared to wear some loss from petty theft in exchange for the automation.  

And this is the big picture on these types of automation: companies are prepared to pay to get these systems in - even, perhaps, to create short-term inefficiencies - because in the long term, they are needed.  

I don't think they always get their projections right though. I'm sure Melbourne public transport vastly underestimated the cost of ticket inspectors, Myki and Myki upgrades. 

But one thing we don't as customers or users necessarily notice, is the huge jump in volumes that happens during the transformation period while things are being automated.  Can you imagine full-service petrol stations now, with the number of cars that line up at petrol bowsers? Instead of each person doing their own (even factoring in those who dawdle inside and browse the shelves before paying), imagine the same petrol station with one or two attendants pumping everyone's petrol, as there used to be.

It's the same with tram conductors (though I hate to admit it, because I've clung to my tram-conductors-are-cheaper-than-ticket-inspectors belief for so long). Could a single conductor on a tram cope with the number of people passing through now?  As it is, three burly ticket inspectors at each super stop can barely cope with the numbers.

Stockmarket operations - straight through processing 

Now to my world. I work in stockbroking operations (the "back office"). I've worked in this industry for 13 years, as a processor, supervisor, manager and project manager. Long before my time, the settlement of trades was manual. I mean completely manual, as in collecting bits of paper from the trading floor, running between broking houses and the stock exchange, and typing things into computers.

These days when we bitch about a manual trade booking, we mean a trade has fed in without a reference number and we have to manually key it in so it books to the right account. 

But the thing is, those manual trade bookings are actually still a real headache. Because we no longer have the teams of people that used to manage all the manual processes, 2 people can manage the electronic bookings and settlements for quite a large stockbroker. But it only takes a few manual trades to add a lot of work to their day. (And it's not always as simple as typing in a number of course).  There is also increased complexity in systems, because enhancements had been added over the years and more permutations in booking and settlement are possible. This was made possible, of course, by automation.

So is trade settlement more efficient?  Absolutely - no doubt. 

Do we use less paper? No - possibly we use more, what with reporting, compliance and audit requirements.

Are there fewer people working? Yes.

Do those people work fewer hours? On the whole, yes - but work is a lot more condensed than it used to be. These days there are no smoking breaks and sometimes no lunch breaks, and people in general work at full speed throughout their day (and that, of course, is called "productivity").

When I review or compile our statistics each month, I see approximately 50% of trades are "STP" (straight through processing, i.e. fully automated). That figure is too low, of course, but still, 50% is a huge number of trades that just do their thing without any intervention required by us at all.

In a nutshell

Does automation make things more efficient?  System-wide, on the whole yes.  At a personal or experiential level? Probably mostly yes, but as customers or users we don't always feel it.

What do you think? 

Do you miss tram conductors? 

Do you hate, love or grudgingly accept automation?