Doram if the package has a weight on it, I expect that to be the weight of the product, package weight not included. As there is acceptable variation, I expect the manufacturer to overshoot by the variation so that the amount of product will be at least what is written on the package. In my experience that is what most sellers do. If I found a 250g bag containing less than 250g of coffee as happened to
Well, yes. Me too. But with a caveat.
How far do manufacturers need to go to ensure that?
The only way to ensure that is probably going to be to double, even triple-weigh every single package. Why? No weighing scale is 100% accurate, and nor can one that is known to be within maximum permissible error range today be assumed to still be so tomorrow or next month. That opens a Pandora’s Box of issues over what’s perfect versus what’s practical, and what it costs to err too far in the direction of perfect.
How pre-packaged foodstuffs (not just coffee) are packaged will vary from one manufacturer to another, the small ones being largely by hand and the larger ones will automatic dispensers/fillers which will be regularly calibrated to ensure reasonable consistency over time. So how often does that measurement calibration need to be checked? Every year, every day or between every bag?
Smaller sites, like specialty coffee roasters, will be using very different equipment, on a much smaller scale but again, whether the dispensing of beans into bag is done by filling with a calibrated portion by the dispenser or by hand-weighing the bag afterwards, how often are those dispensers or scales checked for calibration?
If they rely on one single machine, it can drift out of calibration. If we demand everything be double-checked, we now double the capital cost of the equipment, and need extra staff standing there doing the second weighing.
Even if we demand that every single bag contains at least the advertised weight, without double checking every bag, and using multiple scales, it’s not possible to ensure 100% compliance. Maybe, given the scales are actually pretty good and don’t generally drift off, maybe we hit 99.999%, or 99.9% but 100% is going to add to cost, and we, the consumers, will end up paying for it. But all sorts of things affect that. You’d be surprised (and probaby disgusted) if you’ve never seen it, but the amount of crap that gets down into the space those supermarket checkout scales sits in is astonishing. It varies from breadcrumbs to flakes of dirt from unwashed potatoes, to spillages from leaking milk or orange juice cartons. Get any of that junk in the machanism (affecting balance between platter and load cell) and the weighing can easily be affected. More than a few times, I got called out to a “broken” scale only to find such dirt, or a few times, coins jammed in the mechanism. Remve said dirt/coins and the scales are fine. For today.
On some (typically older) scales, even lifting them out and putting them back in and doing nothing else, can mess up the calibration. Why? The have to be level, and lifting them out (maybe to clean out dirt or retrieve droped coins) can mess it up. Fortunately, most modern scales are self-levelling, within reason. But they still need to be not resting on a coin on one corner or they’ll wobble and that messes up the weight reading.
To put it in context, when calibrating those scales, if someone comes and ‘watches’, and rests on the workbench even several feet away, even that is likely to cause enough instability to not be able to get a stable reading. HINT - when weighing beans on your coffee scales, do not lean on the kitchin counter-top. It can affect it.
So we end up with a set of rules (and laws, with criminal penalties attached) to ensure standards are high enough, but practical. There is a degree of latitude built-in to the system, even when you are calibrating these scales. Ultimately, those doing the calibrating use a set of weights that are set up by reference to the “standard” kilogram. Even defining that is tricky, not least because “weight” (or mass) is determined by gravity, and that depends on distance and the earth’s surface is not uniform. Measure something in one location on the planet’s surface and you can get a different reading to measuring it somewhere else. And yes, that really does feature in the calibration/measuring process. From a purely UK perspective, it’s not an issue as we’re consistent enough across the whole country. So UK calibration weights, supplied by companies approoved by and monitored by Trading Standards, are carefully measured to be within a very small tolerance of a national reference weight, which in turn is calibrated by reference to that single ‘reference’ kilogram weight which, IIRC, is in a carefully controlled sealed enclosure in a secure site somewhere in/near Paris.
So, when I used to go out to shops (in my case, nearly always supermarkets) and either install or “verify” scales to ensure weighing accuracy, the weights I used were checked periodically by Trading Standards and certified for use up to a given date. After which, they need to be recertified (and/or adjusted, if needed) in order to be legal to use them t check (and certify) that a specific set of scales are operating within the legally defined maximum permissible error (MPE). How much error? Depends on the weight. The maximum error if weighing 50mg of a pharmaceutical is different to the weight difference allowed when weiging a 5Kg bag of potatoes. Do you care if your potatoes are 5mg under? I don’t. Oh, and if weighing a lorry on roadside scales, that 5mg is, ummm, undetectable.
But the principles, like MPE, are the same, and from the same legislation, whether checking pharmaceutical or a lorry.
Which brings me right back to how much error, at a practical level, is allowable?
Back in my accountancy (auditing) days, I remember having the principle of “materiality” explained. It went something like this. If there’s an error in the accounts, how much does it matter? From an auditing perspective, not necessarily that of the company, it was about ensuring the accounts represented a “true and fair view” of the finances, which is not quite the same thing as acccurate, or ‘correct’. If you were using accounts to decide whether to invest, or make a bid, or supply goods and services, or extend a loan or credit, it is not going to matter if some accounts clerk mistyped an invoice that should have been £35.61 as £35.16, thereby losing 45p. It also won’t matter if the mistyped £53.00 as £35.00 ….. BUT, if a programming mistake meant that level of mistake, irrelevant at a transaction level was happening system-wide, and therefore affecting millions of transactions, that would be different.
But it still might not be “material”. That same lecturer point out that the UK firm Vauxhall was part of the massive General Motors group, and that it was such a small part of GM that if someone forgot to include Vauxhall in the accounts at all, GMs accounts would probably still be a “true and fair view” because the difference was too small to cause users of the accounts to make a different decision about investing, offering a credit line etc.
Back to step 1 - the “caveat” about expecting manufacturers to ensure I get at least the advertised bag weight in my 250g (or 227g) bag of coffee. Well yes, in theory, but in practice there’s a limit to how much extra I want to pay for my coffee beans to ensure that if, say, 1% of bags are 5g under, I don’t get one. Suppose I buy a 1Kg bag a month at £30 per bag. That’s £360/year. If I lose out by 5g - 10g on several bags a year, does it matter?
So while in principle I would expect to get at least what I pay for, in practice, all I care about is whether any shortfall is “material” to me, or not.
I might not be extending credit to GM and be interested in Vauxhall but, if spending £360/year on coffee beans and I end up losing out by a few grams here and there, I really don’t give a damn. It isn’t material to me.
One final thought …
When we weigh our beans on or coffee scales, even those of us with expensve Acaia scales or the like, how do we know our scales are correct?
I mean, any decent brand will be verifying them when assembling, in the factory. They might (or might not) be verifying every single unit, but they might be batch-testing. But even that just verifies hat they were accurate when packed, in the factory. There’s a reason those supermarket scales are verified, in-situ, after installation, and checked periodically.
Who here (other than me) even has access to properly certified, calibrated weights by which to check their scale?
NOTE - My day-to-day home weights, a cheap set from Amazon, actually aren’t bad. I checked them against calibrated weights. Exxcept for those little snips of metal that are actually the trim weights, and therefore critical for accuracy. They were (or rather, some were) an utter joke.
