We sent undercover shoppers into 81 supermarkets in 27 locations across Australia in March 2024. The research revealed that Aldi’s basket was about 25% cheaper than baskets at Coles or Woolworths. We also found where people live can make a difference when it comes to the cost of groceries.
You can read the report here:
Household budgets across Australia are being squeezed tighter than ever before and the price of basic grocery items has skyrocketed.
Unsurprisingly, the latest CHOICE Consumer Pulse survey reflects this. We found the cost of food and groceries is a concern for a huge 87% of households.
Here’s how we surveyed the supermarkets:
CHOICE’s supermarket price research will continue every quarter for the next three years, helping people to stay on top of where they can find the cheapest groceries.
Unfortunately the survey isn’t comparing ‘apples and apples’. Choice states:
Our basket of brand products cost on average $69.33 at Coles and $68.58 at Woolworths (with no specials), while a comparable basket of Aldi brand products was just $51.51 .
In the past Choice has compared branded products from Woolworths, Coles and possibly others such as IGA with their equivalent store branded products. The later is then used to compare with Aldi which predominantly sells store branded products. This then compares ‘apples with apples’.
Why has Choice chosen to compare branded products (which are known and have been shown to be more expensive than equivalent stored branded products) in the majors with Aldi’s store branded products when it has in the past it has compared supermarket store branded products with Aldi (a direct comparison)?
I actually find this review far from accurate or independent. As a Choice member it’s looks like an ad for ALDI. Without comparing “like for like” eg store branded products at Coles, Woolworths and IGA, ALDI will always be cheaper. It is a store that only really sees its own brands.
This comparison reflects poorly on Choice as the media pick up the headline that is not true. Either exclude ALDI in your comparison or do a comparison in all stores on store branded products in only.
Is there an opportunity to get the sizes of the 14 products included in the comparison basket.
I live in a remote community and we are interested in comparing the prices at our 1 store to those in the survey
You could write to the CHOICE staff responsible for the Community site at community@choice.com.au and they may be able to give you further details of what was chosen. They may not be able to do so though, one reason that they may have to refuse is that if the list leaked then supermarkets could alter prices on those items for any future mystery purchasing to present a lesser difference.
This would be a reason to release the list. If the list represents a consumer’s shopping basket as indicated by Choice and the supermarkets dropped prices on those items as a result of the list being released, it will be in the interest of all consumers.
The supermarkets would have to drop the prices across all stores as they won’t know where the mystery shoppers will be visiting next. Again, this is in the consumer’s interest.
Choice needs to be fully transparent in relation to how the survey was conducted, why particular products were selected and used for price comparison and why branded products were only chosen for the major supermarkets (rather than also including a basket the cheapest or store branded products for comparison with Aldi). If they stand behind the price survey they have done, such information should be released in the consumers interest.
Edit:
It is worth noting Aldi is using the information for marketing purposes.
They certainly listed the basket contents, not the sizes though. They may be willing to releases the sizes they had sought, though anyone can go and compare if they have all 4 supermarkets near them.
In a number of areas where products have been compared in reviews by CHOICE, often ALDI products have had high rating products and some low rating. This variation can also be seen in “premium” branded products from well recognised brands. Just because it is mostly own brand does not make it inferior, often it is cheaper though.
I read with interest the article about Coles Local and Woolworths Metro stores. I’ve noticed that Metro stores in my area stock substantially fewer items than regular Woolworths. I’ve also noticed the prices are sometimes higher and/or there are fewer specials. It really annoys me that there are so many of these stores popping up everywhere as they are not good value for money and mean that other stores can’t be built in the same area.
Thanks for the comments everyone, I’m happy to be able to provide some answers for the questions raised here and hopefully clear up some of the issues. I will respond directly to the comments in this case to provide some clarification around the products selected, the reasoning for undertaking the survey in this way and some of the historical data that precedes this particular investigation.
Here’s something we mention in the above article about ‘How we survey…’ that I think adds some important context:
In compiling our grocery list we include items regularly purchased by average Australians. The intent is to compare the prices of commonly bought items, rather than creating a basket that’s representative of an average weekly shop.
And for reference, here is the list of items:
Here is the full list of grocery items that we priced
Apples
Beef mince
Bread, white, sliced
Butter
Carrots
Cheese, block
Flour
Milk, full cream
Peas, frozen
Penne pasta
Sugar, white
Tea bags
Tinned tomatoes, diced
Weet-Bix
Thankfully, this is an easy one - the survey doesn’t compare branded supermarket products with store brand Aldi items. So for example, with the branded item Weet Bix, these are sold at Aldi. You can probably gather from the list that there is a mix of item categories, including branded, store brand and some fresh items such as carrots, and ‘like-for-like’ items are chosen under instruction by the mystery shoppers.
A great deal of care is taken to choose items that match in size and quality, and we had some input from supermarket professionals along with our own historical research to inform this process. For example, with the tinned tomato grocery list item, the products used had to contain the same percentage of actual tomato. This level of detail was applied for every list item. In reality, we had more difficulty with comparing IGA than Aldi for this survey due to differences in range, and to ensure the comparison was fair you’ll notice our IGA basket comparison only contains 10 items. While it’s no easy task, if we can’t make it ‘like-for-like’ comparable, we’re not going to include items or conduct the survey in this way.
We’re sorry that the article gave you this impression Scott, that was not our intent. Hopefully my previous answer around ‘like-for-like’ and looking at the grocery list answers in part some of your concern.
Any time a product or brand comes at the top of one of our surveys or reviews, CHOICE is accused of misconduct, running advertising or lacking independence. Generally, I can personally understand why people have become more sceptical towards the media in recent years, and it’s not in and of itself necessarily a bad thing to have people apply caution.
In our case, the reality is that the nature of comparative testing is some brands and products will come out on top. The brands that tend to be doing well, also tend to continue to do well over periods of time. So, if you’re looking at a particular product, whether it’s TVs, smartphones or supermarkets, certain brands are frequently at the top. We’re aware how this can be perceived, but our commitment is to the accuracy of the results. Surely anything other than this would be a concern.
In addition, in the case of Aldi, we regularly call them out for their bad products, such as the Baunh TV. Conversely, we often award Coles and Woolworths products as the best in class, for example Coles Classic Granulated Coffee and Woolworths Hot Cross Buns. In this investigation into specials labels, Aldi is comparable to Coles and Wooles.
Why not release the specific basket items? There are a number of reasons for this, but primarily it is to preserve the survey as a real world pricing review. There are some things you learn working in an organisation that has tested products for over 60 years and that is the lengths that manufacturers will take to manipulate our test practices.
I digress but I remember a case when a fridge was being tested in the labs and it had software installed that essentially allowed the appliance to ‘know’ when it was being tested and alter the readings. We uncovered this of course, but it highlights the lengths we need to go to ensure we’re not the victim of any manipulation, which we feel would be a risk in this case.
Consider that we are running the supermarket pricing review for the next three years, and somewhat ironically considering the concerns, we really do need to make this ‘like for like’ without manipulation so that it can work as intended. Unfortunately @craigoxlade, this is also why we can’t give you the size of the items either, we need to ensure that we can continue this procedure under real world conditions, or if you prefer keep some mystery in the mystery shop. Please let us know how you go with your comparison.
If it was a potential that we could somehow lower national market prices in a meaningful way by revealing these items, then I’m relatively sure they would take this course of action. Unfortunately, we don’t believe this would be the case, and instead there’s a very real possibility that it would reduce the quality and validity of this important work.
As balance to this, we believe that grocery list gives a reasonable indication to the fairness of the items being compared, flour, sugar, mince etc. As we’ve seen in the last 24 hours, plenty of people have used the list items to put together their own baskets, and while it’s not the same as visiting 81 supermarkets under quality-controlled conditions, I’ve seen a lot of posts across our channels with results similar to our findings. Even Coles PR was quoted as saying their version of a comparable basket using those items came in at $60, which still puts Aldi as the cheapest.
If you do want to go to the trouble of doing your own version of a comparison using this list, please let us know how you go.
Time and again, we see store brand products top our reviews under lab or blind taste test conditions.
There’s definitely some more work for us to do here, but we have conducted some testing that indicates Woolies Metro’s are more expensive than regular Woolworths. Coles was the same between the types. Metro type stores are often smaller and so is the range, although Coles claim this is not the case in every example. My understanding is that it’s also the type of items that are available that may be impacting here, but I understand in some urban areas there is little option but to visit a Metro.
Interestingly, there wasn’t much variation between city and regional in our pricing survey.
Here’s the article on local vs metro supermarkets in case others want to check it out:
Finally, it might be of interest to see some of the results we have found over time regarding supermarket pricing and satisfaction survey results. The results are visible back until 2018:
I have to agree with Scott’s comments, especially that the report looks like an ad for Aldi . I do sometimes pop into Aldi for particular items that I can only get there but could never do my full shop as they don’t have many products I need, so have to go into Coles or Woolies anyway. Also find the fact there are mainly self service checkouts, and a long queue and no one to help pack at the one with a staff member, and that they don’t accept Amex and charge to use any other credit card, all adds to my time and costs. I have tried some of the Aldi products recommended by Choice but found that, yes, they are cheaper but often don’t do as good a job as my usual favourite. I also like to feel that I am supporting an Australian company so mainly use Coles. The money stays here!
Thank you for providing some context about the survey. Third party media reports have likely given people some wrong impressions.
I found the survey results to be quite surprising. As someone who doesn’t spend a lot of time comparing prices at different supermarkets, my view was that Aldi prices might be 10 to15% lower on some items with owe savings for others. An overall result of around 25% lower prices on the survey basket is likely to include some items with a very large percentage difference.
I find the rationale for not specifying the exact product/brand names and product sizes to be strange. At a basic level, the withholding of information is unlikely to breed trust. Some more specific comments about that decision are below.
You have reported a Coles statement that their version of the basket has a cost of around $60. Woolworths reported that they felt their basket would cost less than $60. Being more transparent about the product details would help clarify this issue. To not do so has people wondering who is right.
If Coles and Woolworths already have a reasonable idea of the brands and sizes, the withholding of that information from the public achieves nothing.
Choice article content and graphics help identify some of those items.
It is unlikely that the Government is funding Choice to repeatedly survey the same items. As the next survey is likely to include different items, disclosing specifics three months after the event shouldn’t create any risks.
Showing details relating to each product would be more helpful to consumers. Most people would not buy those fourteen items on a regular basis. If someone only buys half of those items, a detailed list will tell them how much, if anything, they could save on the items that they buy. Perhaps the bulk of the savings relate to items they don’t buy.
Providing fuller details would give consumers knowledge about whether the main savings relate to price differences of house or national brand products i.e. the public will be better educated.
There can be a large variance in unit prices when a supermarket sells different sizes of the same product. For example, Coles sells three sizes of Week Bix online, with unit prices from 50 cents to 93 cents per 100 grams. The lowest price is 46% lower than the highest. If someone currently buys the cheapest option at Coles, disclosing the full information will tell them whether buying from Aldi will save them money or cost them more.
In relation to Point 7, I would suggest that the use of unit pricing, rather than same sized packaging, will be more meaningful when future surveys results are disclosed.
Coles and Woolworths can provide comparison baskets of the same goods but as they do not know the sizings they could be picking the cheapest unit (regardless of unit pricing) e.g., a bottle of 600 ml of milk instead of a 1, 2, or even 3 litre bottle to make their basket seem cheaper. Interesting that even with those possible size choices, Coles and Woolworths could not match the ALDI savings by a reasonably different dollar and statistically significant margin of difference. Has anyone questioned why they still can’t manage to get within cooee of the basket of ALDI goods?
If someone believes Coles and Woolworths chose the dearest or dearer products when making up their basket, then perhaps their possible tactics of confusing what are in reality simple and easy to understand choices is working? It would seem and CHOICE has clearly outlined how the choices were made in that like for like was chosen for each basket of goods across 81 Supermarket stores. Again I reiterate, still Coles and Woolworths couldn’t match ALDI prices.
CHOICE would need to provide their report to the Government and I’m fairly sure this will include sizings chosen and reasons. This survey is not done just at CHOICE’s sole whim, it is done at the behest of the Federal Government. What has been explained will have to satisfy or remain unsatisfactory to those who wish greater detail in the public debate, I can imagine Supermarkets of some persuasions who will be slathering to get finer detail so they can further game the system.
Senate is seeing exactly this alleged market bending behaviour being reported and allegedly exposed by those subjected to suspected tactics. Not all the detail of those submissions and evidence have been reported yet, no one however seems to be so concerned about the lack of total exposure there though.
I understand completely the reluctance of CHOICE to change what they are doing, it will be up to the Government to either change the rules and or to disclose the finer detail. I encourage those who wish to know the very exact detail to contact the contractor of the services (Federal Labor Government) to see if they will release the details, as CHOICE will have been constrained by the contract they have to maintain some level of confidentiality. I very much doubt that the Labor Government will be free with that exact detail until they are satisfied that it’s release will be in the public interest. Asking CHOICE to break a possibly contracted confidence would also risk harm to CHOICE’s future activities.
They can claim what they like and they will, as their aim is to discredit and confuse. Unless the method of collecting data and timing is identical there is no reason for the numbers to match. Next Monday the answer will be different again.
If the supermarkets are not happy with basket method of assessing their price level they could easily supply a file of all their prices on a given date for direct comparison so no sampling effects occur. I think that rather improbable.
Assuming the packets in the graphics are in fact the same as tested and not just meeting the modern journalistic ideal that there has to be a picture with the text.
You have missed the point of the survey. It is not a guide to which products to buy but a way to compare in a simplified manner aggregates that are much more detailed where all the detail is not accessible. A standard basket is a statistical construct not a recommendation.
Not relevant to the statistical basket.
Quite the contrary. To be a useful series of measurements over time the content must be as consistent as possible. Why would you say that the government would not want that or that Choice would not insist on it? It is not at all likely that the basket would be different in future.
I have been a member of Choice for many years. I am frankly disappointed with this exercise. It is simplistic and largely irrelevant. We do not need to spend millions of dollars to work out that Aldi is cheaper than Woolworths and Coles. They are all cheaper than IGA, Woolworths Metro and Coles Express.
I am not a fan of any of the supermarket chains. My father taught me sixty years ago how Woolworths operates with its suppliers. Then it was merely a variety store. I doubt that the approach has changed or is unique to Woolworths. Their bullying keeps prices down for consumers while keeping their profits up.
It would be worthwhile educating the public on how these businesses operate. Rarely are specials due to overstocking or to end of line clearances. A promotional program runs on a four-week cycle that drives the specials. You will notice that some items are never on special. Others are on special frequently. Those most likely to be on special are discretionary items such as snacks and confectionery.
Products can move from being in a promotional program to not (never on special). A recent example is Birds Eye frozen vegetables. Before the recent inflation, items in this range were $5.00 per kilo. Birds Eye agreed with Woolworths and Coles to raise the everyday price to $6.50 per kilo but on special at $5.20. If you pay the regular price you are effectively subsiding those who buy when it is on special.
An outsider cannot determine what is the ‘average’ price increase or whether the supplier or the retailer benefits most from this arrangement.
This is a survey conducted by CHOICE but paid for by the Federal Government as they contracted a trusted, responsible, and respected organisation to carry out the survey in their behalf. As a member of CHOICE, I am glad that CHOICE had such high standards that they were chosen to carry out the task for the Government of Australia, rather than any other organisation.
So is your problem with CHOICE for using the public money to carry out a service on behalf of the Government. Or is it really with the Government for spending the money to statistically prove whether there really is a difference between Supermarket chains pricing. I am sure that this pricing difference that has been found will become part of the results used to help legislators make changes to our Laws to help bring what many may see as unethical profit taking against both consumers and suppliers, to an end. I am for one, very happy to see this work done. It is also important that this testing is done longer term to both ensure it isn’t just a one off result, and if legislation is passed, how effective that legislative change has been.
Often CHOICE are asked by members to carry out very similar or the same types of surveys at members cost, but here as members we have had complaints levelled at CHOICE when it comes to using public money to carry out a public interest investigation as a contracted party. The apparent difference seems to me, that this is a much more public and publicised survey than if it had just been a CHOICE in-house one and that it seems to point to a significant difference in some chains pricing behaviours such that some chains may be uncomfortable with the results.
I see your point that you see a difference between the chains. That difference you see, could easily be argued that is just your opinion or may just be an anomaly of your location that you see the price differences. To avoid that argument that could be posed in response to any individual experience, this process needs to be carried out over a large sample of places, it needs to be a standardised sample, and it needs to be done by an organisation that can be trusted to carry out the task in an consistent, efficient, and unbiased way. CHOICE seems to fulfill that need, so that you and I can be satisfied that the project was carried out correctly.
I’ve always assumed it comes down to product selection. Fewer lines, less stock space, more efficient store layout and checkout (staff savings), and ….?
Not sure if Aldi’s many stand alone sites also delivered a lower overhead? Whether Aldi’s wholesale costs per unit are any less (product purchase - supply line - warehousing and distribution etc) or product quality is always equivalent? Choice reviews various Aldi products providing mixed results for quality, performance, value.
Our number one competitor to Woolworths locally is ‘Fresh and Save’ has similar floor space, with more utilitarian shelving. The choices are not as many as Woolies, but far greater than Aldi. There are multiple brand choices in all high volume lines. The package sizes may be just one or two. Products less often purchased tend to be limited to two, or sometimes just the one brand.
There is nothing other than time stopping anyone from making a short list of regular buys, and doing a walk through of the nearest. Our big 3 are Woolies, Aldi, Fresh&Save. All within 500m of each other. Coles is a 20 minute journey down the M1. For our small local IGA no need. Most lines are premium priced.
Woolworths and Coles often are “invited” to open a store in a location, or they are an integral part of the decision making about location choice. I remember very clearly the problems faced by ALDI in getting into centres. This difficulty often caused ALDI to build in other areas that had been ignored by some other chains. Sometimes this has caused significant cost of build increases for ALDI, I am sure they would prefer to be able to be built and operate in a similar “environment” as the two larger (for here) chains we often discuss on this Community.
ALDI have criteria about distances from warehousing that has meant often smaller population centres have not had the opportunity to have an ALDI placed near them. Coles and Woolworths in particular have long established penetration into their markets, which often means that they can and often do have economies of scale that dwarf ALDI’s ability to compete on an equal basis. ALDI and others e.g. Costco operate where they can have enough population to support their operations.
Kaufland pulled out due to the difficulty experienced with getting stores into locations, and it would possibly be seen to seem much of that difficulty arose because of some well know chains repeating what had been done to ALDI. ALDI perseverance paid off for ALDI here, but even now it seems they suffer some level of persecution. We don’t have chains like Lidl here, both for reasons similar to those faced by Kaufland and that we just don’t have enough population for the land area to make their presence profitable enough to be here.
Getting back to the survey, it has initially indicated there is a large variance in cost between a basket of standardised choice of items between the different chains. Will this variation continue, will it be possible to see changes brought in if it still is apparent. It only really matters for the comparison that the “basket” that has been chosen is equivalent or equal across all chains and reflects items often or regularly bought by consumers, to ensure results are comparable and relevant.
It doesn’t matter if one store carries more premium lines than another. Shopper preference for particular brands may mean a shopper pays more just because they want to buy a particular brand rather than a generic but still equivalent product. This is somewhat similar to the name brand medications and the generic medicines, effectively the same thing but often you pay more for the premium brand. Some of this buying habit is due to “welded on” preferences rather than any economical, quality, or efficiency of product, sense.
Why would the surcharge make a big difference to the cost of the goods purchased? I don’t think it would add anything like a significant cost to the basket of goods chosen and the cost difference overall. It is easy to work out the impact and at 0.5% per card transaction, it would have a very small impact on the cost of product. If a basket cost $50 the level of impact would be $0.25 added to the total. For the roughly $51 basket it is still only $0.25 as the $0.255 would not be rounded up usually to $0.26. The difference in cost at even the best basket cost put forth by Coles and the ALDI basket, as an example, was close to $9. A $0.25 surcharge would be largely insignificant in the difference in costs. So to call out the surcharge as having an impact that would possibly significantly alter the outcome is overstating that impact significantly. If the difference between baskets had been say $1, then the surcharge would be much more significant in it’s impact.
Sorry Choice, I have read your responses to others complaining about the pointless comparison this is and your explanations simply do not pass the pub test. You might as well compare name brands at Woolies with home brands at Woolies. People buy name brands because they want to pay the extra for the name they trust.
A real comparison would be home brands all around.
