Australia will be the first developed print market in the world to kill audits – and advertisers didn’t say a word
Magazine audits are all but dead after the withdrawal of two of Australia's big three publishers in the past few days. Mumbrella's Tim Burrowes argues that the industry will miss the transparency of the audit bureau when it's gone.
When the announcement dropped from News Corp yesterday morning, I wasn’t sure whether to feel angry or depressed. But I certainly wasn’t surprised.
Effectively, magazine audit transparency is dead.
Australia is about to become the only major English language speaking market in the world where publisher claims about circulations will go unchecked.
The move by Bauer Media last Thursday, rapidly followed by rival publisher News Corp on Monday, to drop out of the magazine audits feels very much like a calculated plan by publishers to kill the Audited Media Association of Australia, or at the very least its magazine print arm.
The timing of just before Christmas is not a coincidence. It means that there is unlikely to be a meaningful backlash from industry bodies – even if they had it in them – until it is too late.
And the fact that both jumped within two working days of each other means they must have been working on this together.
Instead, all that remains for the industry will be readership surveys, which by their very nature, can never hold magazine publishers to account in the way that making them prove circulation claims would.
The part that makes me angry about all this isn’t the actions of the publishers – it’s the fact that they know they’ll be able to get away with it.
If they anticipated that advertisers and media agencies would object to such an anti-transparency strategy they’d never try it on. For this to go on without an uproar from advertisers, shows just how supine they have become on such issues.
Indeed, I don’t blame the publishers at all.
The data that emerges from the Audited Media Association of Australia every quarter, has for nearly a decade contained more bad news for publishers than good.
Once upon a time, this ABC data was an excellent marketing opportunity. It was a chance for publishers to talk about the performance of their weekly titles every quarter, and their monthlies twice a year. For many years there were more winners than losers.
But when consumers began to stop buying, the ABC numbers became a quarterly reminder to the market that magazines were on the slide.
If I was a publisher, I can see why I’d wondering why I was paying an annual fee to remind the market that my core product was in trouble. So I honestly don’t know I’d do it any differently.
I’m sure it was the same reasoning behind Fairfax Media’s decision to pull out of the audit of its digital subscribers after they started falling.
By contrast, the readership surveys deliver stability and good news – the methodology ensures that.
The not-particularly-knowledgeable marketer or media agency staffer – and unfortunately there are plenty of them – may well draw the wrong conclusion about their campaign based on readership data.
Let’s say you’ve placed an ad in a magazine.
With circulation data, you knew exactly how many people had bought that mag, taken it home and, it’s fair to assume, read it thoroughly as they’ve spent their own money on it.
With readership data, somebody standing in a coffee shop may flick through a title for 30 seconds while they await their latte. Asked if they recall reading the mag in the last month, they’d now count as a reader.
The reason Mumbrella has always covered the audit’s numbers and has never regularly covered the two readership surveys, from the publisher-created EMMA (Enhanced Media Metrics Australia) and Roy Morgan Research, is because we don’t really believe the numbers accurately portray the picture.
We don’t think think they’re made up, but they don’t make sense in how they are presented and used.
The example that made me struggle to believe was Bauer’s now dead Zoo magazine. In its dying throes, it’s apparent average readership was more than 15 people per copy, when you compared the audited circulation numbers with the claimed readership numbers. For a weekly, that just didn’t make sense – certainly not as a headline number.
Take a look at the websites of both Roy Morgan Research and EMMA. You won’t find (unless you do better than me) anything resembling their definition of reader, or the question they ask.
But if it was the number of people over a given month who thought they may have glanced at any copy (and remember, this is based on recall not reality) then you can see how the number can be both entirely accurate and entirely misleading.
But the publishers would now like advertisers and agencies to plan media schedules on this basis.
And yet, as I say, none of this is a surprise. My colleague Miranda Ward predicted as much just six weeks ago.
And of course the AMAA is a long way from perfect. It fell a long way behind on digital over a long period of time.
But it always provided a very good sanity check against the potential vanity metric of readership.
But the argument that the reason for pulling out is because of a desire for a holistic number across all media just doesn’t stack up. If publishers wanted that, they could have mandated and enabled the AMAA to do it. They’re on the board, after all.
Particularly in Bauer’s case where its digital strategy is about its ‘To Love’ verticals, not individual mastheads.
Readership metrics are a useful planning tool for the advertiser – and for the publisher a useful sales tool.
By contrast, audits allow the advertiser to be confident that the publisher isn’t stealing from them by lying about how many copies are circulated.
Instead, Australia’s magazines are now about to go from among the most transparent media to one of the least.
I don’t think it will achieve for Bauer and NewsLife Media its aim, by the way. I agree with commenter Jeremy on yesterday’s story:
Bauer and News are doing magazines huge damage by doing this
An AMAA audit costs next to nothing
Why pull out? They’re trying to go opaque to advertisers
Readership research is notoriously inaccurate
Circulation is the only real metric between print and online
It’s the most real metric in media all up probably. It’s point to point with no bs in between like panels, or in the case of online, a million ways to trick your traffic and engagement
And the print guys are dumping it and exiting for the door that says ‘ blind them with same bullshit that online people do’
Mags managed the right way are a cut-through medium now
Not if don’t measure them though
Terrible decisions
Bauer and News will ultimately suffer as a result
Pac Mags is hopefully laughing
But I guess we should wait and see
Jeremy’s right. There’s a huge opportunity for PacMags to be the point of difference in the market. If you were an advertiser and two of the three buying points couldn’t prove how many copies they print, the third option would be attractive.
Except of course that PacMags is rumoured to be for sale, with Bauer a likely buyer. So I wouldn’t be betting on it.
Instead the withdrawal feels like the magazine industry surrendering. We won’t witness quite as closely the death throes of print magazines. But it’s also going to reduce their visibility to the market in a world where your average 22-year-old media buyer is already out of the magazine purchase habit.
It also feels like the AMAA has lost the magazine circulation battle. And it’s losing ground on its digital audits.
Which just leaves newspaper circulation. Given the common ownership – News Corp owns NewsLifeMedia for instance – how long until newspaper audits go the same way?
If News Corp and Fairfax Media see the market tolerate the ending of magazine audits, the dropping of newspaper audits will be just around the corner too.
Which is shortsighted when they provide reassurance for marketers – who seem to trust the AMAA numbers above most others.
As I say, I’ve never felt completely comfortable with the readership numbers from Roy Morgan or EMMA, which is why we’ve been sparing in how we’ve reported them.
But once readership is the only game in town, Mumbrella will have to reassess, too. If you were on a boat in the 18th Century and most of the maps blew overboard, you’d still be glad to have the one remaining sketch.
I suspect that 2017 will be the year that Mumbrella starts writing about readership. I’m depressed, but not surprised.
December 22 update: PacMags withdrew from the audit too. So that’s that.
No surprises. I wonder how long it’s been since any audit was fair.
User ID not verified.
We didn’t say a word because there are none of us left to care.
User ID not verified.
This is absolutely brilliant news if you’re selling digital media.
User ID not verified.
I wonder if people who smoke buy more magazines, because basically no one else goes into a newsagent these days.
User ID not verified.
Circ numbers have long been dubious and readership numbers are almost pure fiction. I remember one survey where Zoo (god rest its soul) dropped thirty something percent of its circ but claimed to have increased readership. Pure fiction.
I expect no one said anything because no one, aside from print execs trying to remain employed, cares anymore.
User ID not verified.
Why should they continue to substantiate their claims. The digital world doesn’t substantiate their claims…
User ID not verified.
Good piece of journalism. Thank God somebody speaks up.
User ID not verified.
So are we entering a post-truth era now for magazines, or were we already in it?
User ID not verified.
Still can’t believe Bauer bought ACP, unbelievable
User ID not verified.
We live in a “post-truth” world now. Circulation doesn’t matter anymore. Nobody listens to techno.
User ID not verified.
Tim,
You report on radio ratings, TV ratings, digital ratings and OOH. But then question the validity of readership? Can you please explain how they are any better or worse, and why the bias? I don’t know, but it just seems hypocritical and a little bit ‘easy target’…
Full disclosure – I work for a magazine company
User ID not verified.
Hi Confused,
I’ve always seen the diary system of radio as a weakness too. Over the years we’ve extensively reported trials of the watches and other devices which capture actual listening behaviour. That feels like it would be a big step forward for radio transparency. I’m also not a fan of the fact that what gets press released to the media on radio survey day is only share numbers – not numerical reach, which is also measured but not shared.
At least with TV ratings, although it’s panel based, it takes place in real time. (Paging John Grono for science on that..?)
And outdoor, other than the launch of MOVE which we wrote about a lot at the time (about six years ago now I think), we don’t actually regularly report on outdoor audiences.
And yes, we do report on digital ratings, including our own audited digital numbers. They’re real, not survey based.
Cheers,
Tim – Mumbrella
I was devastated about missing out on an internship at Vogue but then I remembered magazines are tanking and now I feel even better. Surely big advertisers will turn away in droves rather than trust claimed circulation figures?
Excellent commentary Tim.
User ID not verified.
Hi Tim,
No system of measurement is perfect and that speaks true for all mentioned here, including AMAA figures.
While they are all flawed metrics, there seems to be an inherent assumption in your words here that ALL market research is inherently incorrect, unless taken in real time. And while I know this is not your intent, as marketers we need to trust that the research is conducted in an authentic manner that gives us a reasonable representation of the population measured.
Cheers
Dave
User ID not verified.
The IPSOS readership survey uses a reader definition of (to paraphrase) “reading of printed newspapers / magazines in the last 12 months – including any inserts – anywhere, any time for at least two minutes”.
So if you flicked through an article in Big Rigs because it was the only printed material in the deli while you waited for your chips to get cooked, congratulations – you are now a Big Rigs reader!
User ID not verified.
The move from diaries to Peoplemeters to diaries for TV highlighted how inadequate the human memory is for accurately recalling what was consumed. The more evidence-based data we have for consumption the better.
The argument presented for moving away from circulation looks and feels like BS. I am very disappointed in this development and will be making future evaluations and recommendations for our clients with this lack of evidence very much front of mind.
It removes a very solid piece of evidence from the case for magazines and that cannot be a good thing for them.
User ID not verified.
Print circulation figures were never rock-solid – especially in b2b categories and free to street (street press) but for paid news-stand I would have always recommended to advertisers. Vale
User ID not verified.
Tim,
I’d like to add a few thoughts before we head off on our Christmas breaks. In no particular order:
The definition of readership is the same the world over. Read or looked into. In some surveys there are minimum reading time requirements. In the Roy Morgan survey, there is no minimum time required.
So yes, flicking at the checkout is just as legitimate as reading intently cover to cover. Remember that what advertisers are buying in the ad space is the ‘opportunity to see’ their advertisement. The person flicking at the checkout is just as likely to be in the advertisers target audience as those sitting down to read with a cup of tea. There are other measures to measure reader’s levels of engagement. Let’s not confuse and be subjective on what passes as readership and the opportunity to see.
Its fascinating that all the other surveys mentioned in your article, with the exception of Roy Morgan Single Source, are all owned or funded by the media they measure. Our survey is media neutral. We have media, agencies and advertisers subscribe to the survey we produce. It’s not owned by any one media. This is what gives advertisers and agencies and our media clients the confidence that data they rely on is produced without fear or favour.
It is this independence that is fiercely regarded and renowned. And why some of the media don’t always like the truth that it produces. So their alternative, then, is to fund their own survey.
I’m curious you put so much trust into the circulation results. The publishers provide these results. They have the ability to print extra copies or not and you have been a fierce critic in the past of all the dumping strategies of bulk copies.
Our data passes the ‘sniff test’. Our data makes sense with sensible relativities for readers per copy. You don’t see the wild numbers of pass-on readers mentioned in your article from our survey. Our survey often produces lower readers per copy – due to a more stringent and rigorous approach.
The other sniff test is how does our data stack up against other known measures. You will see on our website that we verify our data with ‘census’ statistics. Such as the amount in superannuation, the average personal income as recorded by the ABS. We ask our respondents these same questions. How much they earn. How much is in their superannuation. These are sensitive and difficult questions to ask. With the latter, I’d have to open those unopened superannuation statements of mine sitting at the front door ( but that’s another story…). The point is -we ask respondents these questions. But we don’t weight their answers to these known universes. ‘All’ we weight to is age, sex and geography as determined by the ABS – but our estimates of these sensitive measures – come to within a few percentage points of the total amount in superannuation and average weekly income as determined by the ABS. Tell me that is not impressive?!! And I challenge all the other survey providers to benchmark their data this way. So as the story goes, if we get these measures right, you can be confident we are getting the right results for readership and so forth. It comes from the same interview.
I’d suggest you take a closer look at all the surveys mentioned and see what qualifies a respondent to participate in that survey. In ours, you only have to be aged 14 years and over. I’d suggest you check to see whether respondents in the other surveys have had to consume the media being measured in a set period of time to even be qualified to be interviewed. Did certain commercial media have to be consumed ? Are results weighted to the Australian population ? Or are they weighted to a pre-set incidence of that media consumption?
We hold our data up to scrutiny by publishing comparisons constantly on our website. We are very transparent on the questions asked and our sampling and weighting frame. We have always welcomed anyone interested to visit our ‘engine room’ in Melbourne to understand the whole process from questionnaire design to data collection, processing and weighting. That invitation still stands.
In signing off, I wish you and your readers an enjoyable Christmas break.
Regards
Lisa Meunier
General Manager- Media, Roy Morgan Research
Lisa.meunier@roymorgan.com
User ID not verified.
Hi Tim et al.
Lots of issues here to cover.
I think the circulation data ran into trouble when the definition of what constituted a “paid copy” was being gamed last decade. For example, I could pick up a copy of the Sunday paper after a Swans game and that counted as paid because it was in the small print of my membership.
The rules re-write in 2011 broke the various types of circulation out into ‘buckets’. For example, it is perfectly legitimate to circulate copies into the airport lounges – in fact, it probably is one of the more high-value distribution outlets. Are these copies paid? Well yes they are – indirectly. Free to the flyer – but they were paid. Various rules such as ‘half cover price’ were debated – none worked. So the decision was made to identify them all – newsstand, subscription, bundled sales, digital sales, school sales. tertiary sales, airline sales, accommodation & hotel sales, and event sales. Sadly, I saw very little reported on these circulation breakdowns, with most press being on the headline ‘masthead’ figure.
Someone used the phrase ‘sanity check’. In some cases circulation is a sanity check, and in some cases it isn’t. For example, it is VERY handy to monitor Readers Per Copy (RPC) – not for the level of the RPC but for the trend of the RPC. And I notice some commenters were comparing a single circulation period (three months) with a readership period (twelve months) – a common but elementary mistake from people who should know better.
RPCs can vary dramatically. And for very good reasons. The longer the publication cycle, or the higher the subscription component (which is why I like the detailed circulation data to be available) there tends to be less pass-on. Older demos tend to pass-on less. Higher cover prices tend to pass-on less. Lads magazines have a very high pass-on rate … largely with the phrase … “Whoa, have a look at them!” (or similar).
Which brings us to the definition of a reader.
The most common/accepted type of question is along the lines of … “Have you, personally, read or looked into any of the following publications for at least two minutes in the past ….”. This is asked with different time frames (in order to calculate audience accumulation”. So, yes it will pick up some VERY casual readers (especially at the check-out), which we as advertisers wouldn’t necessarily consider a reader. But .. that is the international standard, so that is what we use.
But before you panic, we also ask how long the publication was read for. So, you need to look at the average duration read. That average would include all the ‘two-minuters’. I don’t have either the Morgan or emma data at my fingertips but I know it is WAYYYY higher than 2 minutes per issue for both sources and around the half-hour mark for the majority of titles [Morgan and emma – feel free to hop in here].
Circulation also has another, less seldom used, advantage. Generating readership for very small titles, community newspapers and regional newspapers is extremely hard because you simply can’t get sufficient sample size easily. This is where ‘class modelling’ can be used. In essence you ‘pool’ similar titles to get a usable sample size, and also pool their circulation, from which you can then derive the average RPC for the ‘class’ and apply that back to the individual circulations. With this method, at least you can get an indicative (but often volatile) readership estimate. For this reason alone, I would have kept circulation going.
Regarding Tim’s TV question, measurement is not really ‘real-time’ but near to it (and for very good reasons). The data capture is in real time, which is sent overnight as a ‘bundle’ to NielsenTAM. This is then matched to ‘reference signals’ to determine what was watched. This could be ‘live’ viewing (i.e. exactly as broadcast), ‘as-live’ viewing (e.g. you paused the DVR when the ‘phone rang and started watching again when the call was over), or ‘time-shifted’ viewing (e.g. playing back recorded previously broadcast material) which is currently up-to-seven days, but 28-day catch-up can also be done.
The good reasons why it is not real time, are that (i) the data has to be matched to hundreds of possible sources going back over 28 days, (ii) the viewing has to be validated (e.g. remove ‘viewing’ where you have not seen any user activity on any remote for hours), (iii) that validated raw data must then be weighted back to the population.
There was an instance in South America back in the ’90s when a now defunct research company dallied with ‘live ratings’. One night they had a live talent show on, and one of the contestants mistakenly thought she was an opera singer – but sadly she sometimes sounded more like a screeching cockatoo. The live ratings were fed to the programme director who called the station and had her hooked off mid-song! The following morning they found part of the phone-system went down for a short while which caused the ratings drop.
But back to the circulation discussion.
Overall, circulation by itself is rarely, if ever, used as part of the advertising transaction. It is an input to the decision, but the rubber hits the road around the readership. So, on this basis I do understand the decision to make readership primary as it is audience data, and audience data will be at the heart of a unified and harmonised cross-media measurement system that I can only dream of at the moment.
User ID not verified.
Re the “checkout reader”.
Spend a day at Coles. Count how many people have the time to browse anything in the checkout, and if they do, choose a magazine over their phone.
Yes, I realise this is a pretty literal interpretation of checkout reader.
User ID not verified.
Consumer readership audits may have an official ‘twang’ but an audit does not portray the interests of a human being to help guide advertising strategy.
Advertisers may appear to have yielded to the cessation of the official audit, however Advertisers can access more potent tools in collaboration with Roy Morgan Research.
The exciting emergence of IntelliTech; putting consumer media consumption and buying data at the fingertips of advertisers via a digital delivery system is the future.
Imagine sitting at your desk in a portal; clicking magazine brand engagement, frequency and filtering these readers by product/brand “likes” and future buying intentions (plus more), and viewing where these Australians live via a digital map. This is the future. Advertisers will get better insights into reader volumes & engagement, and brand affinity data. The potency of this intelligence is part of the wonderful ground-breaking benefits of technology.
Let’s give the old fashioned audit a big send off and clink our glasses to the future. Merry Christmas!
User ID not verified.
You call out that someone could flick through a magazine while waiting for their coffee and be counted as a reader. But I never hear anything about with TV when the ad breaks kick in and someone leaves the room to make a tea or go to the bathroom. What about all those billboards out there but everyone on the bus is staring down at their smartphones or tablets?
There is so much negativity about print/mags yet the same issues people have with them can be applied to so many other channels.
At the end of the day it isn’t rocket science, you buy the opportunity to reach eye balls.
User ID not verified.
To be fair, when you’re talking about TV or radio ratings, you’re not counting the actual number of television sets or radios. To an advertiser, the readership is much more important than the physical sales.
User ID not verified.
Cheers Matt and to the future of this incredibly powerful and engaging media!
User ID not verified.
Good for you Nick Chan in taking the lead to move to the future for mags and focus on audience consumption as other media do and advertisers buy.
User ID not verified.
I scratched my head when they bought ACP too. What on earth were they thinking?
User ID not verified.
I can help you with these ones as well Anon.
Of course people leave the room during ad-breaks on the TV. They also flick around to other channels.
I did an analysis several years ago, and overall there was a 5%-8% drop in TOTAL viewers during ad breaks, but audience drops on individual programmes was higher. This indicates that there is more ‘flicking around channels’ happening than people leaving the room for the kitchen than the bathroom. Having said that, the instruction to the people in the TV panel is that they must be in the room and the sound must be audible on the TV.
So, when you hear that ‘people go to the bathroom when the ads are on’ – that is correct. But people do not go to the bathroom in EVERY ad break. Using the 8% figure above, that is roughly one-in-twelve. If there are four ad breaks in a one-hour programme that’s a loo break around every three hours. Even at my advanced age that seems reasonable.
Regarding billboards, you are 100% correct. That is exactly why MOVE was designed to NOT report on ‘passers-by’ – that is an ‘Opportunity-To-See’ (OTS) metric and not a ‘Likelihood-To-See’ (LTS) metric
This was done by using hundreds of hours of eye-tracking data to establish when there is a ‘gaze’ on an outdoor sign. We discovered that there are seven main factors correlated to gaze duration.
Some signs are unmissable to the driver – e.g. the silos near Glebe Island Bridge, and basically any overhead sign such as on a pedestrian bridge. Passengers can be distracted with papers, phones etc. and see less (your scenario) but some passengers like to look out the window and see more. Seat position in the vehicle is also critical.
Overall larger formats have an LTS that can approach 100%, while some of the smaller formats would be more like one-third.
So basically, MOVE is an ‘eyes-on’ measurement.
User ID not verified.
Facts. Data. Science. All very 2015.
User ID not verified.
Thank you Chris – that’s what we need more agency people to do is make it clear to the publishers that this is NOT OK and that there will be repercussions. Perhaps too late however – but brave and needed nonetheless.
User ID not verified.
Irrelevant though.
TV and radio ratings are _relative_ to other TV and radio outlets whereas magazine and newspaper advertisers are interested in the absolute number of viewers. Everyone has a TV.
User ID not verified.