Focus groups are dead, thanks to social media
When you’ve got two billion active Facebook users a month ready and waiting to give you their opinion, the focus group doesn’t stand a chance, argues Meltwater’s David Hickey.
The traditional focus group is dead. What had once provided brands and businesses with a sample view of how the general public thinks has now been diminished to numbers and words on a page.
Take the Arnott’s pizza shapes fiasco of 2016 as a key example. According to Arnott’s Marketing Director, Rowena Ditzell, a focus group of around 11,000 people voted to get rid of the old, and bring in a new recipe and flavour.
Pizza Shapes: Don’t believe the hype (AKA the focus group)
A group of people, each with unique taste buds apparently representing the whole of Australia, helped overhaul a much beloved flavour that the majority of Australian consumers did not want to change in the first place.
Spoken like someone who has no clue about how “focus groups” work, what they do, or what they achieve, and with a complete lack of understanding of the limitations of social media for insights.
Maybe go and learn a bit about research before you write this. Here’s a clue: start with “sampling”.
I was about to say something similar…but you’ve saved me the effort 🙂
You said this before I could SD. I don’t actually think David Hickey knows what a focus group is “a focus group of around 11,000 people”. That’s one hell of a focus group!
Legit SD.
It’s fantastic for Zucks that he has managed to get 2 billion MAUs but what the article fails to cover is the technical limitations using any type of social listening to get meaningful insights out of organic Facebook conversation. Facebook does not allow it. Their Pylon data was hardly actionable and the new audience insights API was not built for qualitative analysis. In a market with a small population and Facebook being the dominant network by a significant margin you could argue if anything focus groups should used more to back up any qualitative insights derived from social media.
Not reacting to thousands of people flooding your owned channels would be akin to not realizing there is a riot outside office. If you need a focus group to tell you there is a riot outside your office, I can tell you where you are going wrong for free!
This is an ad!
Person with interest in promoting a view point, states said view point, with little evidence other than anecdote. If you truly believe the focus group is dead, I fear for you.
Never been invited to a focus group. Maybe I’ll get a go now. ☠
I’m interested to know whether those ridiculing your view may have an invested interest in focus groups. How would you know?
You’re trying to bash research while not even understanding the language you’re using. It’s embarrassing. You discredited your article before the second paragraph is even finished.
It is logistically and financially impossible for Arnott’s to run 11,000 focus groups. Yes, they may have spoken to 11,000 consumers through other methods, but not focus groups.
They’ve spoken to 11,000 of the vocal minority.
Yes social media has changed the landscape, but far from burying focus groups, has actually helped them evolve, to be even more relevant.
The fast interaction in direct, honest language makes it very efficient to gather “real-time” feedback in a group, and organise into actionable insights.
The critical difference being focus groups are not about quantity of sample, but quality. Also privacy, where honesty between client and participants will produce more meaningful results.
A small number of participants profiled by demographics, psychographics and purchasing behaviour, is far more valuable than an unprofiled crowd, where you have no idea of who they are.
And a major decision can be checked in a profiled quant survey based on options emerging from the group, for risk management.
We find people of all ages are more relaxed, and love being part of a smaller group, like with friends, and asked their opinions, rather than being an anonymous piece of cold data.
Which BTW is just one aspect of unpredictable political polling where people are not as willing to go public with their views.
These “such and such is DEAD – long live so and so” articles are so played out and tired. TV didn’t kill radio. There is a role for both social listening and more traditional market research. Social media is often great for discovery of issues and ideas, but when you have a really specific and targeted business question to solve, traditional methods like focus groups and surveys can do an excellent job (when done well – not just a Survey Monkey sent out to your 10 closest friends).
Here we go. Something else is ‘dead’. Another agenda to sell.
Some data tells you what is happening. Other data tells you why it’s happening. Focus groups when managed well are excellent at understanding motivations and unconscious reactions. Everything still has its place dependent on its role & capability of partner.
“… what most businesses and brands have yet to understand is that traditional publishers no longer pull all the strings when it comes to consumer news consumption habits, and its influence on the average Aussie is waning.
… 46% of Australians use social media as their main source of news”
Ugh, where do you think the news that people are seeing on social media is coming from?
Traditional publishers expanding their reach on platforms like social media does not a ‘waning influence’ make.
You should read the Reuters report that you referenced.
Pieces like this defending one’s own industry are often like asking a barber if you need a haircut,.. but actually this is quite a well constructed argument, and impressive riposte to the ‘content marketing is dead’ conclusion-jumping people have made after the King-Content write-off.
I can understand why their agency is doing well.. well done Bobbi (though I’ve never heard of you)
Great piece David. I think these comments above are just people from the old Australian media landscape, who are not getting with the times.
Beyond clueless. Seriously, how many people on social media would have been actively discussing Arnott’s Shapes before they were relaunched? Of course there will be a “noise” AFTER an event and I imagine Coke are experiencing that following the launch of “no sugar”, but if a brand wants to change, grow, expand and develop to tackle market and sales challenges and expects to take cues from Social Media they will often find a wall of silence or, at best, comments from a very unrepresentative bunch of consumers. This whole article was just awful in its simplicity and ignorance.
Back in 1985, before they launched ‘New Coke’, Coca Cola did 200,000+ ‘blind taste tests’ to confirm their new formula was preferred over both the old one and Pepsi.
But taste testing is a specific and narrow form of research, which is focused on the product not the brand. Like Arnott’s, they didn’t ask the more important question, which is: “Do you want us to make major changes to your brand?”
Arguably, if they had done a few group discussions – with 8 or 10 customers per group – and talked about this, both Coke and Arnott’s might have avoided their respective cock-ups.
Although there is a strong case to be made that the change was made to hide a change that was already going to happen. The switch from cane sugar (Coke pre-New Coke) to HFCS (Coca-Cola Classic Post New Coke.)
In which case the tests weren’t important anyway.
As someone who has been doing market research for a long time, there are a number of problems with this article. Some of the following will be repeated from other comments, but I think it is important to recap.
1) A focus group of 11000 is pretty impressive figure, but it’s ridiculous and shows a complete lack of understanding about the term ‘focus group’. Going back to the linked article, the comment is that 11000 people were engaged across both qualitative and quantitative research. ‘Focus groups’ only fall into the qualitative side of that split. The vast majority of that 11000 sample would be engaged in quantitative research.
2) Prior to the change to Shapes (as pointed out above) there would have been little to no discussion about Shapes online. At least, once you’d filtered out all the mention of the word ‘shapes’ that didn’t actually relate to the biscuit.
3) You can’t conduct taste testing on social media.
4) Going back to that linked Shapes article, it mentions that overall sales didn’t take the major hit that the social media backlash indicated. There were ‘small declines’ on some lines, but once they put the new and old flavours side-by-side, sales went up. So the learning from that story could easily be, “You are going to get a lot of outrage online, but it really doesn’t mean much at the sales counter. Oh, and make new flavours a line extension rather than a replacement.”
5) As others have mentioned, the social media discussions might be out there, but good luck in finding relevant ones to your brand / product if there isn’t an outrage-fueled meltdown. Especially across all the different social media channels and ones like Facebook that lock down a lot of monitoring access, or make it hard to determine exactly where in the world the comment was made.
6) “Turnbull and Shorten were closer than the polls showed, and social media networks showed how the 2016 election would pan out” – citation needed for that claim. You linked other articles, so please back this one up. I’d be interested to know which social media network is the most accurate in predicting election outcomes. Also, the 2016 Federal Election polls were accurate in a lot of ways (https://www.crikey.com.au/2016/07/25/how-accurate-were-australian-election-polls/), even if the LNP didn’t believe them. Now, they weren’t perfect, but that’s why we have things like margins of error.
7) If you are going to claim that polling and focus groups only provide information on a ‘select’ part of Australian society, social media comments don’t reflect the views of the majority / general public either. Certain social media channels skew to particular types of people and the actual take-up rate of channels is only a small part of the community as well. I wouldn’t take the 10% or so of people claiming to still have a Google+ presence to be representative of Australia on the whole.
8) The quote from that article about social media as a source of news was, “In Australia, social media has declined as a source of news by six percentage points, with only 46% of Australians using it as a way to access news.” That was ‘access’, not ‘main source’ like you stated. Online as main source included the very non-social network channels of News.com.au and ABC News Online.
9) Yes, social media monitoring provides access to a larger data source than any focus group or poll could provide. But it’s a louder, messier source that may not even be talking about the brand or issue you want to know about. Or even be located in the same country. Or even be human, thanks to the rise of bots.
10) In my experience using social media monitoring – and I haven’t used Meltwater – the biggest issue is that only significant activities or events really make enough noise to pick up worthwhile information from general social media. If you are running your own social media services, full of people who are already interested / signed up to your brand, then more limited creative activity can also be picked up.
But you need the big events – like outrage over Shapes – that happen after the fact (i.e. post-hoc, not predictive) to get usable outcomes. And they certainly don’t predict consumer behaviour, because I can’t check if the people who said they’d never buy Shapes again 1) actually bought Shapes to begin with or 2) didn’t actually buy Shapes again. At least, not through social media monitoring.
TL;DR – Focus groups are fine, thanks.
Amazing how many people fall for the false dichotomy.
What a load of rubbish. Media boffins generally know very little about market research, especially qualitative research. I can’t be bothered writing more in response to this dribble. David, please educate yourself before contributing to rubbish media.
Epic.
Fail.
And fact free.
A focus group of 11,000. Love to see the logistics of running that session.
No research provides all the answers.
Quantitative research (whether via social or other channels) can help describe what people think. Qualitative (esp. groups) can help describe why people think a certain way and how they might influence each other.
IMHO assertions that any methodology – past, present or future – that can deliver greater insight to marketers is “dead” can only be triggered by the naïveté of inexperience or be the product of commercial vested interest.
Articles that claim things are dead are dead!
Yeah, right.