The Future Place Blog

What is Social Media Research?

At the moment I am writing a 3500 word article for one of the leading market research journals on the current state of play with respect to social media research. As part of the process of writing it, I have set down what I think the term Social Media Research currently encompasses. I would love to hear other people’s views about the term does and does not cover.

In order to try and structure my thoughts, I have started by creating three broad areas, or rather two broad areas and a catchall. My feeling is that these reflect the current commercial realities. These three groups are:

1. Social Media Monitoring, relating to naturally occurring discourses.
2. Purposed Communities, e.g. MROCs, Community Panels, and brand communities such as MyStarbucksIdea and the more than 22 million people who like the Starbucks Facebook page.
3. Other uses of social media for research.

My feeling is that the first two are relatively clearly bounded areas and currently are much more significant in terms of research spend than the third category. This might change, particularly since that perpetual ‘next big thing’, i.e. mobile research, is in this third group, but as far as the second half of 2011 and 2012 are concerned I think social media monitoring and purposed communities are where the big bucks will be.

In terms of the third category, the others uses of social media include:

1. Research into the use of social media, including studying the way that messages and influence flow in social media.
2. The use of social media as sample source for conventional research.
3. Netnography, conducting ethnographical studies in cyberspace.
4. The integration of social media data with conventional research, for example tracking online panel members’ use of social media.
5. Network analysis and other analytics.
6. WE-Research, for example getting people to use smartphones to capture slices of their life and then to share them socially with researchers and other participants.
7. True social media innovations such as gathering insight through gaming, foursquare, and collaborations like social bookmarking and quora.

Beyond these areas there are plenty of projects that utilise social media, but boundaries have to be drawn somewhere if illumination is going to be created.

What are your definitions of what social media research is, and perhaps what it is not?

Should you try and sue Twitter?

At the moment the British press are reporting that one if its small councils, South Tyneside, has been to the US to subpoena Twitter to get the name of somebody in the UK who is saying things about three of their councillors and one of their officers which they say is untrue. The question is whether this is a good idea.

I wrote this comment after reading Steve Chu’s views on this point where his recommendation was “Surely it's better to keep your head down, get on with your vital work in these tough times, and avoid presenting yourselves as a vindictive bully? If you don't like it, just don't read it.”

In general, I think Steve is right. In a short time from now there could be thousands, perhaps hundreds of thousands, of libels out there, most of which will be heard by nobody and which would cost millions to try and combat. It is simply not practical to pursue every online comment or tweet that might be false or damaging.

In the long term the ISPs and the platforms may arrive at a challenge system which restores some sanity (but hopefully not too much control).

However, for the time being I think people and organisations need to consider the impact of a comment. If a newspaper libels somebody, it is worth pursuing them because newspapers are still believed to be true by many people. If a solitary blogger/tweeter says something, then it is often best to ignore it. Only if that something is being re-tweeted, re-posted or passed on is it even worth considering. If someone is not reaching an audience then taking action against them is likely to spread it further. Before South Tyneside took this action most of the UK had never heard of it, now some of their councillors are nationally famous!

One side effect of the story, so far, is to show that South Tyneside are not very understanding in what the public think is a lot of money. When the expense of going to the US to open court proceedings was challenged, their response was that so far it had cost less than £75,000, but in South Tyneside you could buy a house for that, and the wage of a council worker doing a manual job can be as low as £12,000 a year.

Of course, the world is even more complex if what is said is partly or largely true, which might be the case in some of the footballer and banker stories that have been hitting the press recently, but then suing is just plain daft in this day and age.

 

Are you young and talented?

The FringeFactory is new force for change in market research and they will be holding their first event to coincide with ESOMAR’s Congress in Amsterdam, on 19^th September. The FringeFactory is currently looking for young people who want to show their stuff at their event and you can find out more at their Casting Call page.

The event will take place from 7pm to 9pm on the evening of the 19^th (presumably, straight after the networking drinks). The event will be free to attend, although there may be a charge for drinks/snacks.

If you know any younger researchers, point them in the direction of the FringeFactory.

 

Community Panels are the next big thing in market research

I have been saying this for a while (for example, in chapter 13 of The Handbook of Online and Social Media Research), but I wanted to write a definitive blog post on this item before I become too closely associated with any providers in this area.

Like all useful terms, the phrase Community Panel can be used in a number of subtly different ways, but to me the term relates to something that is positioned between an MROC (online research community) and a private/in-house panel. The typical community panel is large enough to support quantitative research (typically ranging from 5,000 to 50,000), which makes it larger than most MROCs, which are essentially qualitative devices, but it has a level of sophistication and interaction that separates it from traditional panels.

Community Panels differ from traditional in-house panels in terms of their engagement. Traditional panels were largely used for conventional quant surveys, perhaps supplemented with occasional newsletters. By contrast, community panels employ the latest techniques in surveys, communities, online discussions, interactivity, and longitudinal analysis to create a cost effective ‘Voice of the Customer’ resource.

In many ways the community panel is the logical extension of the wider market research move to online access panels. When a large part of the research industry made the move to using online access panels it moved away from concepts such as random access sampling and towards the day-to-day use of convenience sampling (with respondents doing 1 or 2 surveys a week), away from models of validity that depended on sampling and the measurement of sampling error and towards methods such as benchmarking and modelling. The move to online access panels provided the benefits of cost and speed.

For brands that are doing a reasonable amount of research the community panels offers several benefits, even compared with online access panels, these benefits include:

1. Cost: a wide range of companies are reporting that they can conduct three times as much research with the same budget using community panels (this is also true with MROCs). In addition clients with a large research spend are finding they can make absolute savings by using a community panel.
2. Speed: the panel is already accessible to the client, the formats are agreed, the process of designing studies and creating research instruments is quicker because they are not being designed from scratch.
3. Longitudinal analysis: The databases that underpin community panels (from Telligent, to Global Park, to Vision Critical, to Vovici) allow for complex longitudinal analysis, looking at issues such as changes in views, behaviour, and preferences over time.
4. Consistency: all research conducted via the panel can utilise the same scales and processes, increasing clients’ ability to compare results from one study to another, conventional research uses a variety of panel providers and software platforms, introducing extraneous variation.

Of course, very few things come without downsides and the use of community panels introduces two key limitations:

1. The research is branded. In most cases community panels are branded and the members of the community panels know they are dealing with the brand (or an intermediary on behalf of the brand).
2. The research is conducted with customers. There are community panels which include non-customers, but in most cases the active and effective part of a community panel is composed of customers.

My view is that focusing research on customers is not a bad thing. There will always be some research that needs to be conducted with non-customers or in a non-branded way, but I believe these are a minority for most organisations. Customer satisfaction needs to concentrate on customers, but beyond that: ad and image tracking, ad testing, concept screening, ideation, and strategy can all be focused on customers. My feeling is that historically too many brands have focused too much on non-users. After all, in most cases making the product better for customers is likely to make it better for non-users.

The future
I foresee a market where most brands have community panels and where perhaps 80% of their research is conducted via their panel – with non-panel research reserved for those cases where it is actually needed. I think that clients will treat their community panel as a core resource, expecting their suppliers to have a default option of the fieldwork (and the survey scripting) to be conducted via a client’s panel.

For organisations with a large research spend I expect there to be large cost savings, for organisations with smaller budgets they will find their money produces much more research when they switch to a community panel.

 

Can we keep market research and marketing separate?

Research-Live today has a report from the South African Marketing Research Association saying that they are calling for a greater distinctin between research and marketing. In the short term, I think they are right to press for a separation, because of the exemptions for MR that can be achieved, for example greater freedom to contact people, but I fear that in the medium term the battle will be lost.

I say this, and have posted a comment on the Research-Live site, because I am not sure that we will be able to maintain a distinction between marketing and MR much longer. We might be able to prevent MR professionals being involved in marketing, but we can't stop marketing subsuming most aspects of MR.

From CRM, to immersion exercises, to online communities, to social media monitoring, and to response marketing, firms are finding that they get ever larger amounts of their insight from marketing related activities, not from standalone market research. If MR cuts itself off from this process we will be an ever decreasing sector, decreasing in share and I believe in absolute volume.

When I was a child there were lots of coal merchants in my area, but now they are very rare, because almost nobody buys coal. Kodak survived because it expanded out of the film business, Nokia prospered because it got out of the white goods business, I think MR needs to change or die. I think that in the future (maybe the near future) very few organisations will want to buy standalone, traditional market research.

Paying for TweetReach

In the past I have quite often used the free version of TweetReach to see how far a message has moved. The free version looks at just 50 tweets and is often quite interesting. However, tomorrow I am speaking at the LGInsight Acadamy event in Nottingham and I thought it would be good to look at the reach of the #TestItTuesday campaign (Fire and Rescue Services around the UK tweet #TestitTuesday messages every week to improve the awareness of the need to check smoke and carbon monoxide alarms). So on this occasion I paid my US$20 and boght the paid for report.

As you can see by the image below, the full report is really useful, as well as showing that 172 Tweets in the last 7 days reached just over 89000 people the report shows the composition of the tweets, retweets, and replies and it lists all of the Tweets and the people who tweeted them.

$20 is a bit of a steep price I feel, but the full report is definitely more useful than just the last 50 tweets, which is what the free report gives you.

 

 

An Introduction to Ethnogrpahy

A few hours ago we had the first session (timed for Australia/Asia) of the NewMR Ethnography event and the video below was my introduction to that event. If you are reading this on 25 May there may still be time to register for the Europe session (11am UK time) or the US session (2pm New York) at the NewMR website.

 

The NewMR Ethnography event is tomorrow

The NewMR Ethnography event is taking place Wednesday 25 May, the details are available on the NewMR website.

The three sessions are:

Asia time zone:

• Ray Poynter, The Future Place, Ethnography 101
• Prashant Hari, Colmar Brunton, Using WE. Research to access black holes – case study
• Erica Ruyle, Market Strategies International, The Wizard Behind The Curtain: Ethnographic Analysis
• Nick Agafonoff, Real Ethnography, Documentation to Representation – When video trace evidence is presented as ethnography

 Europe time zone:

• Neil McPhee, Nuance Research Ltd, UK, Analysis in Ethnographic work
• Siamack Salari, EverydayLives, Mobile research and the struggle for naturalism and meaning
• Jonathan Gable, BrainJuicer and Christiane Quaas, Boehringer Ingelheim, “Me” to “We” Research: A New Way of Looking
• Jon Puleston, GMI, The secrets of conducting successful Ethnography research online

 America time zone:

• Laura Winn Johnson, CMI, A 360° Brand View: Identifying New Opportunities for Your Business
• Julie Wittes Schlack, Communispace, In Time, In Context, In-Spired: Mindfulness in Mobile Ethnography
• Jamie Gordon, Northstar Research Partners, The 3 C's of Consumer Anthropology
• Kelley Styring, InsightFarm Inc., Digging Deep: Ethnography Driven Innovation

 There is still time to register at the NewMR website. Remember if you register for one of the three sessions you will get the slides for all 12 presentations and you will be able to view the recordings of all 12 presentations.

Also, if you are interested in Ethnography there is a great debate going on in the AMSRS LinkedIn group, discussing “When is it ethnography and when is it not ethnography?”.

This Ethnography event has been sponsored by Square Holes, who describe themselves in the following way:

"The agency prides itself on being youthful and progressive, and events such as this are all about constant evolution of how market researchers better understand consumers.

From an explosive frustration with boring ho-hum research with no creativity nor depth, Square Holes materialised six years ago and has grown solidly working with clients across Australia and beyond with a focus on supporting branding, advertising and digital. Specialist areas include youth, FMCG, health promotion, consumer services, social advertising and sports, art and culture. Square Holes has a strong on-line and off-line offering of traditional and new research tools.

So, if you are keen to partner with such a research agency contact Square Holes. This may be an on the ground partner in Australia or, for those with different time zones, a virtual creative research agency working while you are asleep.”

What is the relationship between market research and science?

The issue of science and market research is one that can cause quite a bit of heat, but here is my two-pennyworth.

Is market research a science?
I think the answer is clearly a no. Although there are many definitions of a science I do not think it is helpful to draw the definition too widely. I define science as “The pursuit of knowledge that seeks generalisable information about the empirical world utilising the paradigm of the scientific method.” Many people would use a more narrow definition, focusing on the creation of generalisable laws, linking cause and effect, but this approach excludes the traditions of parts of the biological sciences, where the main fields of endeavour over the last few hundred years related to finding, measuring, and categorising observable phenomena.

The reason I feel that market research is not a science is that it almost always focuses on the particular, not the general. There are many endeavours that utilise science and which seek to measure and assess the world, but which are not themselves a science. For example, surveyors, tailors, people who inspect cars/engines for faults, journalists, police all examine the world but they are not scientists. Perhaps the best example of this split between using science and being a science is the role of the GP (i.e. the family doctor), doctors use the products of science, but their focus is the specifics of each patient.

It could be argued that a researcher who spent the majority of their time trying to develop a new method of price testing system was engaged in science. But a researcher who was looking at what people in the UK might pay in 2011 for a new type of home delivery might be using science, but they would not be ‘doing science’.

Scientific Method
Even if market research is not a ‘science’ can we say that it does (or should) conform to the scientific method? Again, we need to produce a definition of the scientific method, and my feeling is that it is an approach that attempts to meet the following criteria/beliefs:

1. There is a real world and that world can be measured.
2. The results are generalisable to some extent, i.e. the findings can be extended to a useful number of other situations.
3. The researcher should seek to be neutral, i.e. the researcher is seeking to be objective.
4. The method is noted down in sufficient detail that the research can be replicated.
5. There is a methodological reason for believing that the research is valid (e.g. statistics, models, etc).

Many market researchers would describe most quantitative research as conforming to the model above. I thnk the two biggest weaknesses in this model are criteria 3 and 5.

Can researchers be neutral? Researchers can try to be objective, but the choice of sample, timing, questions, and method are all subjective. Two researchers given the same problem would not necessarily choose the same method, the same type of sample, and same questionnaire to conduct the research. If researchers want to claim they are following a scientific method they need to recognise their limitations and seek to minimise the subjectivity. One method that academia uses in this area is to use questionnaires that have become the ‘validated’ instrument for that test, rather than crafting a new questionnaire for every project – but would this be a popular route for market research?

Criterion 5 above relates to using techniques where there is a reason to believe they are valid. This is a becoming a major area of concern for market research as the neuroscientists, the behavioural economists, and writers such as Mark Earls are all showing that respondents are bad witnesses to their own views and intentions. At present, in market research, there is not a clear mechanism leads to methods that have been shown to lack validity being removed from the toolkit. It sometimes seen that market research is like an alternative medicine vendor who happily sells some cures that that have evidence to support them, some with big question marks over them, and some like homeopathy which have repeatedly been shown not to work.

Alternative Paradigms
Whilst most quantitative market researchers would hold to some version of the scientific method, this is not true of all researchers who study society. Within qualitative research there is a strong strand of constructionist thinking which would raise the following queries about the criteria above.

1. There is not a ‘real’ social world. The social world is made by people, it is an open system, and if some rule were ever to be identified the knowledge of it would permit society to change how it performed.
2. Research tends to be situated, if we study how queuing happens in Starbucks in Nottingham, on a Friday afternoon, we need to be very careful about how generalisable the findings are. To generalise the findings we need to appeal to previous research and to triangulation to show what appears to be common with other situations and what tends to be different.
3. The researcher is not neutral or objective, the researcher changes the research situation and the research is a construction of the researcher. This can be miminised, but it will still be important and should be recognised.
4. The research should be documented, not so it can be replicated (because it can’t be replicated), but so that the reader can take a view on the trust they want to place in the research and findings.
5. The nature of the term ‘valid’ is not consistent to the idea that realities are constructed, socially. The issue of validity becomes re-framed in terms of producing results that are useful and about which we have good reasons to believe.

So, what about science and market research?

My feeling is that two big changes need to happen:

1. In quant research we should accept the limitations of the scientific model in order to benefit from the paradigm, this means addressing the excessive subjectivity in the design of studies and it means discarding approaches if they are shown not to work.
2. In qual we should avoid referring to it as a science, as a scientific method, or as being based on science. We need to explain that the vast majority of society cannot (in the foreseeable future) be reduced to things that can be dealt with by measurement and the scientific method. Clients need to embrace and understand the implications that underpin practical constructionist epistemologies.

Note, I am not saying that clients need to engage in the philosophical arguments about epistemology and ontology. However, at the very least they need to understand:

• Different researchers give different answers (because of design in quant and the paradigm in qual).
• If you are not an expert, judge market research in terms of past performance and recommendations.

 

The New Scientist and "The Grand Delusion"

This week you need to read the print (14 May 2011) or the online version of The New Scientist, no iffs, no buts, you need to read it (the online version is free but you need to register). The article I am imploring you to read is titled "The Grand Delusion".

The Grand Delusion is a seven page article that explains, in easy steps, why the world we think we perceive is just a fabrication constructed in our heads.

Amongst the points raised by the article are:

• During the waking day your eyes are 'offline' for about four hours in total, the total impact of blinks and more importantly saccades (when the eye flicks from one point to another).
• Your colour vision is restricted to an area about the size of your thumbnail when held at arm's length.
• Our biases affect our use of 'facts' not just attitudes.
• Many of our memories are false, but they are indistinguishable from 'true' memories (in 2002 one study found that when asked about the death of Princess Diana nearly half of respondents remembered seeing tv footage of the actual crash- but no such footage exists.
• Most of us tend to think we are better than average at a wide variety of things (74% of drivers think they are better than average, and those who have had a recent crash have an even higher opinion of their skills).

For market researchers all of this news is challenging. As Mark Earls is fond of saying, we are all unreliable witnesses to our own selves.

ps one of the points in the article is that we tend to 'seek out' and believe things that agree with us, a phenomenon called 'confirmation bias', so perhaps I just agree with the article because it is a better informed version of myself? And perhaps I seek out my 'facts' in the New Scientist rather than the Murdoch press because I like the 'facts' in the New Scientist better?