With the launch of Google Consumer Surveys, the behemoth of search has created quite a stir in the world of quantitative research. But as much conversation as GCS has created, it’s probably not such a big surprise. Google already have bountiful experience with online ads (their research offering utilises a similar approach), and have dipped their toes into the insight pool with both Google Analytics and Google Insights For Search; not to mention the survey element added to Google Docs a few years ago. Well, the water must have been nicely warm; now they’ve jumped right in with GCS.

 

The unimaginatively named survey tool officially launched in March this year, but is still functioning on a relatively small scale - only online publishers based in North America with US traffic can sign up to earn cash at the moment. Now there’s an interesting point. Cold hard cash – who pays, who earns, and how much? Let’s have a look at how it works:

 

• Step one: Publisher tells Google they’d like to host surveys for cash. Meanwhile, a product/brand/company owner tells Google they need to find stuff out, but funds are limited.

 

• Step two: Google acts as a kind of dating agency, helping the owners to create their own surveys with their own survey templates and software (pretty simplistic but does the job), charging them one cent per response (so that’s $1 for every 10 responses, or $150 for 1500 responses – which happens to be the recommended amount).  Demographic targeting, or more specific audience targeting (people who like ice cream, for example), costs more at 50 cents per response.

 

• Step three: The survey is set free online. Up to 20 questions can be asked per survey, but the idea is that each respondent will only answer one question each. The survey pops up when consumers try to access premium content on the web, like articles or videos. They can reach the content they desire by answering the question (or, if they’d really rather not take the survey, Google insist there must be an alternative route to get to the content, such as providing their email address). Consumers answer the question, which takes on average 10-15 seconds of their time

 

• Step four: Publishers sit back and let the cash flow in. They get paid a percentage of the money invested in the surveys by the owners, in a similar way to ad hosting. Owners sit back and wait for the target number of consumers to take the survey.

 

• Step five: Publishers cackle at the ease in which they’ve made extra cash. Owners can see the results as each site visitor answers the question, through Google’s own interface. The interface is simple yet effective, enabling owners to look at data by demographics in both table and bars right format. These can be downloaded as image or csv files, which is pretty handy.

So what do we make of it all? Well, in theory, everyone’s a winner. One user said they found it “shockingly easy” – in a good way, we hope. It certainly gives companies on a limited research budget an opportunity to get in touch with consumer opinion. So far, there are some branded case studies on the GCS website, all of which are US-centric (heard of King Arthur flour? 479^o gourmet popcorn? Neither had we). If you dig a little deeper and look at example surveys on the site, you will see that Toluna, a pretty large scale online market research panel provider, has apparently utilised the Google offering to understand awareness of their offering in the US  (Apparently, 90.2% of those answering have never heard of Toluna).

 

There are some drawbacks to GCS, though. Sure, respondent fatigue is virtually eliminated by only requiring them to answer one lonely question, but there’s the annoyance element. People are used to clicking a button and getting what they want, quite literally no questions asked. So will consumers really take the time to read the question in front of them and think about their answer, or will they just click through as quickly as possible to get to where they really want to be? The quality of the data is sure to be affected.

 

From a publisher point of view, there’s the risk that traffic to their survey-hosting sites will suffer as a result; will the percentage of one cent they receive per completion warrant such an impact? Added to all this is the vagueness around how Google work out the demographic detail of consumers: “Using the DoubleClick cookie and the respondent’s IP address, Google Consumer Surveys infers demographic and geographic information for each response so you can easily segment by age, gender, location and more.”Hmm. Besides, how can Google be sure which person in each household is using the computer at the time the survey pops up? 

 

Despite its limitations, it’s an interesting advancement and, for some companies, it’s probably going to feel like the most insightful bargain they’ve ever come across. For now, we’ll watch with curiosity as GCS finds its feet; let’s see what the Google monster does next.

 http://www.google.com/insights/consumersurveys/home

Methodologysurveys

‘Gamification’ – it’s definitely been a top buzz word for 2011. Gaming has always been a big deal, but when social media took off, gaming just got bigger – apps on Facebook, smartphones, tablets etc. People are always connected and, quite a lot of the time, they’re playing a game. But why the need for ‘gamification’? Brands are being told they need to add gaming elements to their offerings – whether through apps, their websites or even their advertising – like McDonald’s have done in Stockholm.

Back in Jan 2011, I viewed a New MR online conference, which looked at the implications of gaming in market research. In our qualitative work this is often a feature of focus groups (ie, making brand mapping into a game, or brand personification tasks). When it comes to quant, we’ve been, um, ‘gamificating’ for years… Back in 2006, we launched the Find Your Tribe survey as part of our UK Tribes work for Channel 4 – segmenting the youth of the UK. In 2008, we gave the survey a make-over and it has seen over 70,000 entries, providing results for many brands on media and purchase trends among young people. The reason for its high completion rate was the gamification of the survey. Engaging, interactive and highly visual, it allowed users to embark on a journey, selecting their favourite brands/media/hobbies/music in a game format with the output at the end presenting the tribe they belonged to (and if they disagreed with the youth tribe they were assigned to they could disagree and let us know!). Gamification works particularly well with youth sectors and the Find Your Tribe survey saw high pass-on rates with most users who participated – inviting friends to take the ‘challenge’ and compare the tribes that they were aligned to.

A lot of our work is often youth orientated – MTV, Nickelodeon, Red Bull, Nuts, Kerrang! Etc, and when it comes to quant for these types of brands we always ensure that the look, feel and language of the surveys have a relevant feel. If it looks good, and is fun to take part in, the respondents will take time to participate – thinking through their answers and therefore providing us with deeper insight. Take Play Respond - a bespoke survey we made with a gamification feel for Viacom. It needed to be visually creative for a youth audience interested in art to take part, and from a quant perspective we added ‘design’ questions that allowed the user to create their own watch (rather like a game). This design element ensured that the respondents took their time over their watch creation, providing some great visual results to present. To make the challenge greater still, we were tasked were producing this in different languages for deployment across a range of markets.

While making sure the look and feel of a survey helps provide a gaming feel, the language used is also key. Gamification language in quant – especially to a youth audience – sees us often use terminology such as ‘guess’ or ‘challenge’, which the audience is familiar with and makes them think a bit harder about their choice of answer (they don’t want to be wrong!)

Let’s be honest, not a lot of young people actively want to take part in an online survey, but through our understanding of what makes them tick, gamification has proved to be a highly successful metric to ensure we get them to take part, think about their responses and perhaps even enjoy the experience!

Dos and Don’ts of gamification in quant surveys

Dos

• Keep it highly visual, interactive and engaging – if it looks nice and flows well people will take the time to answer the questions
• Use it for the right audience and topic, works great for a young audience who understand gaming terminology
• Think about using an ‘output’/answers screen at the end of the survey, whether it’s knowing how other people have answered certain questions or providing an end ‘result’ (ie like Tribes)

Don’ts

• Know your target audience - gamification isn’t for everyone – eg, obviously if you’re talking about a serious topic such as finance
• Don’t go overboard, it’s good to have some gaming terminology/aspects to the survey, but not every question needs it
• It shouldn’t be overtly ‘gamified’ – just subtle enough to draw an interesting slant to the way a question is worded; no one wants to be patronised!

EffectivenessGamingMethodologyStrategyYouthUK Tribes

It's amazing where holiday reading takes you… Somehow we got to reading something on how author/lecturer Silvia Hartmann had modelled the creative processes of Metallica, to gain a better understanding of their 100 million unit shifting success. She discovered that they jam together for hours, meticulously recording everything and subsequently looking for events within the jams with an extraordinary quality to them. They then isolate these events, before looking to build upon them. When Hartmann came to break the approach down into tangible stages, the parallels with developing powerful research projects becomes apparent (if you replace the quest for noisy, grunty creativity with that of achieving commercial advantage, that is)…

• Generate as much material as possible

• Review the material and localise the extraordinary events

• Isolate and lock down these events

• Use the events discovered as a new second level jump off point for generating further material

• Review, localise, isolate and lock down further extraordinary events

• Group the events according to their energetic flavour and matching relationships

• Use the group of events to understand the finished product, and fill in the blanks accordingly

This sounds very much like the approach we aim to deploy on many research projects, chiseling away at the incoming material until we get to findings that genuinely have the power to drive change, then exploring the additional depths achievable through focusing on the most head-turning of findings; with the grouping of events key in terms of presenting organised, structured findings at the project conclusion. It's a sometimes demanding and exacting process but it's one that gets to the evidence-based recommendations that all good research needs to be making. For those about to rock/research, we salute you...

Crowd DNAMusicMethodology