Bad ads create bad experiences for consumers and negatively impact your brand. It is critical to know what creative element or pattern can help drive brand performance and direct response. However, creative A/B tests or lab tests with facial expression analysis are expensive and not scalable. In this research, we used Google Cloud APIs to compute thousands of features from large creative samples to conduct creative meta-analysis. We added those features from machine learning and computer vision with human encoding elements.
We all love and use buzz words. Such as data-driven marketing, big data and 360° customer insights. But what do they really mean? How do you put them in practice? The Belgian National Lottery put it all in action by blending different data sources.
We all love and use buzz words. Such as data driven marketing, big data and 360° customer insights. But what do they really mean? How do you put them in practice? The Belgian National Lottery put it all in action by blending different data sources.
This presentation talks about the hybrid approach combining big data analytics and qualitative research - with a special spotlight on Kenya and Ethiopia.
This presentation talks about the hybrid approach combining big data analytics and qualitative research - with a special spotlight on Kenya and Ethiopia.
Reading articles about market research and listening to what research buyers tell us they need, there continues to be a hunger for research to: 1. Capture authentic emotion 2. Deliver actionable diagnostic insight 3. Inspire the confidence to act with robust findings. MM-Eye has created a hybrid qual-quant research technique that does all 3: 1. Collecting emotionally rich qual data at scale 2. Converting it into quant data 3. Using a next-gen AI-powered text analytic approach we have custom designed for market research. It is called ThoughtScape and we would like to share our approach and the learnings we have gained along the way. ThoughtScape is theoretically robust, being grounded in the latest neuroscience understanding of how the brain works. It is also backed up by a decade of use, having been the core of the global brand research we undertake for Jaguar Land Rover. And to illustrate the unique insight ThoughtScape delivers we've conducted a demonstration study. The topic we have chosen is airline brands and the study directly contrasts two matched samples; one using traditional survey approaches, the other including ThoughtScape. And we'd like to share these results to demonstrate the proof of concept.
In qualitative research, much can be gained by starting with broad questions which allow the respondents to express an attitude, behaviour, or thought rather than leading them with more pointed questions. But groups and depths are limited by time and budget. So can the process of qualitative exploration be applied at scale using AI and voice technology? Find out what happened when we tried. This paper sets out to test the viability of AI power voice technology to do this and concludes that this is indeed possible.
Out-of-Home (OoH) media has always been in trouble regarding measurements. Although great effort, resources, money and time has been spent in many countries, the complexity of the media - due to its geographical dispersion, volume and granularity - have been a headache for most researchers. The OoH medium has been difficult to measure due to its inherent nature to spread across the country, with each location measuring differently so that averages or simple models do not suffice to show the true potential of the medium. Methods for measuring trips to the locations of OoH audiences have evolved from paper questionnaires, to telephone interviews, to computer-assisted, to tablet- assisted, to using small GPS devices carried by respondents. In all cases, it was always previously expensive and slow. Therefore, when new requirements suddenly appear, due to the âdigital kingâ, it makes the current OoH methodologies appear old, even to more senior research participants. Of course, this impacts the perceptions of the media itself, as the media planning tool lacks the advantages of other media such as âfreshâ data. We think that only with the mix of several data sources, each one contributing its strength, can we provide the accurate information that we need for our goal: a worldwide OoH audience measurement.
Out-of-Home (OoH) media has always been in trouble regarding measurements. Although great effort, resources, money and time has been spent in many countries, the complexity of the media - due to its geographical dispersion, volume and granularity - have been a headache for most researchers. The OoH medium has been difficult to measure due to its inherent nature to spread across the country, with each location measuring differently so that averages or simple models do not suffice to show the true potential of the medium. Methods for measuring trips to the locations of OoH audiences have evolved from paper questionnaires, to telephone interviews, to computer-assisted, to tablet- assisted, to using small GPS devices carried by respondents. In all cases, it was always previously expensive and slow. Therefore, when new requirements suddenly appear, due to the digital king, it makes the current OoH methodologies appear old, even to more senior research participants. Of course, this impacts the perceptions of the media itself, as the media planning tool lacks the advantages of other media such as fresh data. We think that only with the mix of several data sources, each one contributing its strength, can we provide the accurate information that we need for our goal: a worldwide OoH audience measurement.
Imagine that you can see everything from the sky, imagine that you can track all the movements of the people in each city, imagine you crunch all the numbers, think how many decisions would help this data, imagine how outdoor media would transform the impact on the real estate, retail or franchise industry. Now is the time of the new paradigm, taking into account the strength of traditional research and adding steroids with Big Data. The project is an hybrid approach with multiple sources, from high-res satellite images, tracking devices and carrier mobile data. Allowing impossible research until now, in developing countries and expanding opportunities in the developed ones. Using high-tech, almost a sci-fi approach, this is one of a kind, finally allowing the same outputs across all countries.
On 1 January 2016, the charity SKO (Stichting Kijk Onderzoek) started supplying daily ratings for programme content viewed online. Initially the ratings report on online programmes were distributed by NPO, RTL Netherlands and SBS Broadcasting (Dutch television channels). With this data, SKO is able to report on the online reach and viewing behavior per target group. SKO is the first party worldwide to offer daily online ratings. Uniquely, the new research effort is based upon data gathered through a hybrid measuring method that employs both census data and panel data. Data supply commenced on January 1 2016 SKO is the first party worldwide that is able to offer detailed, high quality online viewing information through a hybrid method. Phase 1: measuring online viewing behaviour next to TV set viewing behaviour: In order to produce a reliable online video currency, data from the SKO Online Panel and SKO Census needed to be combined. This is the first step of data integration. Initially, the online ratings include NPO, RTL Netherlands and SBS Broadcasting programmes. Later on, other participants such as Discovery Networks, VIACOM, FOX International Channels Benelux, Disney and Sanoma will join SKOâs online effort. Phase 2: integration of online viewing behaviour and TV set viewing behaviour In the second phase of the project â now being worked on â the online video data will be integrated with existing ratings through data fusion. This will result in a cross-media Video Total. From then on, it will be possible to report on the total reach of programmes and video commercials. Users will also be able to keep track of the share of âofflineâ and âonlineâ in the total reach. Of course it is possible, just like in Phase 1, to make analyses per target group.
Imagine that you can see everything from the sky, imagine that you can track all the movements of the people in each city, imagine you crunch all the numbers, think how many decisions would help this data, imagine how outdoor media would transform the impact on the real estate, retail or franchise industry. Now is the time of the new paradigm, taking into account the strength of traditional research and adding steroids with Big Data. The project is an hybrid approach with multiple sources, from high-res satellite images, tracking devices and carrier mobile data. Allowing impossible research until now, in developing countries and expanding opportunities in the developed ones. Using high-tech, almost a sci-fi approach, this is one of a kind, finally allowing the same outputs across all countries.