Answers more than Dashboards
When users talk about man watches, which features are more often mentioned?
Each new product is a mix of features ranging from must-have to delighters (product-market fit). Qualifying the right features is sometimes difficult. VoU gives you another perspective: ranking of features not according to internal, often unsupported, considerations: ranking according to the voice of the user, according to the way in which thousands of people spontaneously speaks about products similar to the one you are preparing.
By analyzing unsolicited conversations you are sure that you are not imposing your model, as it would happen in a standard survey: what you get is the mental model users have about a certain type of product/service.
Complexity is something the users speak a lot about. How do them qualify it?
Words of users
Identifying and categorizing product features is crucial, but in later phases of the new product development life cycle you definitely need to go back to users conversations to understand how they are described in social media.
Too many times we see product presentations which targets the right new product features, but describes it is a way that it is either unintelligible or just not sexy enough. VoU helps you to never miss the language point of view of the user: If 60% of the user on social media describe wines they like as amazing, it is not by chance: it is their mental perception, and you cannot take the risk of ignoring it.
Is complexity/simplicity a driver for buying (Intention to buy)?
Yes, it ranks third, after more predictable drivers such as type and size!
You know which are the most popular features of a given product category, but still you don't know what is the attitude of the users towards it. Of course, you might go on the data set and perform manual sampling, but VoU offers much more than that: an automatic categorization of all feature mentions according to the attitude expressed in the linguistic context.
Of course this presupposes the capability of understanding the text, as we want to capture product oriented attitude not only positive/negative: intention to buy, need of a feature, encountered problems, enthusiasm etc. We achieve this result by coupling Artificial Intelligence techniques with Rule Based Systems: we call it hybridation.
I am planning to produce plastic watches to reduce costs: what are the emotions induced by plastic watches?
Bad idea: compare it with the emotions generated when users discuss of titanium watches
Products and Emotions
Sometimes you need to dig even deeper than attitudes. You want to know the emotions raised by a certain feature. Everybody knows how difficult it is to deal with emotions, even with traditional market research tools. They are expressed via several media such as facial gesture, body posture, voice etc. But a way of expressing them is also text.
We extract emotional nuances in the context of product feature mentions and we classify them according to the Paul Ekman six basic emotions:anger, disgust, fear, happiness, sadness and surprise. It is a difficult task, but after some years of research we attain a precision of 80%, which is comparable to the one of human annotators.
What are emerging trends in man watch materials ?
Wood is the emerging trend with a lot of positive attitudes.
Temporal evolution is a critical feature of new product development, as in most cases trend identification is more important that snapshots of a situation at a given time. As all the conversations we deal with are time-stamped, we can provide both day by day monitoring all along the product life cycle and "back in time simulation", i.e. the capability of simulating in the past certain product and marketing related events you are planning for the future.