Artificial Intelligence: Your Route to Better Consumer Insights

AI can detect business-relevant insights at a very large scale

So much of popular opinion, and even professional or business opinion, of Artificial Intelligence is that it’s all about robots. Usually evil scary robots, from the worlds of science fiction and futuristic fantasy.

But AI isn’t about scary robots from the future. Artificial Intelligence isn’t the future of Consumer Insights. It’s the now of Consumer Insights.

And that’s because AI is already here. It’s already providing valuable services in a variety of disciplines, including marketing, product development, competitive intelligence and others. It exists in usable, manageable forms, in software solutions, and applications that we use, or can use, all the time and for great value.

One key example is in consumer insights. Here, AI automatically detects business-relevant insights at a very large scale across consumer conversations and audiences. What are consumers saying about my brand? What associations do they make between products and different aspects of their lives? What new trends are emerging in a given industry or audience? These are all insights that AI can provide.

In this post, we will explore a few ways AI is already helping brands improve their bottom line and how it is enabling Crimson Hexagon to deliver real value to our users.

Text Analysis & NLP

One thing that AI can do really well is analyze conversations at massive scale. The only problem is that much of the language businesses want to analyze is messy — conversational, slang-filled, error-ladent and more. It is, in another word, unstructured.

Luckily, advanced AI is overcoming this obstacle by using text analysis techniques that can overcome the messiness of human language.

Text analysis, including Natural Language Processing, can be used to map unstructured text based data to categories and meaning such as sentiment and emotional content.

The important advantage here is scalability: detecting patterns in data libraries quickly and at large scales. It’s technology like this that has enabled the field of social media analysis, an important aspect of Crimson Hexagon’s consumer insights platform, to evolve beyond individual community managers or marketers scanning social feeds for posts to respond, promote, or ignore.

One common, and still useful, application of NLP in consumer insights is sentiment analysis.  Here, by understanding whether consumers have positive or negative opinions of a brand, product, campaign, or other topic, important metrics such as brand health, net promoter scores, and competitive benchmarking are possible.  NLP enables this at scale, and in aggregate or individually.

Going beyond sentiment, NLP can also be utilised to apply emotional categorisation to text. Through the lens of Emotion Analysis, Crimson Hexagon offers a deeper understanding of the collective response beyond basic ‘positive’ & ‘negative’ to tell a bigger, more nuanced story over time. The emotions of joy, sadness, disgust, anger, surprise, and fear help to paint that picture.

As an example application of Emotion Analysis, one financial services company needed to track real-time reactions to their TV advert during a highly viewed US sporting event. By setting up a “war room” with multiple screens using Crimson’s live stream feature, and overlaying with emotional categories, the company was able to  successfully aggregate consumer reaction and respond in real time.

Natural Language Generation

Data is often too complex, too unstructured or messy, or at too large a scale, to be understood easily. People aren’t all analysts, statisticians, or data scientists. They want to ‘read’ insights in ways they can understand quickly and easily.

Natural Language Generation can be used to ‘translate’: turn data into human language, creating an understandable narrative.

At Crimson Hexagon ‘narrative insights’ are available throughout our platform, describing in plain text the key insights being observed.

And by combining Natural Language Generation with machine learning (to detect patterns between user interests and data trends)…  Crimson Hexagon can deliver insights, by surfacing them in the platform or through alert emails, before the user even asks!

Machine Learning

‘Machine Learning’ is a subfield of artificial intelligence with the goal of teaching computer systems to perform a specific task.

Computer systems gain experience by observing patterns from examples rather than being programmed with explicit instruction, rules or keywords to follow.  At Crimson Hexagon, our proprietary ‘Brightview’ algorithm is exactly this sort of AI technology, which allows our users to make sense of unstructured text at scale, and at a depth and level of accuracy that would be impossible using other methods.

Machine Learning is perfect for analysing data that is complex (where each item may have many attributes or observations), noisy (where patterns are complicated by random deviations), and/or unstructured (where the data is ambiguous or meaning is difficult to interpret). In other words: the confusion of unsolicited text data in social media and other consumer feedback channels.

Neural Networks & Deep Learning

By using a deep-learning model to classify images in terms of logos, objects, scenes, and actions, non-text data can be better understood.

Deriving from a neural network approach, such AI has many applications, foremost among them image analysis.

By building a neural network to replicate the way human eyes and brains connect, Crimson Hexagon enables the deeper understanding of image based content.  From recognition of logos to the categorisation of actions, objects, and scenes within images, Neural Networks allow, at scale, what would normally take either a lot of eyeballs or a lot of time.

Image: Human Neural Network vs Artificial Intelligence Networks

AI neural network
Human neural network

In summary, Artificial Intelligence is here, now. It’s a hugely useful tool, to be combined with human intuition, creativity, and more traditional methodologies.  Done right, this combination can produce some really useful, powerful, insights.

If we truly value consumer insights, AI provides an indispensable set of tools to aid us.

Want to learn more about how artificial intelligence is helping brands uncover consumer insights? Download our free guide: The Brand’s Guide to Modern Consumer Insights

Request a Demo

Ready to transform your business?

Get a walkthrough of Crimson Hexagon and learn how consumer insights can help you make better business decisions.