According to a recent research 97 percent of customers said they had read reviews in 2017. There are so many ways now for consumers to share data and information that online consumer reviews and feedback data is in fact become the world largest consumer panel. And because it is an anonymous one it’s easy to share loads and loads of data since no one is worried about saying the wrong thing. This has contributed to the explosion of world data to the point that as IBM recently pointed out, 90% of the world’s data was created in the last 2 years.
In fact, its even better then just the world largest consumer panel, as consumers are not concerned that they are listened to and as such convey their opinions more freely…
With so much data available on such a diversified set of consumer goods/services topics – why aren’t all consumer brands listening to it? There’s so much good data available in areas such as customer service and QA (returns, complaints, failures, missing parts, packaging), product or service (popular features, negative reviews, competitive analysis) and market research (analyzing brands, products and sentiment and looking for white spaces for innovation).
It’s a (huge) moving target?
You’d think with all this data available all you have to do is just mine it…but how would you? You could use text analytics software and teach it what keywords and terms to look for. Its been done quite a bit, though its challenging as you need to guess a lot:
- Guess ALL the different ways that different people talk about the same thing: Imagine how kids, teenagers, boys, girls, adults of different ages all speak about the same topic. Would you expect it to be with the same exact words? If we take the example of a phone battery, you’d need to look for all positive and negative expressions about it such as “my phone lasts for days” or “mine doesn’t hold charge for more then 4 hours” or “my phone’s battery is weak”
- Guess the discussion topics that are not on your mind: How can you guess ALL the different topics consumers talk about? The short answer is you can’t, which means you are missing quite a bit
- Guess the next hot topic: Consumers interests shift all the time based on new products, features or change in flavors. You’d need to know something has come up in order for you to look for it, and then obviously you need to guess the different keywords used by the masses to describe it
To chase a moving target you need different methods
So training systems to look for rigid patterns that humans invent in hope that we cover all that consumers talk about seems hopeless. Lets try to spell out what type of systems can actually track a dynamic, moving target of topics:
- It has to be self learning
- Able to identify topics and interests
- Able to understand that different terms talk about the same topic
- Able to learn new terms that surface and their meaning
With Artificial Intelligence and Self Training algorithms you can skip the person-training-machine steps which limit the scope of the machine understanding and is also slow in terms of response time and skip directly to a machine-training-machine scenario, growing to unlimited scale and immediate response to any variation of a meaning.
Most text analytics technologies rely on humans and thus are slow to setup and mainly they miss a lot of the meaning as consumers interests is a fast moving target.
The good news is that the data is there for us to see and mine, assuming we can get to it.
Revuze is an innovative technology vendor that addresses just this with the first self training, fast setup and low touch solution that typically delivers 5-8X the data coverage compared to anything else, and it does it without humans, in a self training solution that can adapt to the moving target of consumer interests