Our data, technology and everything in between.

We created an awesome solution by using reliable public data and applying our signature AI technology to extract insights. Here’s how we did it.

640M

Product insights

18M

Consumer Opinions

300K

Products

25

Product Categories

200

Online Sources

1M

Discussion topics

How we collect and process our data

Here you can see a step-by-step guide of the journey from unstructured consumer opinions all the way to granular and actionable insights:

Step 1

Data collection

First, we collect unstructured consumer opinions — mainley online product reviews — from publicly available sources, such as ecommerce websites.

Step 2

Cleansing

This phase ensures the quality of the data we extract our insights from, as ecommerce websites sometimes manipulate the data to boost the volume of reviews. We therefore use our proprietary cleansing algorithms to cleanse the data, removing fraudulent, duplicated, or promoted opinions.

Step 3

Data structuring

This is where the chaos starts turning into organized and structured data, thanks to our proprietary data structuring algorithms. We have created a unified product catalog from multiple sources, in which the data is structured from the product category level down to each SKU.

Step 4

Classification

Our unique AI discovers the topics of discussion and performs the sentiment analysis, turning the data into actual product insights. This hands-free technology does not require human intervention to predefine topics and keywords, delivering a fully objective and unbiased look into customer satisfaction on each product.

Step 5

Quality assurance (QA)

Our quality assurance human experts and algorithms make sure that the insights we provide are accurate and well organized. Sentimate’s AI technology is 90% accurate for both topics and sentiment classification.

Step 6

Product insights

Users from all over the world use our product insights engine to access millions of reliable and timely product insights to make their business, product, and marketing decisions.

What kind of insights are we able to extract?

Our insights dashboards are so granular, they allow you to make countless discoveries on consumer opinions about any product. On average, our AI engine identifies 50 to 70 topics of discussions for each product.

Number of verified product reviews

See how many verified consumers are discussing the product.

Consumer sentiment

See the average consumer sentiment of the product and for each topic of discussion.

Category rank

See how the product ranks compared to all other products in the corresponding category

Star rating

See not only the average star rating of the product, but also the topics and features that drive the star rating up and down.

SWOT analysis

Identify the product’s strengths, weaknesses, opportunities, and threats benchmarked against the corresponding category

Comparison

Analyze the product in comparison with any other product or with the entire category to identify advantages and disadvantages.

Ready to start using Sentimate?

Create your own user feedback survey