Whitespace with deeper insights

Whitespace discovery for consumer goods

Market Research: Whitespace Discovery

We take a look at how product managers for the niche Robo Lawn Mower category, can analyse consumer feedback to discover whitespace in an emerging product category.

With a rapidly evolving category with new entrants and evolving consumer requirements, it's hard to know what insights to look for and where the whitespace is. Therefore the perfect candidate for demonstrating the power of Deeper Insight pretrained models and data visualisation tools.

Here we'll show you how to find the real consumer sentiment towards your own brand, a competitors brand, and the specifics around whats loved or hated about the products themselves.

All of this analysis is done using passive data collection, and with automated insights discovery. 

With all our market research projects we follow three simple steps:

1. Gather

First, we trained our Skim Engine's Focused Crawler to search review sites and e-commerce sites for relevant feedback from consumers. Using proxying techniques such as IP rotation, we can crawl millions of sources for specific content related to the identified brands.

The Skim Engine™ also removes boilerplate and extracts features from the text with very high precision. We’re able to locate specific comments that contain the brands or product names we’re after from the crawled sources.

whitespace 1

2. Process


Now the data pipelines are setup to feed constant review and comment data for our Robo Lawn Mower brands, we can start to automatically analyse the data.

At Deeper Insights, we combine a set of pretrained and proprietary language models to analyse the data in near real-time. This gives us an unbiased view of the data compared to an analyst or market researcher performing the same task, and within a fraction of the time and cost.

Choice of models:

  • Consumer Sentiment Model
    Identifies the comment level sentiment related to the brand or product
  • Named Entity Recognition Model
    Recognizes the specific product name or brand name for which the sentiment is related
  • Feature Recognition Model
    Recognizes the specific features or aspects of a product
  • Consumer Feedback Model
    Looks at the explainability of the negative or positive sentiment in relation to a feature, product or brand

3. Visualize

At this stage the smart insights are passed through to our data visualisation dashboard to present the facts.


Firstly, by looking at the category as a whole, it's clear that McCulloch are the industry leaders from an overall consumer sentiment perspective. This is accumulated sentiment across all product features from the review data collected.

However, when drilling down into the specific features of each product to identify the single most loved feature, you can see that Husqvarna comes out the winner for Lawn Care. The Consumer Feedback model has also identified the reasons for that, being the “care route” and “care blogs”. Useful insights to feed through to product and marketing teams at McCulloch about their competitor. 


Secondly, we look at the force-directed node graph for common features shared by two or more brands. This way locating potential whitespace. The models have picked out the Guide Wire as a shared feature from both McCulloch and Flymo, and in this case, it has been positively reviewed for both. 


This is just a snapshot in time and therefore would expect that many more insights are gleaned over further weeks and months of analysis. Since the Skim Engine™ never sleeps, we can provide on-demand market research and insights for product managers or digital marketer as and when they need it, leading to bolder business decisions.